Agent_Architecture2_(Input_State_only).ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Cab-Driver Agent"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "**Note:**<br>\n",
    "(i) Considering **Architecture 2** given in the problem statement **(where Input: State only)**.<br>\n",
    "(ii) The corresponding function returning an encoded state is a vector of size \"m+t+d\" (wnere m:no. of locations/cities, t:hours, d:days_of_week). It is named as **'state_encod_arch2'** in the Env.py file and shall be used further in this code."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Architecture 2  (where Input: State only)\n",
    "<img src=\"Architecture_2_Image(Input_State_Only).PNG\" width=\"800\" />"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Using TensorFlow backend.\n"
     ]
    }
   ],
   "source": [
    "# Importing libraries\n",
    "import sys\n",
    "import numpy as np\n",
    "np.set_printoptions(threshold=sys.maxsize)  \n",
    "import pandas as pd\n",
    "import random\n",
    "import time\n",
    "import math\n",
    "from collections import deque\n",
    "import collections\n",
    "import pickle\n",
    "\n",
    "# for building DQN model\n",
    "from tensorflow import keras\n",
    "from keras import layers\n",
    "from keras import Sequential\n",
    "from keras.layers import Dense, Activation, Flatten\n",
    "from keras.optimizers import Adam\n",
    "\n",
    "\n",
    "# for plotting graphs\n",
    "import matplotlib.pyplot as plt\n",
    "\n",
    "# Import the environment\n",
    "from Env import *    \n",
    "\n",
    "# Suppress Warnings\n",
    "import warnings\n",
    "warnings.filterwarnings(\"ignore\")\n",
    "\n",
    "# Suppress Tensorflow Warnings \n",
    "import tensorflow as tf\n",
    "#tf.compat.v1.logging.set_verbosity(tf.compat.v1.logging.ERROR)  "
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Defining Time Matrix"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Loading the time matrix provided\n",
    "Time_matrix = np.load(\"TM.npy\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "(5, 5, 24, 7)\n",
      "6.0\n"
     ]
    }
   ],
   "source": [
    "print(Time_matrix.shape)\n",
    "print(Time_matrix[3][4][17][5]) #Example: (Returns Time Taken) 𝑇𝑖𝑚𝑒−𝑚𝑎𝑡𝑟𝑖𝑥[𝑠𝑡𝑎𝑟𝑡−𝑙𝑜𝑐][𝑒𝑛𝑑−𝑙𝑜𝑐][ℎ𝑜𝑢𝑟−𝑜𝑓−𝑡ℎ𝑒−𝑑𝑎𝑦] [𝑑𝑎𝑦−𝑜𝑓−𝑡ℎ𝑒−𝑤𝑒𝑒𝑘]"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "For<br>𝑠𝑡𝑎𝑟𝑡−𝑙𝑜𝑐: **3**<br>𝑒𝑛𝑑−𝑙𝑜𝑐: **4**<br>ℎ𝑜𝑢𝑟−𝑜𝑓−𝑡ℎ𝑒−𝑑𝑎𝑦: **17**<br>𝑑𝑎𝑦−𝑜𝑓−𝑡ℎ𝑒−𝑤𝑒𝑒𝑘: **5**<br>The time taken by the cab driver is **6.0 hours**.        "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Minimimum time taken: 0.0\n",
      "Maximum time taken: 11.0\n",
      "Average time taken: 3.0542857142857143\n"
     ]
    }
   ],
   "source": [
    "## Understanding the min and max time duration of trips from one point to another\n",
    "\n",
    "print(\"Minimimum time taken:\", Time_matrix.min())\n",
    "print(\"Maximum time taken:\", Time_matrix.max())\n",
    "print(\"Average time taken:\", Time_matrix.mean())"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "The maximum time taken is 11 hours. So, it is safe to say that the maximum time taken by the cab driver to move from one point to another is less than 1 day."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Random State Initialization:\n",
      "(1, 23, 3)\n",
      "(1, 13, 0)\n",
      "(4, 14, 3)\n",
      "(4, 7, 3)\n",
      "(2, 13, 2)\n"
     ]
    }
   ],
   "source": [
    "## Check Random State\n",
    "from Env import CabDriver \n",
    "print(\"Random State Initialization:\")\n",
    "for i in range(5):         ## Checking for 5 epsiodes\n",
    "    env= CabDriver()\n",
    "    random_state_init= env.state_init\n",
    "    print(random_state_init)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Action List (action_indices, action):\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "[(0, (0, 0)),\n",
       " (1, (0, 1)),\n",
       " (2, (0, 2)),\n",
       " (3, (0, 3)),\n",
       " (4, (0, 4)),\n",
       " (5, (1, 0)),\n",
       " (6, (1, 2)),\n",
       " (7, (1, 3)),\n",
       " (8, (1, 4)),\n",
       " (9, (2, 0)),\n",
       " (10, (2, 1)),\n",
       " (11, (2, 3)),\n",
       " (12, (2, 4)),\n",
       " (13, (3, 0)),\n",
       " (14, (3, 1)),\n",
       " (15, (3, 2)),\n",
       " (16, (3, 4)),\n",
       " (17, (4, 0)),\n",
       " (18, (4, 1)),\n",
       " (19, (4, 2)),\n",
       " (20, (4, 3))]"
      ]
     },
     "execution_count": 6,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "## Creating tuples of action_index and actions from action_space\n",
    "from Env import CabDriver\n",
    "env= CabDriver()\n",
    "cab_action_space = env.action_space\n",
    "cab_action_indices =[i for i in range(len(cab_action_space))]\n",
    "action_list= [i for i in zip(cab_action_indices, cab_action_space)]   ## tuples (action_indices, action)\n",
    "print(\"Action List (action_indices, action):\")\n",
    "action_list"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "state_size: 36\n"
     ]
    }
   ],
   "source": [
    "## Understanding state_size\n",
    "state_size= m+t+d               ## Considering DQN_Architecture 2 (Input: State Only)\n",
    "print(\"state_size:\", state_size)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "action_size: 21\n"
     ]
    }
   ],
   "source": [
    "## Understanding action_size\n",
    "action_size= len(env.action_space)\n",
    "print(\"action_size:\",action_size)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Agent Class\n",
    "\n",
    "If you are using this framework, you need to fill the following to complete the following code block:\n",
    "1. State and Action Size\n",
    "2. Hyperparameters\n",
    "3. Create a neural-network model in function 'build_model()'\n",
    "4. Define epsilon-greedy strategy in function 'get_action()'\n",
    "5. Complete the function 'append_sample()'. This function appends the recent experience tuple <state, action, reward, new-state> to the memory\n",
    "6. Complete the 'train_model()' function with following logic:\n",
    "   - If the memory size is greater than mini-batch size, you randomly sample experiences from memory as per the mini-batch size and do the following:\n",
    "      - Initialise your input and output batch for training the model\n",
    "      - Calculate the target Q value for each sample: reward + gamma*max(Q(s'a,))\n",
    "      - Get Q(s', a) values from the last trained model\n",
    "      - Update the input batch as your encoded state and output batch as your Q-values\n",
    "      - Then fit your DQN model using the updated input and output batch."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [],
   "source": [
    "class DQNAgent:\n",
    "    def __init__(self, state_size, action_size):\n",
    "        # Define size of state and action\n",
    "        self.state_size = state_size\n",
    "        self.action_size = action_size\n",
    "\n",
    "        # Write here: Specify you hyper parameters for the DQN\n",
    "        self.discount_factor = 0.95\n",
    "        self.learning_rate =  0.01   \n",
    "        self.epsilon = 1\n",
    "        self.epsilon_max = 1\n",
    "        self.epsilon_decay = -0.00045\n",
    "        self.epsilon_min = 0.0000001\n",
    "        \n",
    "        self.batch_size = 32        \n",
    "        # Create replay memory using deque\n",
    "        self.memory = deque(maxlen=2000)\n",
    "        \n",
    "        # Initialize the value of the states tracked for all samples\n",
    "        self.states_tracked_1 = []\n",
    "        self.states_tracked_2 = []\n",
    "        \n",
    "        \n",
    "        # For Sample state-action pair 1: We are going to track state (2,4,6) and action (2,3) at index 11 in the action space.\n",
    "        self.track_state_1 = np.array(env.state_encod_arch2((2,4,6))).reshape(1, self.state_size)\n",
    "        \n",
    "        # For Sample state-action pair 2: We are going to track state (1,2,3) and action (1,2) at index 6 in the action space.\n",
    "        self.track_state_2 = np.array(env.state_encod_arch2((1,2,3))).reshape(1, self.state_size)\n",
    "        \n",
    "\n",
    "        # Create main model and target model\n",
    "        self.model = self.build_model()\n",
    "\n",
    "    # approximate Q function using Neural Network\n",
    "    def build_model(self):\n",
    "        \"\"\"\n",
    "        Takes in the agent and constructs a neural network to train it and returns model and parameters.\n",
    "        \"\"\"\n",
    "        model = Sequential()\n",
    "        # Write your code here: Add layers to your neural nets \n",
    "        \n",
    "        # Hidden Layers \n",
    "        model.add(Dense(32, input_dim= self.state_size, activation= 'relu', kernel_initializer= 'he_uniform'))\n",
    "        model.add(Dense(32, activation= 'relu', kernel_initializer= 'he_uniform'))\n",
    "        \n",
    "        # the output layer: output is of size of number of actions (action_size) and returns Q values for all actions\n",
    "        model.add(Dense(self.action_size, activation='relu', kernel_initializer='he_uniform'))\n",
    "        \n",
    "        # Compile Model\n",
    "        model.compile(loss='mse',optimizer=Adam(lr=self.learning_rate))\n",
    "        model.summary\n",
    "        \n",
    "        return model\n",
    "\n",
    "\n",
    "\n",
    "    def get_action(self, state):\n",
    "        \n",
    "    # Write your code here:\n",
    "    # get action from model using epsilon-greedy policy\n",
    "    # Decay in ε after we generate each sample from the environment   \n",
    "    \n",
    "        # Get a list of the ride requests driver got.\n",
    "        possible_actions_index, actions= env.requests(state)\n",
    "        \n",
    "        # get action from model using epsilon-greedy policy\n",
    "        z = np.random.random()  ## variable z randomly chooses a value between [0,1)\n",
    "        if z <= self.epsilon:\n",
    "            # explore: choose a random action from all possible actions\n",
    "            return random.randrange(self.action_size)\n",
    "        else:\n",
    "            # choose the action with the highest q(s, a)\n",
    "            # the first index corresponds to the batch size, so\n",
    "            # reshape state to (1, state_size) so that the first index corresponds to the batch size\n",
    "            state = np.array(env.state_encod_arch2(state)).reshape(1, self.state_size)\n",
    "            \n",
    "            # Use the model to predict the Q_values.\n",
    "            q_vals = self.model.predict(state)\n",
    "            \n",
    "            # Truncate the array to only those actions that are part of the ride  requests.\n",
    "            q_vals_possible = np.array([q_vals[0][i] for i in possible_actions_index])\n",
    "\n",
    "            return possible_actions_index[np.argmax(q_vals_possible)]\n",
    "\n",
    "        \n",
    "    def append_sample(self, state, action_index, reward, next_state, done):\n",
    "        \"\"\"appends the new agent run output to replay buffer\"\"\"\n",
    "    # Write your code here:\n",
    "    # save sample <s,a,r,s'> to the replay memory. Also, taking argument 'done' --> terminal state flag.\n",
    "        self.memory.append((state, action_index, reward, next_state, done))\n",
    "    \n",
    "\n",
    "    # pick samples randomly from replay memory (with batch_size) and train the network\n",
    "    def train_model(self):\n",
    "        \"\"\" \n",
    "        Function to train the model on each step run.\n",
    "        Picks the random memory events according to batch size and \n",
    "        runs it through the network to train it.\n",
    "        \"\"\"\n",
    "        # Write your code from here\n",
    "        # 1. Update your 'update_output' and 'update_input' batch           \n",
    "        # 2. Predict the target from earlier model\n",
    "        # 3. Get the target for the Q-network\n",
    "        \n",
    "        if len(self.memory) > self.batch_size:\n",
    "            # Sample batch from the memory\n",
    "            mini_batch = random.sample(self.memory, self.batch_size)\n",
    "            # initialise two matrices - update_input and update_output\n",
    "            update_input = np.zeros((self.batch_size, self.state_size))\n",
    "            update_output = np.zeros((self.batch_size, self.state_size))\n",
    "            actions, rewards, done = [], [], []\n",
    "            \n",
    "             # Update your 'update_output' and 'update_input' batch\n",
    "            # populate update_input and update_output and the lists rewards, actions, done\n",
    "            for i in range(self.batch_size):\n",
    "                state, action, reward, next_state, done_boolean = mini_batch[i]\n",
    "                update_input[i] = env.state_encod_arch2(state)     \n",
    "                actions.append(action)\n",
    "                rewards.append(reward)\n",
    "                update_output[i] = env.state_encod_arch2(next_state)\n",
    "                done.append(done_boolean)\n",
    "\n",
    "            # predict the target q-values from states s\n",
    "            target = self.model.predict(update_input)\n",
    "            # target for q-network\n",
    "            target_qval = self.model.predict(update_output)\n",
    "\n",
    "\n",
    "            # update the target values\n",
    "            for i in range(self.batch_size):\n",
    "                if done[i]:      #terminal state\n",
    "                    target[i][actions[i]] = rewards[i]\n",
    "                else:            # non-terminal state\n",
    "                    target[i][actions[i]] = rewards[i] + self.discount_factor * np.max(target_qval[i])\n",
    "             # 4. Fit your model and track the loss values\n",
    "            # model fit\n",
    "            self.model.fit(update_input, target, batch_size=self.batch_size, epochs=1, verbose=0)\n",
    "    \n",
    "    def save_tracking_states(self):\n",
    "        # Use the model to predict the q_value of the state we are tacking.\n",
    "        q_value_1 = self.model.predict(self.track_state_1)\n",
    "        q_value_2 = self.model.predict(self.track_state_2)\n",
    "\n",
    "        \n",
    "        # Grab the q_value of the action index that we are tracking.\n",
    "        self.states_tracked_1.append(q_value_1[0][11])    ## action (2,3) at index 11 in the action space\n",
    "        self.states_tracked_2.append(q_value_2[0][6])     ## action (1,2) at index 6 in the action space\n",
    "\n",
    "        \n",
    "    def save_weights_numpy(self, name):      ## Saves the 'model_weights' in a  pickle file\n",
    "        weights= self.model.get_weights()   ## Gets model_weights as a list of numpy arrays\n",
    "        try:\n",
    "            fpkl= open(name, 'wb')\n",
    "            pickle.dump(weights, fpkl, protocol= pickle.HIGHEST_PROTOCOL)  \n",
    "            fpkl.close()\n",
    "        except Exception as e:\n",
    "            print(e)\n",
    "    \n",
    "    def save(self, name):\n",
    "        self.model.save(name)      ## Saves the DQN_model in Keras H5 format "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [],
   "source": [
    "episode_time = 24*30   # 24 hrs for 30 days per episode\n",
    "Episodes = 20000       # No. of Episodes\n",
    "m = 5                  # No. Locations\n",
    "t = 24                 # No. of hrs in a day\n",
    "d = 7                  # No. of days in a week \n",
    "\n",
    "# Invoke Env class\n",
    "env = CabDriver()\n",
    "action_space, state_space, state = env.reset()\n",
    "\n",
    "# Set up state and action sizes.\n",
    "state_size = m+t+d   #Using Architecture 2 (input:state)\n",
    "action_size = len(action_space)\n",
    "\n",
    "# Invoke agent class\n",
    "agent = DQNAgent(action_size=action_size, state_size=state_size)\n",
    "\n",
    "# to store rewards in each episode\n",
    "rewards_per_episode, episodes = [], []\n",
    "# Rewards for state\n",
    "rewards_init_state = []"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### DQN block"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "for episode in range(Episodes):\n",
    "\n",
    "    # Write code here\n",
    "    # Call the environment\n",
    "    # Call all the initialised variables of the environment\n",
    "    \n",
    "\n",
    "    #Call the DQN agent\n",
    "    \n",
    "    \n",
    "    while !terminal_state:\n",
    "        \n",
    "        # Write your code here\n",
    "        # 1. Pick epsilon-greedy action from possible actions for the current state\n",
    "        # 2. Evaluate your reward and next state\n",
    "        # 3. Append the experience to the memory\n",
    "        # 4. Train the model by calling function agent.train_model\n",
    "        # 5. Keep a track of rewards, Q-values, loss\n",
    "        "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Saving Model 0\n",
      "Saving Model 0 Weights\n",
      "episode 9, initial_state (3, 17, 4), reward -299.0, memory_length 1146, epsilon 0.9959581905936945 total_time 723.0\n",
      "episode 19, initial_state (0, 22, 4), reward -76.0, memory_length 2000, epsilon 0.9914864481525782 total_time 722.0\n",
      "episode 29, initial_state (1, 18, 0), reward 1.0, memory_length 2000, epsilon 0.9870347833438929 total_time 721.0\n",
      "episode 39, initial_state (4, 20, 2), reward -45.0, memory_length 2000, epsilon 0.9826031060212743 total_time 724.0\n",
      "episode 49, initial_state (2, 15, 3), reward -30.0, memory_length 2000, epsilon 0.978191326443105 total_time 725.0\n",
      "episode 59, initial_state (1, 16, 3), reward 37.0, memory_length 2000, epsilon 0.9737993552706977 total_time 722.0\n",
      "episode 69, initial_state (0, 9, 6), reward 89.0, memory_length 2000, epsilon 0.9694271035664864 total_time 722.0\n",
      "episode 79, initial_state (3, 0, 0), reward 190.0, memory_length 2000, epsilon 0.9650744827922243 total_time 723.0\n",
      "episode 89, initial_state (0, 13, 2), reward 253.0, memory_length 2000, epsilon 0.9607414048071923 total_time 727.0\n",
      "episode 99, initial_state (3, 23, 1), reward 76.0, memory_length 2000, epsilon 0.956427781866413 total_time 724.0\n",
      "episode 109, initial_state (4, 20, 6), reward 19.0, memory_length 2000, epsilon 0.9521335266188744 total_time 734.0\n",
      "episode 119, initial_state (3, 3, 2), reward -1.0, memory_length 2000, epsilon 0.947858552105761 total_time 728.0\n",
      "episode 129, initial_state (2, 10, 6), reward 65.0, memory_length 2000, epsilon 0.9436027717586928 total_time 726.0\n",
      "episode 139, initial_state (0, 23, 2), reward -268.0, memory_length 2000, epsilon 0.9393660993979726 total_time 721.0\n",
      "episode 149, initial_state (1, 8, 1), reward 137.0, memory_length 2000, epsilon 0.93514844923084 total_time 722.0\n",
      "episode 159, initial_state (0, 13, 2), reward 30.0, memory_length 2000, epsilon 0.9309497358497351 total_time 721.0\n",
      "episode 169, initial_state (3, 14, 2), reward 45.0, memory_length 2000, epsilon 0.9267698742305686 total_time 729.0\n",
      "episode 179, initial_state (0, 9, 3), reward 44.0, memory_length 2000, epsilon 0.9226087797309996 total_time 722.0\n",
      "episode 189, initial_state (1, 15, 2), reward 68.0, memory_length 2000, epsilon 0.9184663680887225 total_time 726.0\n",
      "episode 199, initial_state (4, 22, 4), reward 207.0, memory_length 2000, epsilon 0.91434255541976 total_time 735.0\n",
      "episode 209, initial_state (0, 14, 4), reward -18.0, memory_length 2000, epsilon 0.9102372582167645 total_time 724.0\n",
      "episode 219, initial_state (0, 0, 6), reward -16.0, memory_length 2000, epsilon 0.9061503933473274 total_time 728.0\n",
      "episode 229, initial_state (2, 10, 1), reward -60.0, memory_length 2000, epsilon 0.9020818780522954 total_time 721.0\n",
      "episode 239, initial_state (4, 16, 1), reward -301.0, memory_length 2000, epsilon 0.898031629944095 total_time 728.0\n",
      "episode 249, initial_state (1, 12, 3), reward -105.0, memory_length 2000, epsilon 0.8939995670050633 total_time 721.0\n",
      "episode 259, initial_state (4, 19, 6), reward 97.0, memory_length 2000, epsilon 0.889985607585788 total_time 721.0\n",
      "episode 269, initial_state (0, 21, 1), reward -328.0, memory_length 2000, epsilon 0.8859896704034541 total_time 725.0\n",
      "episode 279, initial_state (3, 8, 5), reward 297.0, memory_length 2000, epsilon 0.8820116745401966 total_time 733.0\n",
      "episode 289, initial_state (1, 13, 4), reward 410.0, memory_length 2000, epsilon 0.8780515394414635 total_time 727.0\n",
      "episode 299, initial_state (4, 18, 2), reward 234.0, memory_length 2000, epsilon 0.8741091849143839 total_time 724.0\n",
      "episode 309, initial_state (1, 8, 6), reward 274.0, memory_length 2000, epsilon 0.8701845311261438 total_time 724.0\n",
      "episode 319, initial_state (3, 23, 0), reward 68.0, memory_length 2000, epsilon 0.8662774986023697 total_time 728.0\n",
      "episode 329, initial_state (3, 15, 3), reward 194.0, memory_length 2000, epsilon 0.8623880082255198 total_time 732.0\n",
      "episode 339, initial_state (0, 7, 3), reward 475.0, memory_length 2000, epsilon 0.8585159812332809 total_time 723.0\n",
      "episode 349, initial_state (0, 11, 4), reward 345.0, memory_length 2000, epsilon 0.8546613392169741 total_time 722.0\n",
      "episode 359, initial_state (3, 14, 6), reward -230.0, memory_length 2000, epsilon 0.8508240041199666 total_time 721.0\n",
      "episode 369, initial_state (2, 7, 5), reward 295.0, memory_length 2000, epsilon 0.847003898236092 total_time 729.0\n",
      "episode 379, initial_state (1, 8, 3), reward 72.0, memory_length 2000, epsilon 0.8432009442080755 total_time 721.0\n",
      "episode 389, initial_state (1, 7, 6), reward 162.0, memory_length 2000, epsilon 0.8394150650259677 total_time 724.0\n",
      "episode 399, initial_state (1, 7, 6), reward -53.0, memory_length 2000, epsilon 0.8356461840255863 total_time 733.0\n",
      "episode 409, initial_state (4, 10, 0), reward 156.0, memory_length 2000, epsilon 0.831894224886962 total_time 722.0\n",
      "episode 419, initial_state (1, 0, 3), reward 88.0, memory_length 2000, epsilon 0.8281591116327941 total_time 726.0\n",
      "episode 429, initial_state (3, 19, 3), reward 132.0, memory_length 2000, epsilon 0.8244407686269115 total_time 725.0\n",
      "episode 439, initial_state (1, 6, 5), reward 118.0, memory_length 2000, epsilon 0.8207391205727411 total_time 721.0\n",
      "episode 449, initial_state (4, 12, 5), reward 289.0, memory_length 2000, epsilon 0.8170540925117837 total_time 728.0\n",
      "episode 459, initial_state (4, 2, 0), reward 332.0, memory_length 2000, epsilon 0.813385609822095 total_time 724.0\n",
      "episode 469, initial_state (1, 23, 2), reward -151.0, memory_length 2000, epsilon 0.809733598216775 total_time 723.0\n",
      "episode 479, initial_state (3, 16, 3), reward -75.0, memory_length 2000, epsilon 0.8060979837424641 total_time 721.0\n",
      "episode 489, initial_state (0, 20, 5), reward 79.0, memory_length 2000, epsilon 0.8024786927778449 total_time 724.0\n",
      "episode 499, initial_state (2, 8, 2), reward 250.0, memory_length 2000, epsilon 0.7988756520321516 total_time 726.0\n",
      "episode 509, initial_state (1, 21, 5), reward 604.0, memory_length 2000, epsilon 0.7952887885436859 total_time 724.0\n",
      "episode 519, initial_state (0, 0, 2), reward 99.0, memory_length 2000, epsilon 0.7917180296783399 total_time 729.0\n",
      "episode 529, initial_state (3, 16, 5), reward -27.0, memory_length 2000, epsilon 0.7881633031281245 total_time 721.0\n",
      "episode 539, initial_state (0, 21, 3), reward 501.0, memory_length 2000, epsilon 0.7846245369097054 total_time 726.0\n",
      "episode 549, initial_state (3, 3, 2), reward 308.0, memory_length 2000, epsilon 0.7811016593629457 total_time 724.0\n",
      "episode 559, initial_state (0, 18, 4), reward 123.0, memory_length 2000, epsilon 0.777594599149455 total_time 728.0\n",
      "episode 569, initial_state (4, 11, 4), reward 94.0, memory_length 2000, epsilon 0.7741032852511438 total_time 725.0\n",
      "episode 579, initial_state (3, 8, 4), reward 234.0, memory_length 2000, epsilon 0.7706276469687865 total_time 722.0\n",
      "episode 589, initial_state (3, 22, 0), reward -31.0, memory_length 2000, epsilon 0.7671676139205893 total_time 724.0\n",
      "episode 599, initial_state (3, 18, 3), reward 446.0, memory_length 2000, epsilon 0.7637231160407644 total_time 723.0\n",
      "episode 609, initial_state (0, 10, 0), reward 390.0, memory_length 2000, epsilon 0.7602940835781123 total_time 728.0\n",
      "episode 619, initial_state (2, 15, 3), reward 80.0, memory_length 2000, epsilon 0.7568804470946082 total_time 721.0\n",
      "episode 629, initial_state (1, 16, 5), reward 257.0, memory_length 2000, epsilon 0.7534821374639971 total_time 721.0\n",
      "episode 639, initial_state (1, 5, 3), reward 315.0, memory_length 2000, epsilon 0.7500990858703923 total_time 724.0\n",
      "episode 649, initial_state (3, 0, 1), reward 404.0, memory_length 2000, epsilon 0.7467312238068837 total_time 721.0\n",
      "episode 659, initial_state (0, 14, 2), reward 683.0, memory_length 2000, epsilon 0.7433784830741496 total_time 725.0\n",
      "episode 669, initial_state (0, 8, 4), reward -51.0, memory_length 2000, epsilon 0.7400407957790751 total_time 726.0\n",
      "episode 679, initial_state (1, 20, 5), reward 320.0, memory_length 2000, epsilon 0.7367180943333789 total_time 723.0\n",
      "episode 689, initial_state (4, 11, 4), reward 475.0, memory_length 2000, epsilon 0.7334103114522429 total_time 726.0\n",
      "episode 699, initial_state (0, 8, 5), reward 105.0, memory_length 2000, epsilon 0.7301173801529509 total_time 721.0\n",
      "episode 709, initial_state (1, 12, 5), reward 232.0, memory_length 2000, epsilon 0.7268392337535315 total_time 721.0\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "episode 719, initial_state (1, 8, 2), reward 61.0, memory_length 2000, epsilon 0.7235758058714081 total_time 730.0\n",
      "episode 729, initial_state (3, 11, 3), reward 97.0, memory_length 2000, epsilon 0.7203270304220545 total_time 725.0\n",
      "episode 739, initial_state (2, 1, 0), reward 640.0, memory_length 2000, epsilon 0.717092841617657 total_time 722.0\n",
      "episode 749, initial_state (2, 3, 6), reward 312.0, memory_length 2000, epsilon 0.7138731739657815 total_time 729.0\n",
      "episode 759, initial_state (3, 19, 4), reward 428.0, memory_length 2000, epsilon 0.7106679622680484 total_time 722.0\n",
      "episode 769, initial_state (4, 11, 3), reward 231.0, memory_length 2000, epsilon 0.7074771416188109 total_time 723.0\n",
      "episode 779, initial_state (3, 0, 1), reward 475.0, memory_length 2000, epsilon 0.7043006474038421 total_time 728.0\n",
      "episode 789, initial_state (2, 17, 4), reward 113.0, memory_length 2000, epsilon 0.7011384152990257 total_time 722.0\n",
      "episode 799, initial_state (0, 0, 5), reward 249.0, memory_length 2000, epsilon 0.697990381269053 total_time 726.0\n",
      "episode 809, initial_state (0, 9, 3), reward 82.0, memory_length 2000, epsilon 0.6948564815661279 total_time 732.0\n",
      "episode 819, initial_state (1, 1, 2), reward 265.0, memory_length 2000, epsilon 0.691736652728674 total_time 721.0\n",
      "episode 829, initial_state (3, 3, 2), reward 495.0, memory_length 2000, epsilon 0.6886308315800508 total_time 732.0\n",
      "episode 839, initial_state (1, 12, 2), reward 332.0, memory_length 2000, epsilon 0.6855389552272738 total_time 726.0\n",
      "episode 849, initial_state (2, 13, 3), reward 786.0, memory_length 2000, epsilon 0.6824609610597415 total_time 722.0\n",
      "episode 859, initial_state (3, 8, 1), reward 204.0, memory_length 2000, epsilon 0.6793967867479666 total_time 725.0\n",
      "episode 869, initial_state (3, 22, 1), reward 567.0, memory_length 2000, epsilon 0.6763463702423145 total_time 728.0\n",
      "episode 879, initial_state (0, 21, 3), reward 632.0, memory_length 2000, epsilon 0.6733096497717469 total_time 723.0\n",
      "episode 889, initial_state (3, 4, 2), reward 303.0, memory_length 2000, epsilon 0.6702865638425705 total_time 731.0\n",
      "episode 899, initial_state (0, 23, 1), reward 686.0, memory_length 2000, epsilon 0.6672770512371918 total_time 724.0\n",
      "episode 909, initial_state (0, 22, 2), reward 625.0, memory_length 2000, epsilon 0.6642810510128779 total_time 721.0\n",
      "episode 919, initial_state (0, 3, 4), reward 610.0, memory_length 2000, epsilon 0.6612985025005217 total_time 730.0\n",
      "episode 929, initial_state (2, 21, 4), reward 315.0, memory_length 2000, epsilon 0.6583293453034138 total_time 723.0\n",
      "episode 939, initial_state (2, 17, 0), reward 450.0, memory_length 2000, epsilon 0.6553735192960198 total_time 726.0\n",
      "episode 949, initial_state (2, 4, 5), reward 238.0, memory_length 2000, epsilon 0.652430964622762 total_time 722.0\n",
      "episode 959, initial_state (2, 11, 1), reward 396.0, memory_length 2000, epsilon 0.6495016216968075 total_time 721.0\n",
      "episode 969, initial_state (3, 7, 5), reward 41.0, memory_length 2000, epsilon 0.646585431198862 total_time 729.0\n",
      "episode 979, initial_state (4, 0, 3), reward 492.0, memory_length 2000, epsilon 0.6436823340759683 total_time 721.0\n",
      "episode 989, initial_state (2, 5, 4), reward 533.0, memory_length 2000, epsilon 0.6407922715403106 total_time 721.0\n",
      "episode 999, initial_state (3, 14, 1), reward 536.0, memory_length 2000, epsilon 0.6379151850680236 total_time 721.0\n",
      "Saving Model 1000\n",
      "Saving Model 1000 Weights\n",
      "episode 1009, initial_state (0, 6, 3), reward 253.0, memory_length 2000, epsilon 0.635051016398008 total_time 726.0\n",
      "episode 1019, initial_state (3, 12, 5), reward 633.0, memory_length 2000, epsilon 0.6321997075307504 total_time 723.0\n",
      "episode 1029, initial_state (0, 15, 2), reward 211.0, memory_length 2000, epsilon 0.6293612007271489 total_time 722.0\n",
      "episode 1039, initial_state (1, 9, 1), reward 573.0, memory_length 2000, epsilon 0.6265354385073433 total_time 727.0\n",
      "episode 1049, initial_state (3, 0, 1), reward 262.0, memory_length 2000, epsilon 0.6237223636495525 total_time 729.0\n",
      "episode 1059, initial_state (4, 3, 4), reward 450.0, memory_length 2000, epsilon 0.6209219191889144 total_time 731.0\n",
      "episode 1069, initial_state (2, 13, 3), reward 482.0, memory_length 2000, epsilon 0.6181340484163329 total_time 721.0\n",
      "episode 1079, initial_state (3, 11, 6), reward 775.0, memory_length 2000, epsilon 0.6153586948773296 total_time 724.0\n",
      "episode 1089, initial_state (1, 3, 5), reward 524.0, memory_length 2000, epsilon 0.6125958023709007 total_time 723.0\n",
      "episode 1099, initial_state (4, 14, 4), reward 536.0, memory_length 2000, epsilon 0.6098453149483783 total_time 727.0\n",
      "episode 1109, initial_state (2, 20, 5), reward 216.0, memory_length 2000, epsilon 0.6071071769122982 total_time 725.0\n",
      "episode 1119, initial_state (4, 12, 1), reward 716.0, memory_length 2000, epsilon 0.6043813328152715 total_time 722.0\n",
      "episode 1129, initial_state (3, 16, 4), reward 401.0, memory_length 2000, epsilon 0.6016677274588622 total_time 723.0\n",
      "episode 1139, initial_state (3, 15, 0), reward 462.0, memory_length 2000, epsilon 0.598966305892469 total_time 723.0\n",
      "episode 1149, initial_state (2, 3, 6), reward 349.0, memory_length 2000, epsilon 0.5962770134122131 total_time 722.0\n",
      "episode 1159, initial_state (3, 5, 1), reward 405.0, memory_length 2000, epsilon 0.5935997955598296 total_time 722.0\n",
      "episode 1169, initial_state (4, 6, 1), reward 492.0, memory_length 2000, epsilon 0.5909345981215656 total_time 721.0\n",
      "episode 1179, initial_state (1, 4, 6), reward 105.0, memory_length 2000, epsilon 0.5882813671270819 total_time 721.0\n",
      "episode 1189, initial_state (0, 22, 4), reward 550.0, memory_length 2000, epsilon 0.5856400488483602 total_time 723.0\n",
      "episode 1199, initial_state (3, 23, 5), reward 629.0, memory_length 2000, epsilon 0.5830105897986152 total_time 727.0\n",
      "episode 1209, initial_state (0, 11, 4), reward 580.0, memory_length 2000, epsilon 0.5803929367312111 total_time 723.0\n",
      "episode 1219, initial_state (1, 12, 6), reward 637.0, memory_length 2000, epsilon 0.577787036638584 total_time 723.0\n",
      "episode 1229, initial_state (0, 4, 3), reward 487.0, memory_length 2000, epsilon 0.5751928367511678 total_time 722.0\n",
      "episode 1239, initial_state (0, 21, 1), reward 312.0, memory_length 2000, epsilon 0.5726102845363265 total_time 723.0\n",
      "episode 1249, initial_state (1, 19, 2), reward 526.0, memory_length 2000, epsilon 0.570039327697289 total_time 726.0\n",
      "episode 1259, initial_state (3, 22, 2), reward 496.0, memory_length 2000, epsilon 0.5674799141720918 total_time 724.0\n",
      "episode 1269, initial_state (0, 15, 6), reward 532.0, memory_length 2000, epsilon 0.5649319921325234 total_time 722.0\n",
      "episode 1279, initial_state (3, 14, 5), reward 423.0, memory_length 2000, epsilon 0.5623955099830754 total_time 727.0\n",
      "episode 1289, initial_state (3, 8, 3), reward 407.0, memory_length 2000, epsilon 0.5598704163598978 total_time 726.0\n",
      "episode 1299, initial_state (3, 12, 3), reward 568.0, memory_length 2000, epsilon 0.5573566601297583 total_time 727.0\n",
      "episode 1309, initial_state (4, 17, 4), reward 617.0, memory_length 2000, epsilon 0.5548541903890073 total_time 723.0\n",
      "episode 1319, initial_state (1, 0, 3), reward 759.0, memory_length 2000, epsilon 0.5523629564625472 total_time 728.0\n",
      "episode 1329, initial_state (2, 2, 6), reward 398.0, memory_length 2000, epsilon 0.5498829079028057 total_time 724.0\n",
      "episode 1339, initial_state (2, 13, 0), reward 503.0, memory_length 2000, epsilon 0.5474139944887148 total_time 724.0\n",
      "episode 1349, initial_state (2, 15, 2), reward 693.0, memory_length 2000, epsilon 0.5449561662246933 total_time 733.0\n",
      "episode 1359, initial_state (3, 22, 6), reward 518.0, memory_length 2000, epsilon 0.5425093733396351 total_time 726.0\n",
      "episode 1369, initial_state (4, 5, 0), reward 767.0, memory_length 2000, epsilon 0.5400735662859008 total_time 721.0\n",
      "episode 1379, initial_state (4, 20, 3), reward 646.0, memory_length 2000, epsilon 0.537648695738314 total_time 721.0\n",
      "episode 1389, initial_state (3, 2, 3), reward 497.0, memory_length 2000, epsilon 0.5352347125931634 total_time 726.0\n",
      "episode 1399, initial_state (0, 4, 4), reward 603.0, memory_length 2000, epsilon 0.5328315679672077 total_time 729.0\n",
      "episode 1409, initial_state (0, 9, 6), reward 293.0, memory_length 2000, epsilon 0.5304392131966862 total_time 721.0\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "episode 1419, initial_state (3, 12, 3), reward 216.0, memory_length 2000, epsilon 0.5280575998363332 total_time 733.0\n",
      "episode 1429, initial_state (4, 2, 0), reward 117.0, memory_length 2000, epsilon 0.5256866796583964 total_time 721.0\n",
      "episode 1439, initial_state (1, 10, 1), reward 255.0, memory_length 2000, epsilon 0.5233264046516615 total_time 724.0\n",
      "episode 1449, initial_state (3, 5, 6), reward 541.0, memory_length 2000, epsilon 0.5209767270204789 total_time 723.0\n",
      "episode 1459, initial_state (1, 1, 3), reward 943.0, memory_length 2000, epsilon 0.5186375991837961 total_time 726.0\n",
      "episode 1469, initial_state (3, 0, 2), reward 361.0, memory_length 2000, epsilon 0.5163089737741946 total_time 725.0\n",
      "episode 1479, initial_state (2, 2, 1), reward 905.0, memory_length 2000, epsilon 0.5139908036369305 total_time 728.0\n",
      "episode 1489, initial_state (4, 18, 3), reward 471.0, memory_length 2000, epsilon 0.5116830418289788 total_time 722.0\n",
      "episode 1499, initial_state (4, 8, 0), reward 614.0, memory_length 2000, epsilon 0.5093856416180845 total_time 722.0\n",
      "episode 1509, initial_state (2, 17, 5), reward 1014.0, memory_length 2000, epsilon 0.5070985564818147 total_time 727.0\n",
      "episode 1519, initial_state (2, 0, 2), reward 755.0, memory_length 2000, epsilon 0.5048217401066171 total_time 722.0\n",
      "episode 1529, initial_state (2, 22, 4), reward 428.0, memory_length 2000, epsilon 0.5025551463868823 total_time 721.0\n",
      "episode 1539, initial_state (1, 10, 4), reward 519.0, memory_length 2000, epsilon 0.5002987294240101 total_time 723.0\n",
      "episode 1549, initial_state (2, 14, 2), reward 413.0, memory_length 2000, epsilon 0.49805244352547995 total_time 723.0\n",
      "episode 1559, initial_state (1, 6, 0), reward 630.0, memory_length 2000, epsilon 0.49581624320392553 total_time 726.0\n",
      "episode 1569, initial_state (4, 17, 6), reward 792.0, memory_length 2000, epsilon 0.49359008317621406 total_time 723.0\n",
      "episode 1579, initial_state (0, 14, 3), reward 237.0, memory_length 2000, epsilon 0.4913739183625287 total_time 721.0\n",
      "episode 1589, initial_state (0, 9, 4), reward 1016.0, memory_length 2000, epsilon 0.48916770388545644 total_time 722.0\n",
      "episode 1599, initial_state (1, 4, 6), reward 432.0, memory_length 2000, epsilon 0.4869713950690786 total_time 726.0\n",
      "episode 1609, initial_state (3, 19, 4), reward 343.0, memory_length 2000, epsilon 0.48478494743806666 total_time 723.0\n",
      "episode 1619, initial_state (2, 14, 4), reward 770.0, memory_length 2000, epsilon 0.48260831671678134 total_time 721.0\n",
      "episode 1629, initial_state (0, 13, 4), reward 675.0, memory_length 2000, epsilon 0.4804414588283762 total_time 731.0\n",
      "episode 1639, initial_state (2, 6, 5), reward 587.0, memory_length 2000, epsilon 0.4782843298939049 total_time 722.0\n",
      "episode 1649, initial_state (0, 14, 2), reward 624.0, memory_length 2000, epsilon 0.47613688623143285 total_time 722.0\n",
      "episode 1659, initial_state (3, 22, 1), reward 412.0, memory_length 2000, epsilon 0.4739990843551525 total_time 721.0\n",
      "episode 1669, initial_state (1, 11, 0), reward 606.0, memory_length 2000, epsilon 0.47187088097450275 total_time 726.0\n",
      "episode 1679, initial_state (2, 5, 6), reward 916.0, memory_length 2000, epsilon 0.4697522329932925 total_time 727.0\n",
      "episode 1689, initial_state (2, 18, 2), reward 246.0, memory_length 2000, epsilon 0.4676430975088276 total_time 722.0\n",
      "episode 1699, initial_state (2, 2, 6), reward 811.0, memory_length 2000, epsilon 0.4655434318110426 total_time 724.0\n",
      "episode 1709, initial_state (0, 2, 2), reward 582.0, memory_length 2000, epsilon 0.4634531933816352 total_time 725.0\n",
      "episode 1719, initial_state (3, 12, 6), reward 821.0, memory_length 2000, epsilon 0.4613723398932059 total_time 725.0\n",
      "episode 1729, initial_state (3, 19, 5), reward 559.0, memory_length 2000, epsilon 0.45930082920840043 total_time 725.0\n",
      "episode 1739, initial_state (0, 22, 0), reward 644.0, memory_length 2000, epsilon 0.4572386193790566 total_time 726.0\n",
      "episode 1749, initial_state (3, 20, 3), reward 797.0, memory_length 2000, epsilon 0.45518566864535476 total_time 728.0\n",
      "episode 1759, initial_state (4, 3, 6), reward 681.0, memory_length 2000, epsilon 0.45314193543497266 total_time 723.0\n",
      "episode 1769, initial_state (0, 18, 2), reward 586.0, memory_length 2000, epsilon 0.4511073783622429 total_time 726.0\n",
      "episode 1779, initial_state (2, 22, 0), reward 802.0, memory_length 2000, epsilon 0.44908195622731506 total_time 729.0\n",
      "episode 1789, initial_state (2, 22, 3), reward 630.0, memory_length 2000, epsilon 0.44706562801532185 total_time 721.0\n",
      "episode 1799, initial_state (3, 17, 1), reward 586.0, memory_length 2000, epsilon 0.44505835289554796 total_time 723.0\n",
      "episode 1809, initial_state (4, 6, 6), reward 339.0, memory_length 2000, epsilon 0.44306009022060383 total_time 732.0\n",
      "episode 1819, initial_state (2, 23, 2), reward 864.0, memory_length 2000, epsilon 0.44107079952560174 total_time 732.0\n",
      "episode 1829, initial_state (0, 17, 6), reward 572.0, memory_length 2000, epsilon 0.4390904405273373 total_time 723.0\n",
      "episode 1839, initial_state (2, 15, 6), reward 885.0, memory_length 2000, epsilon 0.4371189731234731 total_time 722.0\n",
      "episode 1849, initial_state (4, 22, 2), reward 670.0, memory_length 2000, epsilon 0.43515635739172687 total_time 721.0\n",
      "episode 1859, initial_state (0, 21, 0), reward 372.0, memory_length 2000, epsilon 0.43320255358906284 total_time 726.0\n",
      "episode 1869, initial_state (2, 17, 2), reward 1035.0, memory_length 2000, epsilon 0.4312575221508874 total_time 727.0\n",
      "episode 1879, initial_state (0, 1, 5), reward 743.0, memory_length 2000, epsilon 0.4293212236902474 total_time 726.0\n",
      "episode 1889, initial_state (2, 15, 3), reward 891.0, memory_length 2000, epsilon 0.4273936189970328 total_time 731.0\n",
      "episode 1899, initial_state (1, 0, 5), reward 529.0, memory_length 2000, epsilon 0.42547466903718284 total_time 727.0\n",
      "episode 1909, initial_state (2, 3, 4), reward 784.0, memory_length 2000, epsilon 0.4235643349518951 total_time 725.0\n",
      "episode 1919, initial_state (0, 23, 0), reward 1112.0, memory_length 2000, epsilon 0.42166257805683915 total_time 728.0\n",
      "episode 1929, initial_state (1, 14, 6), reward 994.0, memory_length 2000, epsilon 0.4197693598413728 total_time 722.0\n",
      "episode 1939, initial_state (0, 21, 1), reward 754.0, memory_length 2000, epsilon 0.4178846419677626 total_time 721.0\n",
      "episode 1949, initial_state (2, 6, 1), reward 479.0, memory_length 2000, epsilon 0.4160083862704071 total_time 727.0\n",
      "episode 1959, initial_state (0, 11, 2), reward 419.0, memory_length 2000, epsilon 0.41414055475506445 total_time 725.0\n",
      "episode 1969, initial_state (2, 12, 1), reward 695.0, memory_length 2000, epsilon 0.4122811095980825 total_time 723.0\n",
      "episode 1979, initial_state (3, 14, 2), reward 621.0, memory_length 2000, epsilon 0.4104300131456334 total_time 722.0\n",
      "episode 1989, initial_state (0, 20, 2), reward 1287.0, memory_length 2000, epsilon 0.40858722791295055 total_time 741.0\n",
      "episode 1999, initial_state (1, 3, 0), reward 955.0, memory_length 2000, epsilon 0.4067527165835702 total_time 732.0\n",
      "Saving Model 2000\n",
      "Saving Model 2000 Weights\n",
      "episode 2009, initial_state (1, 10, 5), reward 1026.0, memory_length 2000, epsilon 0.40492644200857514 total_time 729.0\n",
      "episode 2019, initial_state (2, 19, 4), reward 222.0, memory_length 2000, epsilon 0.4031083672058429 total_time 722.0\n",
      "episode 2029, initial_state (1, 6, 0), reward 828.0, memory_length 2000, epsilon 0.4012984553592966 total_time 721.0\n",
      "episode 2039, initial_state (0, 17, 3), reward 561.0, memory_length 2000, epsilon 0.3994966698181594 total_time 722.0\n",
      "episode 2049, initial_state (0, 6, 5), reward 484.0, memory_length 2000, epsilon 0.3977029740962125 total_time 721.0\n",
      "episode 2059, initial_state (3, 18, 6), reward 532.0, memory_length 2000, epsilon 0.39591733187105643 total_time 728.0\n",
      "episode 2069, initial_state (0, 13, 2), reward 692.0, memory_length 2000, epsilon 0.394139706983375 total_time 723.0\n",
      "episode 2079, initial_state (3, 20, 5), reward 825.0, memory_length 2000, epsilon 0.39237006343620334 total_time 721.0\n",
      "episode 2089, initial_state (3, 20, 5), reward 978.0, memory_length 2000, epsilon 0.39060836539419946 total_time 725.0\n",
      "episode 2099, initial_state (0, 7, 2), reward 735.0, memory_length 2000, epsilon 0.3888545771829175 total_time 731.0\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "episode 2109, initial_state (3, 7, 3), reward 964.0, memory_length 2000, epsilon 0.38710866328808646 total_time 727.0\n",
      "episode 2119, initial_state (3, 2, 2), reward 691.0, memory_length 2000, epsilon 0.38537058835489024 total_time 723.0\n",
      "episode 2129, initial_state (1, 6, 0), reward 860.0, memory_length 2000, epsilon 0.38364031718725206 total_time 725.0\n",
      "episode 2139, initial_state (0, 19, 3), reward 927.0, memory_length 2000, epsilon 0.3819178147471215 total_time 732.0\n",
      "episode 2149, initial_state (2, 7, 1), reward 757.0, memory_length 2000, epsilon 0.3802030461537656 total_time 731.0\n",
      "episode 2159, initial_state (4, 12, 2), reward 625.0, memory_length 2000, epsilon 0.37849597668306145 total_time 727.0\n",
      "episode 2169, initial_state (1, 15, 2), reward 862.0, memory_length 2000, epsilon 0.37679657176679415 total_time 722.0\n",
      "episode 2179, initial_state (4, 6, 2), reward 861.0, memory_length 2000, epsilon 0.3751047969919559 total_time 725.0\n",
      "episode 2189, initial_state (0, 6, 3), reward 864.0, memory_length 2000, epsilon 0.37342061810004984 total_time 726.0\n",
      "episode 2199, initial_state (2, 4, 1), reward 779.0, memory_length 2000, epsilon 0.37174400098639576 total_time 721.0\n",
      "episode 2209, initial_state (0, 19, 2), reward 878.0, memory_length 2000, epsilon 0.37007491169943985 total_time 726.0\n",
      "episode 2219, initial_state (4, 7, 4), reward 1189.0, memory_length 2000, epsilon 0.36841331644006703 total_time 725.0\n",
      "episode 2229, initial_state (2, 20, 3), reward 1301.0, memory_length 2000, epsilon 0.36675918156091647 total_time 727.0\n",
      "episode 2239, initial_state (0, 18, 5), reward 1035.0, memory_length 2000, epsilon 0.3651124735657004 total_time 732.0\n",
      "episode 2249, initial_state (3, 8, 5), reward 761.0, memory_length 2000, epsilon 0.36347315910852557 total_time 729.0\n",
      "episode 2259, initial_state (4, 18, 0), reward 800.0, memory_length 2000, epsilon 0.36184120499321826 total_time 722.0\n",
      "episode 2269, initial_state (4, 0, 1), reward 718.0, memory_length 2000, epsilon 0.3602165781726519 total_time 728.0\n",
      "episode 2279, initial_state (3, 14, 6), reward 1088.0, memory_length 2000, epsilon 0.35859924574807783 total_time 721.0\n",
      "episode 2289, initial_state (3, 16, 5), reward 955.0, memory_length 2000, epsilon 0.35698917496845917 total_time 727.0\n",
      "episode 2299, initial_state (0, 9, 1), reward 946.0, memory_length 2000, epsilon 0.3553863332298075 total_time 727.0\n",
      "episode 2309, initial_state (1, 18, 5), reward 900.0, memory_length 2000, epsilon 0.3537906880745231 total_time 726.0\n",
      "episode 2319, initial_state (4, 21, 4), reward 1168.0, memory_length 2000, epsilon 0.35220220719073686 total_time 723.0\n",
      "episode 2329, initial_state (0, 8, 5), reward 954.0, memory_length 2000, epsilon 0.35062085841165663 total_time 730.0\n",
      "episode 2339, initial_state (3, 17, 2), reward 824.0, memory_length 2000, epsilon 0.3490466097149157 total_time 721.0\n",
      "episode 2349, initial_state (2, 5, 4), reward 981.0, memory_length 2000, epsilon 0.34747942922192393 total_time 723.0\n",
      "episode 2359, initial_state (4, 16, 5), reward 1018.0, memory_length 2000, epsilon 0.34591928519722304 total_time 723.0\n",
      "episode 2369, initial_state (0, 3, 3), reward 873.0, memory_length 2000, epsilon 0.34436614604784305 total_time 721.0\n",
      "episode 2379, initial_state (1, 5, 2), reward 1157.0, memory_length 2000, epsilon 0.3428199803226632 total_time 732.0\n",
      "episode 2389, initial_state (0, 4, 1), reward 800.0, memory_length 2000, epsilon 0.3412807567117746 total_time 721.0\n",
      "episode 2399, initial_state (1, 2, 0), reward 331.0, memory_length 2000, epsilon 0.33974844404584675 total_time 726.0\n",
      "episode 2409, initial_state (4, 13, 4), reward 650.0, memory_length 2000, epsilon 0.3382230112954956 total_time 722.0\n",
      "episode 2419, initial_state (0, 2, 2), reward 1158.0, memory_length 2000, epsilon 0.336704427570656 total_time 725.0\n",
      "episode 2429, initial_state (4, 23, 3), reward 942.0, memory_length 2000, epsilon 0.3351926621199554 total_time 721.0\n",
      "episode 2439, initial_state (2, 17, 1), reward 909.0, memory_length 2000, epsilon 0.333687684330092 total_time 729.0\n",
      "episode 2449, initial_state (1, 19, 6), reward 694.0, memory_length 2000, epsilon 0.33218946372521396 total_time 732.0\n",
      "episode 2459, initial_state (0, 14, 2), reward 830.0, memory_length 2000, epsilon 0.3306979699663029 total_time 724.0\n",
      "episode 2469, initial_state (4, 0, 4), reward 738.0, memory_length 2000, epsilon 0.3292131728505593 total_time 726.0\n",
      "episode 2479, initial_state (4, 0, 2), reward 804.0, memory_length 2000, epsilon 0.32773504231079065 total_time 722.0\n",
      "episode 2489, initial_state (3, 16, 3), reward 1080.0, memory_length 2000, epsilon 0.3262635484148032 total_time 726.0\n",
      "episode 2499, initial_state (0, 14, 4), reward 816.0, memory_length 2000, epsilon 0.3247986613647952 total_time 723.0\n",
      "episode 2509, initial_state (4, 16, 4), reward 1004.0, memory_length 2000, epsilon 0.3233403514967538 total_time 727.0\n",
      "episode 2519, initial_state (4, 0, 4), reward 1224.0, memory_length 2000, epsilon 0.32188858927985425 total_time 731.0\n",
      "episode 2529, initial_state (1, 2, 6), reward 1131.0, memory_length 2000, epsilon 0.32044334531586227 total_time 729.0\n",
      "episode 2539, initial_state (1, 14, 2), reward 550.0, memory_length 2000, epsilon 0.3190045903385381 total_time 724.0\n",
      "episode 2549, initial_state (1, 21, 5), reward 825.0, memory_length 2000, epsilon 0.31757229521304425 total_time 723.0\n",
      "episode 2559, initial_state (0, 23, 3), reward 924.0, memory_length 2000, epsilon 0.3161464309353555 total_time 722.0\n",
      "episode 2569, initial_state (4, 18, 6), reward 1085.0, memory_length 2000, epsilon 0.3147269686316715 total_time 722.0\n",
      "episode 2579, initial_state (0, 10, 6), reward 1046.0, memory_length 2000, epsilon 0.3133138795578322 total_time 721.0\n",
      "episode 2589, initial_state (4, 16, 4), reward 865.0, memory_length 2000, epsilon 0.31190713509873536 total_time 728.0\n",
      "episode 2599, initial_state (0, 12, 3), reward 1304.0, memory_length 2000, epsilon 0.31050670676775777 total_time 722.0\n",
      "episode 2609, initial_state (4, 17, 5), reward 1087.0, memory_length 2000, epsilon 0.30911256620617783 total_time 721.0\n",
      "episode 2619, initial_state (3, 21, 2), reward 956.0, memory_length 2000, epsilon 0.3077246851826015 total_time 729.0\n",
      "episode 2629, initial_state (1, 4, 3), reward 937.0, memory_length 2000, epsilon 0.3063430355923907 total_time 727.0\n",
      "episode 2639, initial_state (4, 16, 5), reward 1136.0, memory_length 2000, epsilon 0.3049675894570939 total_time 721.0\n",
      "episode 2649, initial_state (0, 1, 6), reward 1107.0, memory_length 2000, epsilon 0.3035983189238799 total_time 729.0\n",
      "episode 2659, initial_state (2, 20, 6), reward 969.0, memory_length 2000, epsilon 0.3022351962649737 total_time 722.0\n",
      "episode 2669, initial_state (1, 5, 0), reward 1183.0, memory_length 2000, epsilon 0.3008781938770948 total_time 727.0\n",
      "episode 2679, initial_state (2, 3, 4), reward 1063.0, memory_length 2000, epsilon 0.29952728428089853 total_time 728.0\n",
      "episode 2689, initial_state (2, 21, 3), reward 811.0, memory_length 2000, epsilon 0.2981824401204193 total_time 731.0\n",
      "episode 2699, initial_state (2, 4, 1), reward 843.0, memory_length 2000, epsilon 0.29684363416251697 total_time 722.0\n",
      "episode 2709, initial_state (3, 7, 5), reward 1137.0, memory_length 2000, epsilon 0.29551083929632516 total_time 726.0\n",
      "episode 2719, initial_state (3, 23, 2), reward 793.0, memory_length 2000, epsilon 0.2941840285327023 total_time 725.0\n",
      "episode 2729, initial_state (4, 7, 1), reward 802.0, memory_length 2000, epsilon 0.292863175003685 total_time 724.0\n",
      "episode 2739, initial_state (1, 13, 3), reward 1200.0, memory_length 2000, epsilon 0.29154825196194417 total_time 721.0\n",
      "episode 2749, initial_state (3, 6, 2), reward 797.0, memory_length 2000, epsilon 0.2902392327802434 total_time 725.0\n",
      "episode 2759, initial_state (2, 20, 4), reward 997.0, memory_length 2000, epsilon 0.28893609095089945 total_time 723.0\n",
      "episode 2769, initial_state (4, 2, 5), reward 912.0, memory_length 2000, epsilon 0.28763880008524567 total_time 721.0\n",
      "episode 2779, initial_state (2, 6, 4), reward 1319.0, memory_length 2000, epsilon 0.2863473339130978 total_time 721.0\n",
      "episode 2789, initial_state (4, 1, 4), reward 834.0, memory_length 2000, epsilon 0.28506166628222174 total_time 722.0\n",
      "episode 2799, initial_state (4, 8, 4), reward 1083.0, memory_length 2000, epsilon 0.283781771157804 total_time 727.0\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "episode 2809, initial_state (0, 13, 1), reward 1070.0, memory_length 2000, epsilon 0.2825076226219246 total_time 721.0\n",
      "episode 2819, initial_state (4, 0, 5), reward 846.0, memory_length 2000, epsilon 0.28123919487303195 total_time 726.0\n",
      "episode 2829, initial_state (0, 14, 1), reward 629.0, memory_length 2000, epsilon 0.2799764622254211 total_time 721.0\n",
      "episode 2839, initial_state (2, 21, 0), reward 897.0, memory_length 2000, epsilon 0.2787193991087126 total_time 728.0\n",
      "episode 2849, initial_state (3, 2, 5), reward 915.0, memory_length 2000, epsilon 0.2774679800673354 total_time 725.0\n",
      "episode 2859, initial_state (2, 11, 4), reward 1180.0, memory_length 2000, epsilon 0.2762221797600112 total_time 721.0\n",
      "episode 2869, initial_state (3, 20, 5), reward 1099.0, memory_length 2000, epsilon 0.2749819729592411 total_time 724.0\n",
      "episode 2879, initial_state (4, 16, 2), reward 784.0, memory_length 2000, epsilon 0.2737473345507952 total_time 724.0\n",
      "episode 2889, initial_state (2, 13, 4), reward 465.0, memory_length 2000, epsilon 0.27251823953320337 total_time 721.0\n",
      "episode 2899, initial_state (0, 14, 2), reward 845.0, memory_length 2000, epsilon 0.27129466301724947 total_time 721.0\n",
      "episode 2909, initial_state (0, 2, 3), reward 1107.0, memory_length 2000, epsilon 0.2700765802254673 total_time 724.0\n",
      "episode 2919, initial_state (1, 5, 3), reward 1176.0, memory_length 2000, epsilon 0.26886396649163885 total_time 726.0\n",
      "episode 2929, initial_state (2, 18, 0), reward 963.0, memory_length 2000, epsilon 0.2676567972602944 total_time 724.0\n",
      "episode 2939, initial_state (1, 11, 1), reward 794.0, memory_length 2000, epsilon 0.26645504808621584 total_time 725.0\n",
      "episode 2949, initial_state (4, 17, 5), reward 977.0, memory_length 2000, epsilon 0.2652586946339412 total_time 727.0\n",
      "episode 2959, initial_state (2, 7, 6), reward 933.0, memory_length 2000, epsilon 0.26406771267727236 total_time 723.0\n",
      "episode 2969, initial_state (2, 8, 4), reward 700.0, memory_length 2000, epsilon 0.2628820780987839 total_time 723.0\n",
      "episode 2979, initial_state (4, 10, 3), reward 795.0, memory_length 2000, epsilon 0.26170176688933516 total_time 725.0\n",
      "episode 2989, initial_state (4, 16, 4), reward 1081.0, memory_length 2000, epsilon 0.2605267551475837 total_time 726.0\n",
      "episode 2999, initial_state (2, 1, 3), reward 560.0, memory_length 2000, epsilon 0.2593570190795017 total_time 722.0\n",
      "Saving Model 3000\n",
      "Saving Model 3000 Weights\n",
      "episode 3009, initial_state (1, 15, 2), reward 1278.0, memory_length 2000, epsilon 0.2581925349978938 total_time 721.0\n",
      "episode 3019, initial_state (4, 23, 6), reward 1121.0, memory_length 2000, epsilon 0.25703327932191755 total_time 726.0\n",
      "episode 3029, initial_state (1, 1, 2), reward 1091.0, memory_length 2000, epsilon 0.25587922857660583 total_time 723.0\n",
      "episode 3039, initial_state (2, 20, 2), reward 995.0, memory_length 2000, epsilon 0.2547303593923917 total_time 726.0\n",
      "episode 3049, initial_state (1, 20, 2), reward 793.0, memory_length 2000, epsilon 0.2535866485046348 total_time 728.0\n",
      "episode 3059, initial_state (4, 10, 5), reward 1018.0, memory_length 2000, epsilon 0.25244807275315073 total_time 722.0\n",
      "episode 3069, initial_state (2, 16, 1), reward 955.0, memory_length 2000, epsilon 0.2513146090817416 total_time 724.0\n",
      "episode 3079, initial_state (4, 20, 0), reward 1067.0, memory_length 2000, epsilon 0.2501862345377291 total_time 723.0\n",
      "episode 3089, initial_state (2, 10, 4), reward 1169.0, memory_length 2000, epsilon 0.24906292627149043 total_time 726.0\n",
      "episode 3099, initial_state (1, 22, 5), reward 992.0, memory_length 2000, epsilon 0.2479446615359947 total_time 722.0\n",
      "episode 3109, initial_state (3, 14, 1), reward 1095.0, memory_length 2000, epsilon 0.2468314176863428 total_time 723.0\n",
      "episode 3119, initial_state (0, 3, 3), reward 1026.0, memory_length 2000, epsilon 0.2457231721793088 total_time 721.0\n",
      "episode 3129, initial_state (1, 3, 5), reward 968.0, memory_length 2000, epsilon 0.24461990257288316 total_time 729.0\n",
      "episode 3139, initial_state (4, 13, 2), reward 1058.0, memory_length 2000, epsilon 0.2435215865258188 total_time 725.0\n",
      "episode 3149, initial_state (0, 10, 2), reward 702.0, memory_length 2000, epsilon 0.24242820179717817 total_time 726.0\n",
      "episode 3159, initial_state (0, 20, 3), reward 1045.0, memory_length 2000, epsilon 0.24133972624588318 total_time 726.0\n",
      "episode 3169, initial_state (1, 18, 1), reward 1134.0, memory_length 2000, epsilon 0.24025613783026661 total_time 725.0\n",
      "episode 3179, initial_state (2, 22, 0), reward 1100.0, memory_length 2000, epsilon 0.23917741460762618 total_time 726.0\n",
      "episode 3189, initial_state (1, 14, 4), reward 595.0, memory_length 2000, epsilon 0.23810353473377974 total_time 726.0\n",
      "episode 3199, initial_state (0, 1, 1), reward 1070.0, memory_length 2000, epsilon 0.23703447646262304 total_time 726.0\n",
      "episode 3209, initial_state (1, 5, 1), reward 1135.0, memory_length 2000, epsilon 0.23597021814568964 total_time 723.0\n",
      "episode 3219, initial_state (0, 17, 3), reward 941.0, memory_length 2000, epsilon 0.23491073823171224 total_time 727.0\n",
      "episode 3229, initial_state (1, 2, 0), reward 1311.0, memory_length 2000, epsilon 0.23385601526618638 total_time 728.0\n",
      "episode 3239, initial_state (4, 16, 4), reward 802.0, memory_length 2000, epsilon 0.23280602789093593 total_time 721.0\n",
      "episode 3249, initial_state (2, 20, 3), reward 906.0, memory_length 2000, epsilon 0.23176075484368075 total_time 722.0\n",
      "episode 3259, initial_state (0, 9, 6), reward 1328.0, memory_length 2000, epsilon 0.23072017495760577 total_time 735.0\n",
      "episode 3269, initial_state (1, 14, 2), reward 1400.0, memory_length 2000, epsilon 0.2296842671609329 total_time 723.0\n",
      "episode 3279, initial_state (4, 19, 4), reward 1175.0, memory_length 2000, epsilon 0.22865301047649375 total_time 727.0\n",
      "episode 3289, initial_state (0, 5, 5), reward 1235.0, memory_length 2000, epsilon 0.22762638402130528 total_time 726.0\n",
      "episode 3299, initial_state (1, 14, 0), reward 814.0, memory_length 2000, epsilon 0.2266043670061466 total_time 721.0\n",
      "episode 3309, initial_state (1, 17, 0), reward 1180.0, memory_length 2000, epsilon 0.22558693873513835 total_time 726.0\n",
      "episode 3319, initial_state (4, 2, 4), reward 1036.0, memory_length 2000, epsilon 0.22457407860532322 total_time 724.0\n",
      "episode 3329, initial_state (4, 12, 5), reward 888.0, memory_length 2000, epsilon 0.22356576610624895 total_time 727.0\n",
      "episode 3339, initial_state (0, 13, 6), reward 929.0, memory_length 2000, epsilon 0.22256198081955297 total_time 723.0\n",
      "episode 3349, initial_state (3, 11, 1), reward 1171.0, memory_length 2000, epsilon 0.221562702418549 total_time 723.0\n",
      "episode 3359, initial_state (3, 11, 0), reward 1261.0, memory_length 2000, epsilon 0.2205679106678152 total_time 727.0\n",
      "episode 3369, initial_state (1, 17, 0), reward 1077.0, memory_length 2000, epsilon 0.21957758542278463 total_time 723.0\n",
      "episode 3379, initial_state (3, 1, 6), reward 1024.0, memory_length 2000, epsilon 0.21859170662933722 total_time 727.0\n",
      "episode 3389, initial_state (2, 23, 0), reward 1338.0, memory_length 2000, epsilon 0.21761025432339381 total_time 724.0\n",
      "episode 3399, initial_state (4, 10, 0), reward 1140.0, memory_length 2000, epsilon 0.21663320863051158 total_time 722.0\n",
      "episode 3409, initial_state (0, 10, 1), reward 689.0, memory_length 2000, epsilon 0.21566054976548188 total_time 722.0\n",
      "episode 3419, initial_state (0, 9, 5), reward 1093.0, memory_length 2000, epsilon 0.21469225803192948 total_time 723.0\n",
      "episode 3429, initial_state (2, 14, 2), reward 962.0, memory_length 2000, epsilon 0.2137283138219136 total_time 723.0\n",
      "episode 3439, initial_state (0, 9, 5), reward 1171.0, memory_length 2000, epsilon 0.21276869761553116 total_time 729.0\n",
      "episode 3449, initial_state (2, 3, 5), reward 898.0, memory_length 2000, epsilon 0.21181338998052113 total_time 722.0\n",
      "episode 3459, initial_state (0, 19, 5), reward 977.0, memory_length 2000, epsilon 0.2108623715718713 total_time 725.0\n",
      "episode 3469, initial_state (3, 11, 2), reward 1197.0, memory_length 2000, epsilon 0.20991562313142628 total_time 728.0\n",
      "episode 3479, initial_state (2, 10, 0), reward 744.0, memory_length 2000, epsilon 0.20897312548749794 total_time 724.0\n",
      "episode 3489, initial_state (4, 10, 2), reward 1164.0, memory_length 2000, epsilon 0.20803485955447676 total_time 722.0\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "episode 3499, initial_state (0, 19, 0), reward 1149.0, memory_length 2000, epsilon 0.2071008063324455 total_time 723.0\n",
      "episode 3509, initial_state (4, 8, 1), reward 955.0, memory_length 2000, epsilon 0.2061709469067944 total_time 723.0\n",
      "episode 3519, initial_state (1, 2, 3), reward 1324.0, memory_length 2000, epsilon 0.20524526244783847 total_time 729.0\n",
      "episode 3529, initial_state (4, 11, 4), reward 784.0, memory_length 2000, epsilon 0.20432373421043573 total_time 721.0\n",
      "episode 3539, initial_state (1, 11, 0), reward 874.0, memory_length 2000, epsilon 0.2034063435336079 total_time 725.0\n",
      "episode 3549, initial_state (1, 4, 5), reward 1044.0, memory_length 2000, epsilon 0.20249307184016238 total_time 721.0\n",
      "episode 3559, initial_state (4, 1, 6), reward 1116.0, memory_length 2000, epsilon 0.20158390063631618 total_time 735.0\n",
      "episode 3569, initial_state (3, 13, 1), reward 990.0, memory_length 2000, epsilon 0.2006788115113214 total_time 723.0\n",
      "episode 3579, initial_state (3, 11, 3), reward 935.0, memory_length 2000, epsilon 0.19977778613709227 total_time 727.0\n",
      "episode 3589, initial_state (0, 20, 4), reward 849.0, memory_length 2000, epsilon 0.19888080626783422 total_time 722.0\n",
      "episode 3599, initial_state (3, 1, 0), reward 1110.0, memory_length 2000, epsilon 0.19798785373967417 total_time 725.0\n",
      "episode 3609, initial_state (0, 16, 1), reward 1215.0, memory_length 2000, epsilon 0.19709891047029302 total_time 730.0\n",
      "episode 3619, initial_state (0, 5, 1), reward 1095.0, memory_length 2000, epsilon 0.19621395845855913 total_time 726.0\n",
      "episode 3629, initial_state (3, 11, 2), reward 1180.0, memory_length 2000, epsilon 0.195332979784164 total_time 721.0\n",
      "episode 3639, initial_state (1, 20, 1), reward 1315.0, memory_length 2000, epsilon 0.1944559566072594 total_time 726.0\n",
      "episode 3649, initial_state (3, 2, 5), reward 1210.0, memory_length 2000, epsilon 0.19358287116809603 total_time 722.0\n",
      "episode 3659, initial_state (4, 20, 6), reward 1035.0, memory_length 2000, epsilon 0.1927137057866639 total_time 725.0\n",
      "episode 3669, initial_state (0, 0, 2), reward 804.0, memory_length 2000, epsilon 0.19184844286233438 total_time 723.0\n",
      "episode 3679, initial_state (2, 9, 5), reward 1184.0, memory_length 2000, epsilon 0.1909870648735036 total_time 725.0\n",
      "episode 3689, initial_state (2, 9, 5), reward 1050.0, memory_length 2000, epsilon 0.19012955437723789 total_time 726.0\n",
      "episode 3699, initial_state (3, 21, 4), reward 949.0, memory_length 2000, epsilon 0.18927589400892042 total_time 722.0\n",
      "episode 3709, initial_state (0, 0, 5), reward 1214.0, memory_length 2000, epsilon 0.18842606648189958 total_time 721.0\n",
      "episode 3719, initial_state (4, 8, 0), reward 1196.0, memory_length 2000, epsilon 0.18758005458713883 total_time 724.0\n",
      "episode 3729, initial_state (1, 20, 2), reward 1266.0, memory_length 2000, epsilon 0.1867378411928684 total_time 722.0\n",
      "episode 3739, initial_state (2, 19, 2), reward 1312.0, memory_length 2000, epsilon 0.18589940924423837 total_time 723.0\n",
      "episode 3749, initial_state (3, 4, 1), reward 1283.0, memory_length 2000, epsilon 0.18506474176297302 total_time 729.0\n",
      "episode 3759, initial_state (1, 3, 1), reward 1322.0, memory_length 2000, epsilon 0.1842338218470274 total_time 723.0\n",
      "episode 3769, initial_state (4, 17, 2), reward 1218.0, memory_length 2000, epsilon 0.18340663267024482 total_time 721.0\n",
      "episode 3779, initial_state (4, 1, 6), reward 1215.0, memory_length 2000, epsilon 0.18258315748201615 total_time 725.0\n",
      "episode 3789, initial_state (0, 19, 6), reward 1153.0, memory_length 2000, epsilon 0.18176337960694067 total_time 722.0\n",
      "episode 3799, initial_state (0, 5, 3), reward 1150.0, memory_length 2000, epsilon 0.18094728244448846 total_time 726.0\n",
      "episode 3809, initial_state (1, 7, 4), reward 1376.0, memory_length 2000, epsilon 0.18013484946866407 total_time 726.0\n",
      "episode 3819, initial_state (1, 11, 6), reward 1565.0, memory_length 2000, epsilon 0.17932606422767192 total_time 725.0\n",
      "episode 3829, initial_state (4, 14, 4), reward 1125.0, memory_length 2000, epsilon 0.17852091034358333 total_time 721.0\n",
      "episode 3839, initial_state (1, 21, 0), reward 1310.0, memory_length 2000, epsilon 0.17771937151200456 total_time 727.0\n",
      "episode 3849, initial_state (1, 3, 5), reward 1126.0, memory_length 2000, epsilon 0.17692143150174697 total_time 722.0\n",
      "episode 3859, initial_state (1, 12, 3), reward 1043.0, memory_length 2000, epsilon 0.17612707415449796 total_time 728.0\n",
      "episode 3869, initial_state (2, 21, 3), reward 1122.0, memory_length 2000, epsilon 0.1753362833844942 total_time 723.0\n",
      "episode 3879, initial_state (4, 14, 3), reward 965.0, memory_length 2000, epsilon 0.17454904317819556 total_time 724.0\n",
      "episode 3889, initial_state (3, 3, 2), reward 1311.0, memory_length 2000, epsilon 0.17376533759396093 total_time 721.0\n",
      "episode 3899, initial_state (2, 22, 4), reward 1213.0, memory_length 2000, epsilon 0.17298515076172546 total_time 723.0\n",
      "episode 3909, initial_state (0, 18, 4), reward 1141.0, memory_length 2000, epsilon 0.17220846688267916 total_time 725.0\n",
      "episode 3919, initial_state (2, 9, 1), reward 1283.0, memory_length 2000, epsilon 0.17143527022894695 total_time 729.0\n",
      "episode 3929, initial_state (1, 12, 1), reward 1217.0, memory_length 2000, epsilon 0.1706655451432701 total_time 723.0\n",
      "episode 3939, initial_state (0, 22, 5), reward 709.0, memory_length 2000, epsilon 0.16989927603868935 total_time 724.0\n",
      "episode 3949, initial_state (4, 6, 1), reward 1063.0, memory_length 2000, epsilon 0.16913644739822917 total_time 726.0\n",
      "episode 3959, initial_state (0, 15, 3), reward 1322.0, memory_length 2000, epsilon 0.16837704377458354 total_time 724.0\n",
      "episode 3969, initial_state (4, 1, 0), reward 1248.0, memory_length 2000, epsilon 0.16762104978980308 total_time 724.0\n",
      "episode 3979, initial_state (0, 5, 0), reward 820.0, memory_length 2000, epsilon 0.16686845013498383 total_time 724.0\n",
      "episode 3989, initial_state (2, 3, 2), reward 1105.0, memory_length 2000, epsilon 0.16611922956995692 total_time 724.0\n",
      "episode 3999, initial_state (1, 12, 4), reward 1081.0, memory_length 2000, epsilon 0.1653733729229805 total_time 722.0\n",
      "Saving Model 4000\n",
      "Saving Model 4000 Weights\n",
      "episode 4009, initial_state (3, 13, 4), reward 1541.0, memory_length 2000, epsilon 0.16463086509043187 total_time 722.0\n",
      "episode 4019, initial_state (3, 2, 4), reward 1200.0, memory_length 2000, epsilon 0.1638916910365021 total_time 726.0\n",
      "episode 4029, initial_state (3, 23, 2), reward 1004.0, memory_length 2000, epsilon 0.16315583579289122 total_time 729.0\n",
      "episode 4039, initial_state (3, 13, 2), reward 851.0, memory_length 2000, epsilon 0.16242328445850554 total_time 728.0\n",
      "episode 4049, initial_state (1, 19, 5), reward 1005.0, memory_length 2000, epsilon 0.1616940221991555 total_time 726.0\n",
      "episode 4059, initial_state (0, 5, 2), reward 1337.0, memory_length 2000, epsilon 0.1609680342472553 total_time 726.0\n",
      "episode 4069, initial_state (3, 7, 1), reward 1099.0, memory_length 2000, epsilon 0.16024530590152425 total_time 730.0\n",
      "episode 4079, initial_state (3, 2, 0), reward 1179.0, memory_length 2000, epsilon 0.1595258225266885 total_time 728.0\n",
      "episode 4089, initial_state (3, 7, 6), reward 817.0, memory_length 2000, epsilon 0.15880956955318531 total_time 721.0\n",
      "episode 4099, initial_state (1, 12, 4), reward 1186.0, memory_length 2000, epsilon 0.15809653247686736 total_time 721.0\n",
      "episode 4109, initial_state (0, 23, 5), reward 1342.0, memory_length 2000, epsilon 0.15738669685870949 total_time 724.0\n",
      "episode 4119, initial_state (2, 20, 0), reward 867.0, memory_length 2000, epsilon 0.1566800483245162 total_time 721.0\n",
      "episode 4129, initial_state (2, 23, 2), reward 1021.0, memory_length 2000, epsilon 0.1559765725646306 total_time 721.0\n",
      "episode 4139, initial_state (4, 18, 4), reward 1161.0, memory_length 2000, epsilon 0.15527625533364442 total_time 728.0\n",
      "episode 4149, initial_state (3, 15, 5), reward 1399.0, memory_length 2000, epsilon 0.15457908245010982 total_time 722.0\n",
      "episode 4159, initial_state (2, 0, 3), reward 1114.0, memory_length 2000, epsilon 0.15388503979625207 total_time 723.0\n",
      "episode 4169, initial_state (0, 21, 5), reward 1149.0, memory_length 2000, epsilon 0.15319411331768376 total_time 722.0\n",
      "episode 4179, initial_state (0, 4, 3), reward 1000.0, memory_length 2000, epsilon 0.15250628902312005 total_time 727.0\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "episode 4189, initial_state (2, 4, 2), reward 1287.0, memory_length 2000, epsilon 0.1518215529840955 total_time 727.0\n",
      "episode 4199, initial_state (2, 14, 5), reward 960.0, memory_length 2000, epsilon 0.1511398913346819 total_time 727.0\n",
      "episode 4209, initial_state (1, 17, 0), reward 938.0, memory_length 2000, epsilon 0.15046129027120755 total_time 722.0\n",
      "episode 4219, initial_state (3, 1, 5), reward 1025.0, memory_length 2000, epsilon 0.14978573605197776 total_time 721.0\n",
      "episode 4229, initial_state (4, 3, 0), reward 1078.0, memory_length 2000, epsilon 0.14911321499699645 total_time 722.0\n",
      "episode 4239, initial_state (1, 1, 0), reward 1286.0, memory_length 2000, epsilon 0.14844371348768934 total_time 726.0\n",
      "episode 4249, initial_state (3, 6, 3), reward 857.0, memory_length 2000, epsilon 0.1477772179666279 total_time 727.0\n",
      "episode 4259, initial_state (0, 23, 3), reward 1114.0, memory_length 2000, epsilon 0.14711371493725517 total_time 723.0\n",
      "episode 4269, initial_state (1, 8, 4), reward 1224.0, memory_length 2000, epsilon 0.14645319096361206 total_time 727.0\n",
      "episode 4279, initial_state (2, 20, 1), reward 1032.0, memory_length 2000, epsilon 0.1457956326700655 total_time 724.0\n",
      "episode 4289, initial_state (2, 17, 0), reward 1272.0, memory_length 2000, epsilon 0.14514102674103765 total_time 724.0\n",
      "episode 4299, initial_state (3, 13, 4), reward 1396.0, memory_length 2000, epsilon 0.14448935992073608 total_time 725.0\n",
      "episode 4309, initial_state (0, 10, 6), reward 864.0, memory_length 2000, epsilon 0.14384061901288533 total_time 729.0\n",
      "episode 4319, initial_state (2, 18, 1), reward 756.0, memory_length 2000, epsilon 0.1431947908804599 total_time 721.0\n",
      "episode 4329, initial_state (4, 23, 3), reward 934.0, memory_length 2000, epsilon 0.14255186244541806 total_time 723.0\n",
      "episode 4339, initial_state (2, 7, 3), reward 1382.0, memory_length 2000, epsilon 0.14191182068843697 total_time 730.0\n",
      "episode 4349, initial_state (1, 8, 6), reward 1585.0, memory_length 2000, epsilon 0.14127465264864922 total_time 729.0\n",
      "episode 4359, initial_state (0, 10, 6), reward 1080.0, memory_length 2000, epsilon 0.14064034542338025 total_time 722.0\n",
      "episode 4369, initial_state (0, 9, 2), reward 1227.0, memory_length 2000, epsilon 0.14000888616788704 total_time 721.0\n",
      "episode 4379, initial_state (4, 7, 4), reward 1027.0, memory_length 2000, epsilon 0.13938026209509807 total_time 724.0\n",
      "episode 4389, initial_state (3, 10, 0), reward 1107.0, memory_length 2000, epsilon 0.13875446047535442 total_time 723.0\n",
      "episode 4399, initial_state (0, 10, 0), reward 1133.0, memory_length 2000, epsilon 0.13813146863615192 total_time 722.0\n",
      "episode 4409, initial_state (3, 3, 2), reward 1338.0, memory_length 2000, epsilon 0.1375112739618845 total_time 727.0\n",
      "episode 4419, initial_state (4, 13, 2), reward 1109.0, memory_length 2000, epsilon 0.13689386389358882 total_time 725.0\n",
      "episode 4429, initial_state (1, 21, 2), reward 1102.0, memory_length 2000, epsilon 0.13627922592868993 total_time 726.0\n",
      "episode 4439, initial_state (3, 18, 0), reward 1019.0, memory_length 2000, epsilon 0.13566734762074797 total_time 724.0\n",
      "episode 4449, initial_state (4, 14, 3), reward 1314.0, memory_length 2000, epsilon 0.13505821657920636 total_time 731.0\n",
      "episode 4459, initial_state (3, 2, 3), reward 933.0, memory_length 2000, epsilon 0.13445182046914067 total_time 732.0\n",
      "episode 4469, initial_state (3, 11, 4), reward 924.0, memory_length 2000, epsilon 0.13384814701100894 total_time 722.0\n",
      "episode 4479, initial_state (0, 22, 1), reward 1208.0, memory_length 2000, epsilon 0.13324718398040308 total_time 724.0\n",
      "episode 4489, initial_state (1, 17, 2), reward 1370.0, memory_length 2000, epsilon 0.13264891920780106 total_time 722.0\n",
      "episode 4499, initial_state (4, 10, 6), reward 1140.0, memory_length 2000, epsilon 0.13205334057832088 total_time 725.0\n",
      "episode 4509, initial_state (3, 21, 0), reward 1004.0, memory_length 2000, epsilon 0.13146043603147495 total_time 727.0\n",
      "episode 4519, initial_state (1, 15, 3), reward 1265.0, memory_length 2000, epsilon 0.13087019356092583 total_time 721.0\n",
      "episode 4529, initial_state (0, 3, 4), reward 1287.0, memory_length 2000, epsilon 0.13028260121424345 total_time 724.0\n",
      "episode 4539, initial_state (2, 14, 0), reward 1035.0, memory_length 2000, epsilon 0.1296976470926627 total_time 722.0\n",
      "episode 4549, initial_state (3, 17, 3), reward 1065.0, memory_length 2000, epsilon 0.12911531935084253 total_time 731.0\n",
      "episode 4559, initial_state (4, 8, 4), reward 1262.0, memory_length 2000, epsilon 0.12853560619662638 total_time 723.0\n",
      "episode 4569, initial_state (4, 3, 2), reward 984.0, memory_length 2000, epsilon 0.12795849589080294 total_time 722.0\n",
      "episode 4579, initial_state (0, 22, 6), reward 1217.0, memory_length 2000, epsilon 0.12738397674686888 total_time 727.0\n",
      "episode 4589, initial_state (1, 19, 4), reward 1134.0, memory_length 2000, epsilon 0.12681203713079195 total_time 731.0\n",
      "episode 4599, initial_state (2, 16, 2), reward 1060.0, memory_length 2000, epsilon 0.12624266546077528 total_time 722.0\n",
      "episode 4609, initial_state (3, 15, 1), reward 779.0, memory_length 2000, epsilon 0.1256758502070232 total_time 726.0\n",
      "episode 4619, initial_state (3, 23, 1), reward 1257.0, memory_length 2000, epsilon 0.1251115798915073 total_time 726.0\n",
      "episode 4629, initial_state (2, 22, 4), reward 1016.0, memory_length 2000, epsilon 0.12454984308773452 total_time 721.0\n",
      "episode 4639, initial_state (1, 17, 4), reward 1131.0, memory_length 2000, epsilon 0.12399062842051543 total_time 722.0\n",
      "episode 4649, initial_state (2, 13, 2), reward 1056.0, memory_length 2000, epsilon 0.12343392456573378 total_time 724.0\n",
      "episode 4659, initial_state (2, 2, 0), reward 1440.0, memory_length 2000, epsilon 0.12287972025011756 total_time 725.0\n",
      "episode 4669, initial_state (3, 23, 2), reward 1349.0, memory_length 2000, epsilon 0.1223280042510105 total_time 721.0\n",
      "episode 4679, initial_state (3, 16, 4), reward 928.0, memory_length 2000, epsilon 0.12177876539614463 total_time 729.0\n",
      "episode 4689, initial_state (1, 20, 2), reward 1256.0, memory_length 2000, epsilon 0.12123199256341449 total_time 721.0\n",
      "episode 4699, initial_state (2, 8, 6), reward 1130.0, memory_length 2000, epsilon 0.12068767468065145 total_time 731.0\n",
      "episode 4709, initial_state (4, 14, 3), reward 1364.0, memory_length 2000, epsilon 0.1201458007253998 total_time 721.0\n",
      "episode 4719, initial_state (0, 2, 2), reward 1258.0, memory_length 2000, epsilon 0.11960635972469351 total_time 721.0\n",
      "episode 4729, initial_state (4, 18, 4), reward 1305.0, memory_length 2000, epsilon 0.11906934075483375 total_time 728.0\n",
      "episode 4739, initial_state (1, 11, 6), reward 1185.0, memory_length 2000, epsilon 0.11853473294116816 total_time 726.0\n",
      "episode 4749, initial_state (3, 15, 3), reward 1337.0, memory_length 2000, epsilon 0.11800252545787014 total_time 725.0\n",
      "episode 4759, initial_state (0, 0, 1), reward 1091.0, memory_length 2000, epsilon 0.11747270752771999 total_time 725.0\n",
      "episode 4769, initial_state (3, 13, 0), reward 1274.0, memory_length 2000, epsilon 0.1169452684218866 total_time 726.0\n",
      "episode 4779, initial_state (3, 13, 4), reward 1418.0, memory_length 2000, epsilon 0.11642019745970993 total_time 723.0\n",
      "episode 4789, initial_state (2, 5, 2), reward 1043.0, memory_length 2000, epsilon 0.11589748400848512 total_time 721.0\n",
      "episode 4799, initial_state (3, 23, 0), reward 1137.0, memory_length 2000, epsilon 0.11537711748324699 total_time 722.0\n",
      "episode 4809, initial_state (0, 0, 0), reward 1332.0, memory_length 2000, epsilon 0.11485908734655546 total_time 726.0\n",
      "episode 4819, initial_state (0, 3, 3), reward 1109.0, memory_length 2000, epsilon 0.11434338310828272 total_time 724.0\n",
      "episode 4829, initial_state (2, 0, 6), reward 1212.0, memory_length 2000, epsilon 0.1138299943254002 total_time 723.0\n",
      "episode 4839, initial_state (3, 17, 5), reward 1465.0, memory_length 2000, epsilon 0.11331891060176755 total_time 724.0\n",
      "episode 4849, initial_state (3, 4, 4), reward 1239.0, memory_length 2000, epsilon 0.11281012158792193 total_time 729.0\n",
      "episode 4859, initial_state (4, 7, 2), reward 1357.0, memory_length 2000, epsilon 0.11230361698086838 total_time 722.0\n",
      "episode 4869, initial_state (0, 9, 6), reward 1125.0, memory_length 2000, epsilon 0.11179938652387135 total_time 724.0\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "episode 4879, initial_state (1, 17, 0), reward 903.0, memory_length 2000, epsilon 0.11129742000624678 total_time 727.0\n",
      "episode 4889, initial_state (0, 4, 5), reward 1394.0, memory_length 2000, epsilon 0.11079770726315558 total_time 723.0\n",
      "episode 4899, initial_state (3, 21, 1), reward 889.0, memory_length 2000, epsilon 0.11030023817539766 total_time 721.0\n",
      "episode 4909, initial_state (3, 0, 6), reward 1192.0, memory_length 2000, epsilon 0.10980500266920691 total_time 721.0\n",
      "episode 4919, initial_state (2, 21, 0), reward 1203.0, memory_length 2000, epsilon 0.10931199071604747 total_time 726.0\n",
      "episode 4929, initial_state (1, 21, 5), reward 1143.0, memory_length 2000, epsilon 0.10882119233241047 total_time 726.0\n",
      "episode 4939, initial_state (4, 22, 6), reward 1107.0, memory_length 2000, epsilon 0.10833259757961175 total_time 731.0\n",
      "episode 4949, initial_state (3, 7, 4), reward 1128.0, memory_length 2000, epsilon 0.10784619656359101 total_time 721.0\n",
      "episode 4959, initial_state (1, 4, 0), reward 1063.0, memory_length 2000, epsilon 0.10736197943471096 total_time 727.0\n",
      "episode 4969, initial_state (1, 1, 2), reward 937.0, memory_length 2000, epsilon 0.10687993638755819 total_time 726.0\n",
      "episode 4979, initial_state (2, 3, 1), reward 1358.0, memory_length 2000, epsilon 0.10640005766074462 total_time 721.0\n",
      "episode 4989, initial_state (0, 0, 5), reward 1103.0, memory_length 2000, epsilon 0.10592233353670952 total_time 721.0\n",
      "episode 4999, initial_state (2, 9, 2), reward 1201.0, memory_length 2000, epsilon 0.10544675434152313 total_time 726.0\n",
      "Saving Model 5000\n",
      "Saving Model 5000 Weights\n",
      "episode 5009, initial_state (2, 9, 4), reward 1256.0, memory_length 2000, epsilon 0.10497331044469045 total_time 726.0\n",
      "episode 5019, initial_state (0, 23, 1), reward 1316.0, memory_length 2000, epsilon 0.1045019922589564 total_time 722.0\n",
      "episode 5029, initial_state (2, 4, 3), reward 1251.0, memory_length 2000, epsilon 0.10403279024011165 total_time 721.0\n",
      "episode 5039, initial_state (0, 1, 4), reward 1180.0, memory_length 2000, epsilon 0.10356569488679924 total_time 724.0\n",
      "episode 5049, initial_state (0, 1, 3), reward 1054.0, memory_length 2000, epsilon 0.10310069674032231 total_time 722.0\n",
      "episode 5059, initial_state (0, 23, 5), reward 1066.0, memory_length 2000, epsilon 0.10263778638445256 total_time 722.0\n",
      "episode 5069, initial_state (0, 20, 5), reward 774.0, memory_length 2000, epsilon 0.10217695444523936 total_time 725.0\n",
      "episode 5079, initial_state (2, 22, 3), reward 1081.0, memory_length 2000, epsilon 0.10171819159082031 total_time 723.0\n",
      "episode 5089, initial_state (3, 15, 0), reward 1117.0, memory_length 2000, epsilon 0.10126148853123183 total_time 724.0\n",
      "episode 5099, initial_state (0, 3, 0), reward 1259.0, memory_length 2000, epsilon 0.10080683601822141 total_time 724.0\n",
      "episode 5109, initial_state (2, 17, 5), reward 1584.0, memory_length 2000, epsilon 0.10035422484506014 total_time 729.0\n",
      "episode 5119, initial_state (4, 1, 0), reward 1134.0, memory_length 2000, epsilon 0.09990364584635623 total_time 727.0\n",
      "episode 5129, initial_state (3, 8, 2), reward 999.0, memory_length 2000, epsilon 0.09945508989786962 total_time 723.0\n",
      "episode 5139, initial_state (4, 5, 5), reward 1098.0, memory_length 2000, epsilon 0.09900854791632704 total_time 725.0\n",
      "episode 5149, initial_state (1, 10, 1), reward 1522.0, memory_length 2000, epsilon 0.09856401085923804 total_time 724.0\n",
      "episode 5159, initial_state (4, 23, 1), reward 1343.0, memory_length 2000, epsilon 0.09812146972471208 total_time 722.0\n",
      "episode 5169, initial_state (3, 5, 2), reward 1194.0, memory_length 2000, epsilon 0.09768091555127602 total_time 722.0\n",
      "episode 5179, initial_state (1, 16, 4), reward 1342.0, memory_length 2000, epsilon 0.0972423394176928 total_time 726.0\n",
      "episode 5189, initial_state (4, 13, 0), reward 999.0, memory_length 2000, epsilon 0.09680573244278078 total_time 726.0\n",
      "episode 5199, initial_state (3, 23, 0), reward 1144.0, memory_length 2000, epsilon 0.09637108578523372 total_time 726.0\n",
      "episode 5209, initial_state (4, 4, 6), reward 955.0, memory_length 2000, epsilon 0.09593839064344202 total_time 724.0\n",
      "episode 5219, initial_state (1, 2, 0), reward 1187.0, memory_length 2000, epsilon 0.09550763825531422 total_time 728.0\n",
      "episode 5229, initial_state (3, 2, 6), reward 1194.0, memory_length 2000, epsilon 0.09507881989809976 total_time 724.0\n",
      "episode 5239, initial_state (3, 23, 1), reward 1215.0, memory_length 2000, epsilon 0.09465192688821228 total_time 728.0\n",
      "episode 5249, initial_state (2, 12, 5), reward 1009.0, memory_length 2000, epsilon 0.09422695058105371 total_time 724.0\n",
      "episode 5259, initial_state (4, 21, 3), reward 1383.0, memory_length 2000, epsilon 0.0938038823708393 total_time 723.0\n",
      "episode 5269, initial_state (0, 20, 6), reward 1076.0, memory_length 2000, epsilon 0.09338271369042338 total_time 721.0\n",
      "episode 5279, initial_state (1, 3, 2), reward 1244.0, memory_length 2000, epsilon 0.09296343601112574 total_time 722.0\n",
      "episode 5289, initial_state (0, 23, 1), reward 1143.0, memory_length 2000, epsilon 0.09254604084255907 total_time 728.0\n",
      "episode 5299, initial_state (1, 18, 2), reward 1373.0, memory_length 2000, epsilon 0.09213051973245691 total_time 723.0\n",
      "episode 5309, initial_state (3, 16, 6), reward 1005.0, memory_length 2000, epsilon 0.09171686426650259 total_time 721.0\n",
      "episode 5319, initial_state (3, 6, 3), reward 1065.0, memory_length 2000, epsilon 0.09130506606815886 total_time 725.0\n",
      "episode 5329, initial_state (2, 12, 0), reward 1195.0, memory_length 2000, epsilon 0.09089511679849803 total_time 728.0\n",
      "episode 5339, initial_state (4, 10, 4), reward 1332.0, memory_length 2000, epsilon 0.09048700815603346 total_time 724.0\n",
      "episode 5349, initial_state (0, 22, 1), reward 1383.0, memory_length 2000, epsilon 0.0900807318765511 total_time 729.0\n",
      "episode 5359, initial_state (1, 20, 1), reward 1177.0, memory_length 2000, epsilon 0.0896762797329425 total_time 721.0\n",
      "episode 5369, initial_state (4, 2, 1), reward 1344.0, memory_length 2000, epsilon 0.08927364353503792 total_time 721.0\n",
      "episode 5379, initial_state (0, 11, 0), reward 1488.0, memory_length 2000, epsilon 0.08887281512944055 total_time 726.0\n",
      "episode 5389, initial_state (0, 16, 0), reward 1563.0, memory_length 2000, epsilon 0.08847378639936147 total_time 723.0\n",
      "episode 5399, initial_state (4, 7, 3), reward 1023.0, memory_length 2000, epsilon 0.08807654926445534 total_time 724.0\n",
      "episode 5409, initial_state (4, 5, 5), reward 1035.0, memory_length 2000, epsilon 0.08768109568065653 total_time 731.0\n",
      "episode 5419, initial_state (2, 21, 6), reward 1383.0, memory_length 2000, epsilon 0.08728741764001652 total_time 725.0\n",
      "episode 5429, initial_state (2, 7, 0), reward 1373.0, memory_length 2000, epsilon 0.08689550717054143 total_time 732.0\n",
      "episode 5439, initial_state (1, 4, 6), reward 1436.0, memory_length 2000, epsilon 0.08650535633603094 total_time 725.0\n",
      "episode 5449, initial_state (1, 22, 4), reward 1176.0, memory_length 2000, epsilon 0.08611695723591734 total_time 725.0\n",
      "episode 5459, initial_state (0, 0, 0), reward 1005.0, memory_length 2000, epsilon 0.08573030200510551 total_time 728.0\n",
      "episode 5469, initial_state (1, 0, 2), reward 1396.0, memory_length 2000, epsilon 0.08534538281381387 total_time 727.0\n",
      "episode 5479, initial_state (3, 7, 2), reward 1260.0, memory_length 2000, epsilon 0.0849621918674156 total_time 727.0\n",
      "episode 5489, initial_state (3, 22, 4), reward 1548.0, memory_length 2000, epsilon 0.08458072140628095 total_time 722.0\n",
      "episode 5499, initial_state (1, 7, 1), reward 1472.0, memory_length 2000, epsilon 0.0842009637056201 total_time 724.0\n",
      "episode 5509, initial_state (0, 5, 2), reward 1198.0, memory_length 2000, epsilon 0.08382291107532656 total_time 725.0\n",
      "episode 5519, initial_state (1, 21, 3), reward 1423.0, memory_length 2000, epsilon 0.08344655585982166 total_time 730.0\n",
      "episode 5529, initial_state (3, 4, 4), reward 1262.0, memory_length 2000, epsilon 0.0830718904378995 total_time 730.0\n",
      "episode 5539, initial_state (4, 20, 3), reward 1072.0, memory_length 2000, epsilon 0.0826989072225724 total_time 726.0\n",
      "episode 5549, initial_state (0, 10, 6), reward 1053.0, memory_length 2000, epsilon 0.08232759866091754 total_time 727.0\n",
      "episode 5559, initial_state (1, 13, 6), reward 1171.0, memory_length 2000, epsilon 0.08195795723392385 total_time 725.0\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "episode 5569, initial_state (4, 7, 2), reward 1207.0, memory_length 2000, epsilon 0.08158997545633977 total_time 724.0\n",
      "episode 5579, initial_state (1, 5, 0), reward 1062.0, memory_length 2000, epsilon 0.08122364587652181 total_time 725.0\n",
      "episode 5589, initial_state (3, 12, 5), reward 1269.0, memory_length 2000, epsilon 0.08085896107628338 total_time 723.0\n",
      "episode 5599, initial_state (1, 3, 0), reward 882.0, memory_length 2000, epsilon 0.08049591367074488 total_time 724.0\n",
      "episode 5609, initial_state (4, 12, 5), reward 1214.0, memory_length 2000, epsilon 0.08013449630818388 total_time 728.0\n",
      "episode 5619, initial_state (4, 4, 4), reward 1013.0, memory_length 2000, epsilon 0.07977470166988646 total_time 731.0\n",
      "episode 5629, initial_state (1, 11, 5), reward 1436.0, memory_length 2000, epsilon 0.07941652246999892 total_time 722.0\n",
      "episode 5639, initial_state (0, 6, 5), reward 1283.0, memory_length 2000, epsilon 0.07905995145538018 total_time 724.0\n",
      "episode 5649, initial_state (2, 21, 6), reward 1153.0, memory_length 2000, epsilon 0.07870498140545505 total_time 722.0\n",
      "episode 5659, initial_state (0, 23, 3), reward 1508.0, memory_length 2000, epsilon 0.0783516051320679 total_time 734.0\n",
      "episode 5669, initial_state (0, 20, 2), reward 1206.0, memory_length 2000, epsilon 0.07799981547933707 total_time 729.0\n",
      "episode 5679, initial_state (3, 16, 0), reward 1198.0, memory_length 2000, epsilon 0.07764960532351012 total_time 729.0\n",
      "episode 5689, initial_state (0, 0, 3), reward 1382.0, memory_length 2000, epsilon 0.07730096757281937 total_time 725.0\n",
      "episode 5699, initial_state (3, 17, 6), reward 1153.0, memory_length 2000, epsilon 0.0769538951673385 total_time 725.0\n",
      "episode 5709, initial_state (1, 16, 3), reward 1382.0, memory_length 2000, epsilon 0.07660838107883944 total_time 723.0\n",
      "episode 5719, initial_state (0, 19, 6), reward 1310.0, memory_length 2000, epsilon 0.07626441831065005 total_time 721.0\n",
      "episode 5729, initial_state (3, 19, 3), reward 1148.0, memory_length 2000, epsilon 0.07592199989751258 total_time 726.0\n",
      "episode 5739, initial_state (4, 7, 1), reward 1518.0, memory_length 2000, epsilon 0.0755811189054424 total_time 721.0\n",
      "episode 5749, initial_state (2, 9, 1), reward 1046.0, memory_length 2000, epsilon 0.07524176843158781 total_time 723.0\n",
      "episode 5759, initial_state (1, 6, 2), reward 1404.0, memory_length 2000, epsilon 0.07490394160409013 total_time 725.0\n",
      "episode 5769, initial_state (3, 19, 4), reward 1351.0, memory_length 2000, epsilon 0.07456763158194452 total_time 722.0\n",
      "episode 5779, initial_state (1, 13, 1), reward 1090.0, memory_length 2000, epsilon 0.07423283155486153 total_time 728.0\n",
      "episode 5789, initial_state (1, 4, 0), reward 1426.0, memory_length 2000, epsilon 0.07389953474312926 total_time 723.0\n",
      "episode 5799, initial_state (2, 22, 4), reward 1280.0, memory_length 2000, epsilon 0.07356773439747578 total_time 724.0\n",
      "episode 5809, initial_state (0, 17, 3), reward 1320.0, memory_length 2000, epsilon 0.07323742379893283 total_time 725.0\n",
      "episode 5819, initial_state (2, 19, 2), reward 1401.0, memory_length 2000, epsilon 0.07290859625869943 total_time 725.0\n",
      "episode 5829, initial_state (0, 14, 4), reward 882.0, memory_length 2000, epsilon 0.07258124511800669 total_time 723.0\n",
      "episode 5839, initial_state (2, 8, 4), reward 1160.0, memory_length 2000, epsilon 0.07225536374798286 total_time 722.0\n",
      "episode 5849, initial_state (0, 23, 4), reward 1436.0, memory_length 2000, epsilon 0.071930945549519 total_time 727.0\n",
      "episode 5859, initial_state (4, 0, 3), reward 1028.0, memory_length 2000, epsilon 0.07160798395313554 total_time 726.0\n",
      "episode 5869, initial_state (1, 7, 6), reward 1463.0, memory_length 2000, epsilon 0.0712864724188491 total_time 731.0\n",
      "episode 5879, initial_state (0, 1, 1), reward 1862.0, memory_length 2000, epsilon 0.07096640443604013 total_time 724.0\n",
      "episode 5889, initial_state (1, 19, 6), reward 1109.0, memory_length 2000, epsilon 0.07064777352332108 total_time 722.0\n",
      "episode 5899, initial_state (2, 17, 3), reward 1625.0, memory_length 2000, epsilon 0.070330573228405 total_time 721.0\n",
      "episode 5909, initial_state (4, 4, 6), reward 1091.0, memory_length 2000, epsilon 0.07001479712797512 total_time 731.0\n",
      "episode 5919, initial_state (2, 17, 2), reward 1269.0, memory_length 2000, epsilon 0.06970043882755464 total_time 721.0\n",
      "episode 5929, initial_state (4, 1, 2), reward 1631.0, memory_length 2000, epsilon 0.0693874919613772 total_time 724.0\n",
      "episode 5939, initial_state (0, 1, 6), reward 1460.0, memory_length 2000, epsilon 0.06907595019225808 total_time 724.0\n",
      "episode 5949, initial_state (3, 19, 4), reward 1329.0, memory_length 2000, epsilon 0.0687658072114658 total_time 725.0\n",
      "episode 5959, initial_state (0, 11, 6), reward 1306.0, memory_length 2000, epsilon 0.06845705673859441 total_time 723.0\n",
      "episode 5969, initial_state (4, 19, 3), reward 1647.0, memory_length 2000, epsilon 0.0681496925214363 total_time 726.0\n",
      "episode 5979, initial_state (1, 6, 6), reward 1449.0, memory_length 2000, epsilon 0.06784370833585553 total_time 725.0\n",
      "episode 5989, initial_state (2, 2, 1), reward 1245.0, memory_length 2000, epsilon 0.06753909798566192 total_time 722.0\n",
      "episode 5999, initial_state (0, 3, 3), reward 1352.0, memory_length 2000, epsilon 0.06723585530248544 total_time 728.0\n",
      "Saving Model 6000\n",
      "Saving Model 6000 Weights\n",
      "episode 6009, initial_state (1, 21, 1), reward 1455.0, memory_length 2000, epsilon 0.06693397414565139 total_time 723.0\n",
      "episode 6019, initial_state (2, 14, 2), reward 951.0, memory_length 2000, epsilon 0.06663344840205608 total_time 728.0\n",
      "episode 6029, initial_state (4, 5, 5), reward 1076.0, memory_length 2000, epsilon 0.06633427198604289 total_time 735.0\n",
      "episode 6039, initial_state (0, 21, 1), reward 1396.0, memory_length 2000, epsilon 0.06603643883927916 total_time 724.0\n",
      "episode 6049, initial_state (0, 10, 2), reward 1377.0, memory_length 2000, epsilon 0.06573994293063355 total_time 721.0\n",
      "episode 6059, initial_state (0, 13, 1), reward 1192.0, memory_length 2000, epsilon 0.06544477825605372 total_time 723.0\n",
      "episode 6069, initial_state (0, 9, 5), reward 1026.0, memory_length 2000, epsilon 0.06515093883844497 total_time 723.0\n",
      "episode 6079, initial_state (4, 4, 5), reward 1287.0, memory_length 2000, epsilon 0.06485841872754898 total_time 723.0\n",
      "episode 6089, initial_state (1, 5, 0), reward 1328.0, memory_length 2000, epsilon 0.06456721199982357 total_time 723.0\n",
      "episode 6099, initial_state (0, 10, 0), reward 1467.0, memory_length 2000, epsilon 0.06427731275832257 total_time 730.0\n",
      "episode 6109, initial_state (1, 6, 5), reward 1152.0, memory_length 2000, epsilon 0.06398871513257637 total_time 722.0\n",
      "episode 6119, initial_state (3, 15, 6), reward 1634.0, memory_length 2000, epsilon 0.06370141327847323 total_time 721.0\n",
      "episode 6129, initial_state (4, 16, 5), reward 1215.0, memory_length 2000, epsilon 0.06341540137814075 total_time 721.0\n",
      "episode 6139, initial_state (3, 8, 3), reward 1559.0, memory_length 2000, epsilon 0.06313067363982822 total_time 727.0\n",
      "episode 6149, initial_state (1, 0, 2), reward 1438.0, memory_length 2000, epsilon 0.0628472242977892 total_time 721.0\n",
      "episode 6159, initial_state (4, 22, 6), reward 1495.0, memory_length 2000, epsilon 0.06256504761216479 total_time 721.0\n",
      "episode 6169, initial_state (1, 23, 0), reward 1494.0, memory_length 2000, epsilon 0.06228413786886749 total_time 727.0\n",
      "episode 6179, initial_state (4, 14, 4), reward 1490.0, memory_length 2000, epsilon 0.06200448937946541 total_time 723.0\n",
      "episode 6189, initial_state (4, 17, 6), reward 1550.0, memory_length 2000, epsilon 0.06172609648106707 total_time 722.0\n",
      "episode 6199, initial_state (3, 21, 0), reward 1684.0, memory_length 2000, epsilon 0.061448953536206774 total_time 731.0\n",
      "episode 6209, initial_state (0, 15, 4), reward 1190.0, memory_length 2000, epsilon 0.0611730549327304 total_time 722.0\n",
      "episode 6219, initial_state (1, 18, 1), reward 1409.0, memory_length 2000, epsilon 0.0608983950836818 total_time 731.0\n",
      "episode 6229, initial_state (0, 5, 2), reward 1209.0, memory_length 2000, epsilon 0.06062496842718969 total_time 722.0\n",
      "episode 6239, initial_state (3, 18, 6), reward 1170.0, memory_length 2000, epsilon 0.060352769426354855 total_time 722.0\n",
      "episode 6249, initial_state (3, 9, 0), reward 1338.0, memory_length 2000, epsilon 0.0600817925691383 total_time 726.0\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "episode 6259, initial_state (1, 14, 5), reward 1307.0, memory_length 2000, epsilon 0.059812032368249356 total_time 725.0\n",
      "episode 6269, initial_state (4, 13, 5), reward 1202.0, memory_length 2000, epsilon 0.05954348336103475 total_time 728.0\n",
      "episode 6279, initial_state (1, 18, 0), reward 1141.0, memory_length 2000, epsilon 0.05927614010936795 total_time 721.0\n",
      "episode 6289, initial_state (2, 18, 4), reward 1121.0, memory_length 2000, epsilon 0.05900999719953891 total_time 721.0\n",
      "episode 6299, initial_state (4, 21, 0), reward 1365.0, memory_length 2000, epsilon 0.05874504924214464 total_time 725.0\n",
      "episode 6309, initial_state (2, 19, 5), reward 1421.0, memory_length 2000, epsilon 0.05848129087197997 total_time 722.0\n",
      "episode 6319, initial_state (4, 11, 3), reward 1592.0, memory_length 2000, epsilon 0.05821871674792885 total_time 722.0\n",
      "episode 6329, initial_state (4, 15, 1), reward 1343.0, memory_length 2000, epsilon 0.05795732155285634 total_time 726.0\n",
      "episode 6339, initial_state (0, 15, 1), reward 1364.0, memory_length 2000, epsilon 0.05769709999350076 total_time 726.0\n",
      "episode 6349, initial_state (0, 22, 0), reward 1563.0, memory_length 2000, epsilon 0.05743804680036667 total_time 726.0\n",
      "episode 6359, initial_state (3, 21, 5), reward 1130.0, memory_length 2000, epsilon 0.057180156727618066 total_time 723.0\n",
      "episode 6369, initial_state (4, 4, 4), reward 1322.0, memory_length 2000, epsilon 0.05692342455297213 total_time 727.0\n",
      "episode 6379, initial_state (4, 16, 6), reward 1064.0, memory_length 2000, epsilon 0.05666784507759359 total_time 726.0\n",
      "episode 6389, initial_state (1, 12, 3), reward 1451.0, memory_length 2000, epsilon 0.05641341312598929 total_time 727.0\n",
      "episode 6399, initial_state (3, 18, 4), reward 1140.0, memory_length 2000, epsilon 0.05616012354590355 total_time 725.0\n",
      "episode 6409, initial_state (0, 15, 3), reward 1365.0, memory_length 2000, epsilon 0.055907971208213725 total_time 727.0\n",
      "episode 6419, initial_state (0, 12, 0), reward 1113.0, memory_length 2000, epsilon 0.05565695100682632 total_time 726.0\n",
      "episode 6429, initial_state (3, 14, 3), reward 1377.0, memory_length 2000, epsilon 0.05540705785857371 total_time 727.0\n",
      "episode 6439, initial_state (2, 18, 1), reward 1311.0, memory_length 2000, epsilon 0.05515828670311112 total_time 723.0\n",
      "episode 6449, initial_state (2, 10, 2), reward 1252.0, memory_length 2000, epsilon 0.0549106325028141 total_time 726.0\n",
      "episode 6459, initial_state (2, 17, 6), reward 1377.0, memory_length 2000, epsilon 0.05466409024267668 total_time 734.0\n",
      "episode 6469, initial_state (2, 16, 6), reward 1101.0, memory_length 2000, epsilon 0.05441865493020963 total_time 723.0\n",
      "episode 6479, initial_state (4, 9, 2), reward 1433.0, memory_length 2000, epsilon 0.0541743215953395 total_time 723.0\n",
      "episode 6489, initial_state (2, 5, 2), reward 1371.0, memory_length 2000, epsilon 0.053931085290307934 total_time 722.0\n",
      "episode 6499, initial_state (3, 12, 2), reward 1493.0, memory_length 2000, epsilon 0.053688941089571396 total_time 721.0\n",
      "episode 6509, initial_state (0, 17, 4), reward 1346.0, memory_length 2000, epsilon 0.053447884089701594 total_time 730.0\n",
      "episode 6519, initial_state (2, 13, 0), reward 1306.0, memory_length 2000, epsilon 0.053207909409286 total_time 721.0\n",
      "episode 6529, initial_state (2, 1, 4), reward 1326.0, memory_length 2000, epsilon 0.052969012188829157 total_time 728.0\n",
      "episode 6539, initial_state (3, 23, 2), reward 1261.0, memory_length 2000, epsilon 0.052731187590654205 total_time 722.0\n",
      "episode 6549, initial_state (2, 13, 0), reward 1302.0, memory_length 2000, epsilon 0.052494430798804864 total_time 725.0\n",
      "episode 6559, initial_state (0, 8, 1), reward 1573.0, memory_length 2000, epsilon 0.05225873701894803 total_time 724.0\n",
      "episode 6569, initial_state (3, 8, 3), reward 1014.0, memory_length 2000, epsilon 0.05202410147827664 total_time 728.0\n",
      "episode 6579, initial_state (4, 11, 3), reward 1376.0, memory_length 2000, epsilon 0.0517905194254129 total_time 723.0\n",
      "episode 6589, initial_state (1, 11, 5), reward 1139.0, memory_length 2000, epsilon 0.05155798613031234 total_time 729.0\n",
      "episode 6599, initial_state (3, 10, 5), reward 1365.0, memory_length 2000, epsilon 0.05132649688416772 total_time 721.0\n",
      "episode 6609, initial_state (1, 19, 5), reward 1073.0, memory_length 2000, epsilon 0.05109604699931393 total_time 721.0\n",
      "episode 6619, initial_state (0, 12, 5), reward 1315.0, memory_length 2000, epsilon 0.05086663180913294 total_time 724.0\n",
      "episode 6629, initial_state (3, 21, 0), reward 1473.0, memory_length 2000, epsilon 0.05063824666795927 total_time 725.0\n",
      "episode 6639, initial_state (3, 6, 1), reward 1251.0, memory_length 2000, epsilon 0.05041088695098604 total_time 727.0\n",
      "episode 6649, initial_state (0, 0, 5), reward 1577.0, memory_length 2000, epsilon 0.0501845480541712 total_time 721.0\n",
      "episode 6659, initial_state (0, 2, 2), reward 1268.0, memory_length 2000, epsilon 0.04995922539414434 total_time 721.0\n",
      "episode 6669, initial_state (4, 13, 4), reward 1258.0, memory_length 2000, epsilon 0.04973491440811393 total_time 724.0\n",
      "episode 6679, initial_state (0, 0, 2), reward 1334.0, memory_length 2000, epsilon 0.049511610553774794 total_time 724.0\n",
      "episode 6689, initial_state (2, 20, 6), reward 1272.0, memory_length 2000, epsilon 0.04928930930921628 total_time 723.0\n",
      "episode 6699, initial_state (0, 15, 0), reward 1166.0, memory_length 2000, epsilon 0.04906800617283061 total_time 721.0\n",
      "episode 6709, initial_state (1, 8, 1), reward 1297.0, memory_length 2000, epsilon 0.048847696663221654 total_time 723.0\n",
      "episode 6719, initial_state (3, 0, 1), reward 1375.0, memory_length 2000, epsilon 0.048628376319114375 total_time 723.0\n",
      "episode 6729, initial_state (3, 22, 2), reward 1526.0, memory_length 2000, epsilon 0.048410040699264245 total_time 724.0\n",
      "episode 6739, initial_state (3, 13, 1), reward 1167.0, memory_length 2000, epsilon 0.04819268538236753 total_time 727.0\n",
      "episode 6749, initial_state (3, 7, 3), reward 1392.0, memory_length 2000, epsilon 0.04797630596697168 total_time 721.0\n",
      "episode 6759, initial_state (3, 21, 3), reward 1092.0, memory_length 2000, epsilon 0.04776089807138607 total_time 724.0\n",
      "episode 6769, initial_state (1, 4, 2), reward 1140.0, memory_length 2000, epsilon 0.0475464573335935 total_time 724.0\n",
      "episode 6779, initial_state (2, 18, 0), reward 1472.0, memory_length 2000, epsilon 0.04733297941116166 total_time 725.0\n",
      "episode 6789, initial_state (0, 18, 2), reward 1369.0, memory_length 2000, epsilon 0.047120459981155355 total_time 726.0\n",
      "episode 6799, initial_state (0, 12, 3), reward 1282.0, memory_length 2000, epsilon 0.04690889474004888 total_time 723.0\n",
      "episode 6809, initial_state (4, 3, 0), reward 1035.0, memory_length 2000, epsilon 0.04669827940363884 total_time 723.0\n",
      "episode 6819, initial_state (0, 23, 6), reward 1098.0, memory_length 2000, epsilon 0.04648860970695748 total_time 722.0\n",
      "episode 6829, initial_state (0, 15, 2), reward 1220.0, memory_length 2000, epsilon 0.04627988140418632 total_time 733.0\n",
      "episode 6839, initial_state (0, 2, 2), reward 1409.0, memory_length 2000, epsilon 0.04607209026857003 total_time 726.0\n",
      "episode 6849, initial_state (3, 9, 2), reward 1693.0, memory_length 2000, epsilon 0.045865232092331065 total_time 724.0\n",
      "episode 6859, initial_state (2, 1, 2), reward 1098.0, memory_length 2000, epsilon 0.045659302686584254 total_time 726.0\n",
      "episode 6869, initial_state (3, 9, 4), reward 1170.0, memory_length 2000, epsilon 0.045454297881252104 total_time 728.0\n",
      "episode 6879, initial_state (3, 21, 3), reward 1249.0, memory_length 2000, epsilon 0.045250213524980334 total_time 723.0\n",
      "episode 6889, initial_state (4, 23, 0), reward 1604.0, memory_length 2000, epsilon 0.0450470454850537 total_time 729.0\n",
      "episode 6899, initial_state (2, 8, 1), reward 1283.0, memory_length 2000, epsilon 0.0448447896473125 total_time 725.0\n",
      "episode 6909, initial_state (4, 0, 3), reward 1067.0, memory_length 2000, epsilon 0.04464344191606906 total_time 729.0\n",
      "episode 6919, initial_state (4, 9, 6), reward 970.0, memory_length 2000, epsilon 0.04444299821402498 total_time 730.0\n",
      "episode 6929, initial_state (1, 11, 6), reward 937.0, memory_length 2000, epsilon 0.04424345448218844 total_time 723.0\n",
      "episode 6939, initial_state (4, 8, 1), reward 1231.0, memory_length 2000, epsilon 0.044044806679792034 total_time 724.0\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "episode 6949, initial_state (2, 13, 6), reward 1549.0, memory_length 2000, epsilon 0.04384705078421098 total_time 723.0\n",
      "episode 6959, initial_state (2, 10, 0), reward 997.0, memory_length 2000, epsilon 0.04365018279088166 total_time 729.0\n",
      "episode 6969, initial_state (4, 17, 5), reward 1502.0, memory_length 2000, epsilon 0.04345419871322045 total_time 728.0\n",
      "episode 6979, initial_state (4, 5, 5), reward 1361.0, memory_length 2000, epsilon 0.043259094582543095 total_time 721.0\n",
      "episode 6989, initial_state (2, 18, 1), reward 1292.0, memory_length 2000, epsilon 0.043064866447984275 total_time 722.0\n",
      "episode 6999, initial_state (1, 9, 0), reward 1284.0, memory_length 2000, epsilon 0.04287151037641762 total_time 725.0\n",
      "Saving Model 7000\n",
      "Saving Model 7000 Weights\n",
      "episode 7009, initial_state (1, 11, 3), reward 1576.0, memory_length 2000, epsilon 0.042679022452376106 total_time 729.0\n",
      "episode 7019, initial_state (3, 22, 5), reward 1265.0, memory_length 2000, epsilon 0.04248739877797266 total_time 725.0\n",
      "episode 7029, initial_state (4, 1, 5), reward 1167.0, memory_length 2000, epsilon 0.04229663547282135 total_time 725.0\n",
      "episode 7039, initial_state (4, 11, 6), reward 1062.0, memory_length 2000, epsilon 0.042106728673958704 total_time 731.0\n",
      "episode 7049, initial_state (2, 6, 3), reward 1177.0, memory_length 2000, epsilon 0.041917674535765566 total_time 724.0\n",
      "episode 7059, initial_state (3, 5, 3), reward 1463.0, memory_length 2000, epsilon 0.0417294692298892 total_time 728.0\n",
      "episode 7069, initial_state (2, 12, 6), reward 1704.0, memory_length 2000, epsilon 0.04154210894516569 total_time 723.0\n",
      "episode 7079, initial_state (0, 0, 2), reward 1121.0, memory_length 2000, epsilon 0.04135558988754289 total_time 727.0\n",
      "episode 7089, initial_state (1, 14, 2), reward 1058.0, memory_length 2000, epsilon 0.04116990828000354 total_time 729.0\n",
      "episode 7099, initial_state (1, 3, 1), reward 1191.0, memory_length 2000, epsilon 0.040985060362488675 total_time 722.0\n",
      "episode 7109, initial_state (4, 13, 1), reward 1742.0, memory_length 2000, epsilon 0.04080104239182171 total_time 725.0\n",
      "episode 7119, initial_state (0, 10, 0), reward 1306.0, memory_length 2000, epsilon 0.040617850641632404 total_time 721.0\n",
      "episode 7129, initial_state (3, 18, 0), reward 1385.0, memory_length 2000, epsilon 0.04043548140228158 total_time 722.0\n",
      "episode 7139, initial_state (4, 3, 5), reward 1194.0, memory_length 2000, epsilon 0.04025393098078592 total_time 723.0\n",
      "episode 7149, initial_state (4, 9, 3), reward 1315.0, memory_length 2000, epsilon 0.04007319570074316 total_time 729.0\n",
      "episode 7159, initial_state (3, 12, 1), reward 1463.0, memory_length 2000, epsilon 0.03989327190225773 total_time 737.0\n",
      "episode 7169, initial_state (0, 14, 0), reward 1062.0, memory_length 2000, epsilon 0.039714155941866516 total_time 724.0\n",
      "episode 7179, initial_state (1, 21, 4), reward 1011.0, memory_length 2000, epsilon 0.03953584419246524 total_time 722.0\n",
      "episode 7189, initial_state (3, 12, 5), reward 1101.0, memory_length 2000, epsilon 0.03935833304323489 total_time 724.0\n",
      "episode 7199, initial_state (2, 21, 6), reward 1220.0, memory_length 2000, epsilon 0.03918161889956859 total_time 725.0\n",
      "episode 7209, initial_state (3, 2, 2), reward 1218.0, memory_length 2000, epsilon 0.03900569818299891 total_time 722.0\n",
      "episode 7219, initial_state (0, 10, 2), reward 1775.0, memory_length 2000, epsilon 0.03883056733112536 total_time 721.0\n",
      "episode 7229, initial_state (1, 8, 5), reward 1315.0, memory_length 2000, epsilon 0.038656222797542154 total_time 728.0\n",
      "episode 7239, initial_state (4, 19, 6), reward 1197.0, memory_length 2000, epsilon 0.038482661051766565 total_time 722.0\n",
      "episode 7249, initial_state (1, 19, 4), reward 1508.0, memory_length 2000, epsilon 0.03830987857916728 total_time 727.0\n",
      "episode 7259, initial_state (0, 22, 4), reward 1326.0, memory_length 2000, epsilon 0.03813787188089335 total_time 725.0\n",
      "episode 7269, initial_state (0, 18, 5), reward 1411.0, memory_length 2000, epsilon 0.03796663747380326 total_time 725.0\n",
      "episode 7279, initial_state (3, 4, 2), reward 1026.0, memory_length 2000, epsilon 0.03779617189039438 total_time 724.0\n",
      "episode 7289, initial_state (3, 6, 4), reward 1459.0, memory_length 2000, epsilon 0.03762647167873284 total_time 725.0\n",
      "episode 7299, initial_state (3, 1, 4), reward 1440.0, memory_length 2000, epsilon 0.037457533402383586 total_time 721.0\n",
      "episode 7309, initial_state (3, 6, 6), reward 1789.0, memory_length 2000, epsilon 0.0372893536403407 total_time 728.0\n",
      "episode 7319, initial_state (3, 10, 1), reward 1384.0, memory_length 2000, epsilon 0.0371219289869583 total_time 723.0\n",
      "episode 7329, initial_state (4, 17, 2), reward 1639.0, memory_length 2000, epsilon 0.03695525605188139 total_time 724.0\n",
      "episode 7339, initial_state (0, 0, 1), reward 1225.0, memory_length 2000, epsilon 0.03678933145997735 total_time 731.0\n",
      "episode 7349, initial_state (1, 6, 6), reward 1672.0, memory_length 2000, epsilon 0.03662415185126756 total_time 721.0\n",
      "episode 7359, initial_state (3, 14, 6), reward 1372.0, memory_length 2000, epsilon 0.03645971388085925 total_time 723.0\n",
      "episode 7369, initial_state (1, 4, 2), reward 1412.0, memory_length 2000, epsilon 0.03629601421887794 total_time 722.0\n",
      "episode 7379, initial_state (0, 5, 6), reward 1418.0, memory_length 2000, epsilon 0.03613304955039984 total_time 729.0\n",
      "episode 7389, initial_state (3, 11, 5), reward 1240.0, memory_length 2000, epsilon 0.03597081657538487 total_time 724.0\n",
      "episode 7399, initial_state (4, 17, 3), reward 1101.0, memory_length 2000, epsilon 0.035809312008609755 total_time 725.0\n",
      "episode 7409, initial_state (0, 12, 0), reward 1521.0, memory_length 2000, epsilon 0.03564853257960147 total_time 721.0\n",
      "episode 7419, initial_state (1, 9, 3), reward 997.0, memory_length 2000, epsilon 0.0354884750325711 total_time 722.0\n",
      "episode 7429, initial_state (4, 3, 1), reward 1376.0, memory_length 2000, epsilon 0.03532913612634785 total_time 723.0\n",
      "episode 7439, initial_state (3, 15, 5), reward 1494.0, memory_length 2000, epsilon 0.03517051263431341 total_time 728.0\n",
      "episode 7449, initial_state (0, 14, 6), reward 1508.0, memory_length 2000, epsilon 0.03501260134433667 total_time 725.0\n",
      "episode 7459, initial_state (1, 7, 1), reward 1266.0, memory_length 2000, epsilon 0.034855399058708575 total_time 722.0\n",
      "episode 7469, initial_state (0, 9, 6), reward 1152.0, memory_length 2000, epsilon 0.0346989025940775 total_time 729.0\n",
      "episode 7479, initial_state (4, 6, 6), reward 1527.0, memory_length 2000, epsilon 0.03454310878138469 total_time 732.0\n",
      "episode 7489, initial_state (0, 15, 6), reward 1368.0, memory_length 2000, epsilon 0.03438801446580009 total_time 723.0\n",
      "episode 7499, initial_state (0, 6, 6), reward 1332.0, memory_length 2000, epsilon 0.03423361650665853 total_time 722.0\n",
      "episode 7509, initial_state (4, 13, 1), reward 1505.0, memory_length 2000, epsilon 0.03407991177739605 total_time 723.0\n",
      "episode 7519, initial_state (1, 4, 4), reward 1319.0, memory_length 2000, epsilon 0.03392689716548662 total_time 728.0\n",
      "episode 7529, initial_state (1, 4, 0), reward 1233.0, memory_length 2000, epsilon 0.03377456957237916 total_time 722.0\n",
      "episode 7539, initial_state (1, 0, 6), reward 1279.0, memory_length 2000, epsilon 0.033622925913434666 total_time 722.0\n",
      "episode 7549, initial_state (2, 1, 2), reward 1394.0, memory_length 2000, epsilon 0.033471963117863865 total_time 723.0\n",
      "episode 7559, initial_state (1, 5, 2), reward 1320.0, memory_length 2000, epsilon 0.03332167812866502 total_time 723.0\n",
      "episode 7569, initial_state (2, 3, 2), reward 1347.0, memory_length 2000, epsilon 0.03317206790256193 total_time 721.0\n",
      "episode 7579, initial_state (1, 9, 6), reward 1270.0, memory_length 2000, epsilon 0.03302312940994242 total_time 729.0\n",
      "episode 7589, initial_state (2, 3, 4), reward 1431.0, memory_length 2000, epsilon 0.032874859634796916 total_time 728.0\n",
      "episode 7599, initial_state (1, 11, 2), reward 1559.0, memory_length 2000, epsilon 0.0327272555746574 total_time 723.0\n",
      "episode 7609, initial_state (3, 8, 2), reward 1158.0, memory_length 2000, epsilon 0.032580314240536645 total_time 724.0\n",
      "episode 7619, initial_state (1, 18, 2), reward 1352.0, memory_length 2000, epsilon 0.03243403265686757 total_time 730.0\n",
      "episode 7629, initial_state (0, 9, 4), reward 1324.0, memory_length 2000, epsilon 0.03228840786144314 total_time 725.0\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "episode 7639, initial_state (4, 16, 5), reward 1017.0, memory_length 2000, epsilon 0.03214343690535624 total_time 728.0\n",
      "episode 7649, initial_state (4, 2, 6), reward 1402.0, memory_length 2000, epsilon 0.031999116852940074 total_time 721.0\n",
      "episode 7659, initial_state (2, 3, 3), reward 1531.0, memory_length 2000, epsilon 0.03185544478170866 total_time 729.0\n",
      "episode 7669, initial_state (4, 19, 3), reward 1498.0, memory_length 2000, epsilon 0.031712417782297615 total_time 722.0\n",
      "episode 7679, initial_state (1, 10, 3), reward 1092.0, memory_length 2000, epsilon 0.03157003295840534 total_time 723.0\n",
      "episode 7689, initial_state (3, 11, 5), reward 1621.0, memory_length 2000, epsilon 0.03142828742673429 total_time 729.0\n",
      "episode 7699, initial_state (4, 3, 0), reward 1244.0, memory_length 2000, epsilon 0.03128717831693258 total_time 722.0\n",
      "episode 7709, initial_state (2, 1, 4), reward 1494.0, memory_length 2000, epsilon 0.03114670277153593 total_time 729.0\n",
      "episode 7719, initial_state (0, 4, 2), reward 1495.0, memory_length 2000, epsilon 0.031006857945909738 total_time 725.0\n",
      "episode 7729, initial_state (4, 16, 5), reward 1653.0, memory_length 2000, epsilon 0.030867641008191508 total_time 726.0\n",
      "episode 7739, initial_state (4, 23, 4), reward 1467.0, memory_length 2000, epsilon 0.030729049139233508 total_time 721.0\n",
      "episode 7749, initial_state (1, 0, 2), reward 1208.0, memory_length 2000, epsilon 0.030591079532545633 total_time 724.0\n",
      "episode 7759, initial_state (1, 15, 6), reward 1056.0, memory_length 2000, epsilon 0.030453729394238658 total_time 723.0\n",
      "episode 7769, initial_state (0, 0, 6), reward 1130.0, memory_length 2000, epsilon 0.030316995942967558 total_time 724.0\n",
      "episode 7779, initial_state (1, 13, 0), reward 1366.0, memory_length 2000, epsilon 0.030180876409875292 total_time 724.0\n",
      "episode 7789, initial_state (3, 5, 3), reward 1667.0, memory_length 2000, epsilon 0.030045368038536673 total_time 725.0\n",
      "episode 7799, initial_state (3, 15, 1), reward 1309.0, memory_length 2000, epsilon 0.029910468084902525 total_time 722.0\n",
      "episode 7809, initial_state (0, 0, 1), reward 1410.0, memory_length 2000, epsilon 0.02977617381724419 total_time 723.0\n",
      "episode 7819, initial_state (4, 22, 1), reward 1742.0, memory_length 2000, epsilon 0.029642482516098172 total_time 727.0\n",
      "episode 7829, initial_state (1, 0, 0), reward 1372.0, memory_length 2000, epsilon 0.029509391474211027 total_time 725.0\n",
      "episode 7839, initial_state (4, 17, 3), reward 1484.0, memory_length 2000, epsilon 0.02937689799648464 total_time 723.0\n",
      "episode 7849, initial_state (2, 5, 6), reward 1150.0, memory_length 2000, epsilon 0.029244999399921527 total_time 721.0\n",
      "episode 7859, initial_state (0, 7, 6), reward 1313.0, memory_length 2000, epsilon 0.029113693013570616 total_time 728.0\n",
      "episode 7869, initial_state (3, 2, 1), reward 1242.0, memory_length 2000, epsilon 0.028982976178473112 total_time 726.0\n",
      "episode 7879, initial_state (2, 3, 4), reward 1230.0, memory_length 2000, epsilon 0.028852846247608605 total_time 726.0\n",
      "episode 7889, initial_state (3, 13, 1), reward 1006.0, memory_length 2000, epsilon 0.028723300585841584 total_time 726.0\n",
      "episode 7899, initial_state (2, 22, 3), reward 1311.0, memory_length 2000, epsilon 0.028594336569867936 total_time 723.0\n",
      "episode 7909, initial_state (2, 22, 5), reward 1216.0, memory_length 2000, epsilon 0.028465951588161946 total_time 724.0\n",
      "episode 7919, initial_state (1, 16, 5), reward 1242.0, memory_length 2000, epsilon 0.02833814304092336 total_time 726.0\n",
      "episode 7929, initial_state (3, 4, 5), reward 1123.0, memory_length 2000, epsilon 0.02821090834002471 total_time 722.0\n",
      "episode 7939, initial_state (2, 6, 6), reward 1285.0, memory_length 2000, epsilon 0.028084244908958958 total_time 730.0\n",
      "episode 7949, initial_state (1, 11, 4), reward 1166.0, memory_length 2000, epsilon 0.027958150182787318 total_time 725.0\n",
      "episode 7959, initial_state (4, 15, 0), reward 1427.0, memory_length 2000, epsilon 0.027832621608087242 total_time 725.0\n",
      "episode 7969, initial_state (2, 13, 3), reward 1437.0, memory_length 2000, epsilon 0.027707656642900834 total_time 721.0\n",
      "episode 7979, initial_state (0, 20, 5), reward 1576.0, memory_length 2000, epsilon 0.02758325275668325 total_time 724.0\n",
      "episode 7989, initial_state (3, 5, 2), reward 1448.0, memory_length 2000, epsilon 0.02745940743025156 total_time 722.0\n",
      "episode 7999, initial_state (4, 9, 4), reward 1571.0, memory_length 2000, epsilon 0.027336118155733682 total_time 726.0\n",
      "Saving Model 8000\n",
      "Saving Model 8000 Weights\n",
      "episode 8009, initial_state (1, 8, 0), reward 1094.0, memory_length 2000, epsilon 0.02721338243651757 total_time 725.0\n",
      "episode 8019, initial_state (4, 14, 0), reward 1579.0, memory_length 2000, epsilon 0.027091197787200736 total_time 722.0\n",
      "episode 8029, initial_state (0, 2, 0), reward 1625.0, memory_length 2000, epsilon 0.02696956173353984 total_time 724.0\n",
      "episode 8039, initial_state (4, 13, 3), reward 1659.0, memory_length 2000, epsilon 0.02684847181240064 total_time 725.0\n",
      "episode 8049, initial_state (0, 17, 4), reward 1328.0, memory_length 2000, epsilon 0.026727925571708118 total_time 727.0\n",
      "episode 8059, initial_state (2, 20, 2), reward 1265.0, memory_length 2000, epsilon 0.026607920570396747 total_time 722.0\n",
      "episode 8069, initial_state (0, 11, 2), reward 1314.0, memory_length 2000, epsilon 0.026488454378361167 total_time 724.0\n",
      "episode 8079, initial_state (1, 22, 0), reward 1238.0, memory_length 2000, epsilon 0.02636952457640692 total_time 727.0\n",
      "episode 8089, initial_state (4, 10, 2), reward 1517.0, memory_length 2000, epsilon 0.026251128756201423 total_time 723.0\n",
      "episode 8099, initial_state (4, 1, 0), reward 1571.0, memory_length 2000, epsilon 0.026133264520225304 total_time 723.0\n",
      "episode 8109, initial_state (1, 21, 4), reward 1393.0, memory_length 2000, epsilon 0.026015929481723724 total_time 725.0\n",
      "episode 8119, initial_state (1, 13, 5), reward 1404.0, memory_length 2000, epsilon 0.02589912126465816 total_time 731.0\n",
      "episode 8129, initial_state (3, 15, 0), reward 1582.0, memory_length 2000, epsilon 0.025782837503658236 total_time 724.0\n",
      "episode 8139, initial_state (3, 3, 3), reward 1071.0, memory_length 2000, epsilon 0.025667075843973795 total_time 722.0\n",
      "episode 8149, initial_state (2, 20, 6), reward 1344.0, memory_length 2000, epsilon 0.025551833941427293 total_time 724.0\n",
      "episode 8159, initial_state (4, 3, 4), reward 1405.0, memory_length 2000, epsilon 0.02543710946236625 total_time 727.0\n",
      "episode 8169, initial_state (3, 5, 1), reward 1512.0, memory_length 2000, epsilon 0.025322900083616044 total_time 726.0\n",
      "episode 8179, initial_state (1, 17, 1), reward 933.0, memory_length 2000, epsilon 0.025209203492432875 total_time 724.0\n",
      "episode 8189, initial_state (3, 4, 3), reward 1255.0, memory_length 2000, epsilon 0.025096017386456854 total_time 723.0\n",
      "episode 8199, initial_state (0, 4, 4), reward 1567.0, memory_length 2000, epsilon 0.02498333947366549 total_time 726.0\n",
      "episode 8209, initial_state (4, 8, 1), reward 1180.0, memory_length 2000, epsilon 0.0248711674723272 total_time 726.0\n",
      "episode 8219, initial_state (3, 6, 0), reward 1488.0, memory_length 2000, epsilon 0.024759499110955107 total_time 722.0\n",
      "episode 8229, initial_state (4, 12, 0), reward 1410.0, memory_length 2000, epsilon 0.0246483321282611 total_time 726.0\n",
      "episode 8239, initial_state (3, 2, 0), reward 1585.0, memory_length 2000, epsilon 0.024537664273109955 total_time 724.0\n",
      "episode 8249, initial_state (3, 3, 0), reward 1468.0, memory_length 2000, epsilon 0.024427493304473833 total_time 732.0\n",
      "episode 8259, initial_state (2, 2, 6), reward 1404.0, memory_length 2000, epsilon 0.024317816991386867 total_time 727.0\n",
      "episode 8269, initial_state (4, 1, 1), reward 1447.0, memory_length 2000, epsilon 0.02420863311289995 total_time 725.0\n",
      "episode 8279, initial_state (1, 8, 4), reward 1416.0, memory_length 2000, epsilon 0.024099939458035836 total_time 723.0\n",
      "episode 8289, initial_state (3, 14, 2), reward 1072.0, memory_length 2000, epsilon 0.023991733825744266 total_time 721.0\n",
      "episode 8299, initial_state (4, 23, 6), reward 1328.0, memory_length 2000, epsilon 0.02388401402485751 total_time 724.0\n",
      "episode 8309, initial_state (2, 9, 4), reward 1148.0, memory_length 2000, epsilon 0.02377677787404593 total_time 725.0\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "episode 8319, initial_state (3, 1, 6), reward 1450.0, memory_length 2000, epsilon 0.023670023201773777 total_time 722.0\n",
      "episode 8329, initial_state (0, 9, 6), reward 1153.0, memory_length 2000, epsilon 0.023563747846255312 total_time 723.0\n",
      "episode 8339, initial_state (0, 12, 0), reward 1189.0, memory_length 2000, epsilon 0.023457949655410957 total_time 726.0\n",
      "episode 8349, initial_state (2, 19, 5), reward 1090.0, memory_length 2000, epsilon 0.02335262648682371 total_time 723.0\n",
      "episode 8359, initial_state (2, 20, 4), reward 1322.0, memory_length 2000, epsilon 0.023247776207695834 total_time 721.0\n",
      "episode 8369, initial_state (1, 15, 1), reward 1313.0, memory_length 2000, epsilon 0.023143396694805567 total_time 724.0\n",
      "episode 8379, initial_state (3, 1, 6), reward 997.0, memory_length 2000, epsilon 0.023039485834464225 total_time 721.0\n",
      "episode 8389, initial_state (3, 9, 0), reward 1553.0, memory_length 2000, epsilon 0.02293604152247334 total_time 721.0\n",
      "episode 8399, initial_state (0, 15, 3), reward 1144.0, memory_length 2000, epsilon 0.02283306166408204 total_time 726.0\n",
      "episode 8409, initial_state (4, 18, 2), reward 1289.0, memory_length 2000, epsilon 0.02273054417394469 total_time 725.0\n",
      "episode 8419, initial_state (1, 17, 3), reward 1446.0, memory_length 2000, epsilon 0.0226284869760786 total_time 727.0\n",
      "episode 8429, initial_state (3, 8, 2), reward 1226.0, memory_length 2000, epsilon 0.022526888003822024 total_time 723.0\n",
      "episode 8439, initial_state (0, 0, 5), reward 1382.0, memory_length 2000, epsilon 0.022425745199792325 total_time 734.0\n",
      "episode 8449, initial_state (2, 16, 3), reward 1445.0, memory_length 2000, epsilon 0.022325056515844238 total_time 727.0\n",
      "episode 8459, initial_state (0, 10, 3), reward 1185.0, memory_length 2000, epsilon 0.02222481991302848 total_time 728.0\n",
      "episode 8469, initial_state (2, 5, 2), reward 1662.0, memory_length 2000, epsilon 0.02212503336155043 total_time 724.0\n",
      "episode 8479, initial_state (3, 10, 2), reward 1206.0, memory_length 2000, epsilon 0.022025694840728995 total_time 723.0\n",
      "episode 8489, initial_state (1, 4, 1), reward 1398.0, memory_length 2000, epsilon 0.02192680233895575 total_time 723.0\n",
      "episode 8499, initial_state (0, 17, 5), reward 1134.0, memory_length 2000, epsilon 0.02182835385365414 total_time 729.0\n",
      "episode 8509, initial_state (2, 23, 2), reward 1593.0, memory_length 2000, epsilon 0.021730347391238976 total_time 721.0\n",
      "episode 8519, initial_state (0, 20, 6), reward 1406.0, memory_length 2000, epsilon 0.02163278096707606 total_time 730.0\n",
      "episode 8529, initial_state (2, 15, 1), reward 1498.0, memory_length 2000, epsilon 0.021535652605441942 total_time 722.0\n",
      "episode 8539, initial_state (3, 17, 2), reward 1358.0, memory_length 2000, epsilon 0.02143896033948401 total_time 722.0\n",
      "episode 8549, initial_state (0, 5, 1), reward 1261.0, memory_length 2000, epsilon 0.021342702211180543 total_time 722.0\n",
      "episode 8559, initial_state (2, 11, 3), reward 1108.0, memory_length 2000, epsilon 0.021246876271301177 total_time 731.0\n",
      "episode 8569, initial_state (3, 3, 6), reward 1538.0, memory_length 2000, epsilon 0.021151480579367353 total_time 725.0\n",
      "episode 8579, initial_state (2, 12, 4), reward 1501.0, memory_length 2000, epsilon 0.02105651320361304 total_time 721.0\n",
      "episode 8589, initial_state (2, 10, 1), reward 1077.0, memory_length 2000, epsilon 0.020961972220945632 total_time 727.0\n",
      "episode 8599, initial_state (4, 20, 2), reward 1212.0, memory_length 2000, epsilon 0.020867855716907018 total_time 723.0\n",
      "episode 8609, initial_state (0, 0, 4), reward 1585.0, memory_length 2000, epsilon 0.020774161785634748 total_time 724.0\n",
      "episode 8619, initial_state (4, 2, 3), reward 1279.0, memory_length 2000, epsilon 0.02068088852982354 total_time 722.0\n",
      "episode 8629, initial_state (4, 22, 3), reward 1359.0, memory_length 2000, epsilon 0.020588034060686752 total_time 734.0\n",
      "episode 8639, initial_state (3, 0, 0), reward 1148.0, memory_length 2000, epsilon 0.020495596497918222 total_time 722.0\n",
      "episode 8649, initial_state (1, 7, 0), reward 1278.0, memory_length 2000, epsilon 0.020403573969654153 total_time 727.0\n",
      "episode 8659, initial_state (0, 8, 0), reward 1473.0, memory_length 2000, epsilon 0.020311964612435188 total_time 722.0\n",
      "episode 8669, initial_state (0, 4, 4), reward 1181.0, memory_length 2000, epsilon 0.020220766571168728 total_time 723.0\n",
      "episode 8679, initial_state (3, 16, 2), reward 1154.0, memory_length 2000, epsilon 0.020129977999091303 total_time 725.0\n",
      "episode 8689, initial_state (0, 20, 3), reward 1521.0, memory_length 2000, epsilon 0.020039597057731236 total_time 722.0\n",
      "episode 8699, initial_state (0, 7, 0), reward 1302.0, memory_length 2000, epsilon 0.01994962191687138 total_time 729.0\n",
      "episode 8709, initial_state (4, 4, 6), reward 1204.0, memory_length 2000, epsilon 0.01986005075451205 total_time 722.0\n",
      "episode 8719, initial_state (3, 18, 5), reward 1144.0, memory_length 2000, epsilon 0.019770881756834155 total_time 723.0\n",
      "episode 8729, initial_state (2, 14, 1), reward 1271.0, memory_length 2000, epsilon 0.019682113118162443 total_time 728.0\n",
      "episode 8739, initial_state (0, 12, 3), reward 1562.0, memory_length 2000, epsilon 0.019593743040928938 total_time 727.0\n",
      "episode 8749, initial_state (3, 0, 1), reward 1775.0, memory_length 2000, epsilon 0.019505769735636575 total_time 727.0\n",
      "episode 8759, initial_state (1, 7, 4), reward 1380.0, memory_length 2000, epsilon 0.019418191420822895 total_time 723.0\n",
      "episode 8769, initial_state (4, 6, 4), reward 1513.0, memory_length 2000, epsilon 0.019331006323024036 total_time 727.0\n",
      "episode 8779, initial_state (2, 5, 2), reward 1265.0, memory_length 2000, epsilon 0.019244212676738804 total_time 733.0\n",
      "episode 8789, initial_state (0, 16, 3), reward 1229.0, memory_length 2000, epsilon 0.019157808724392877 total_time 727.0\n",
      "episode 8799, initial_state (1, 2, 1), reward 1171.0, memory_length 2000, epsilon 0.01907179271630328 total_time 722.0\n",
      "episode 8809, initial_state (0, 15, 3), reward 1009.0, memory_length 2000, epsilon 0.018986162910642895 total_time 724.0\n",
      "episode 8819, initial_state (2, 4, 4), reward 1404.0, memory_length 2000, epsilon 0.018900917573405243 total_time 721.0\n",
      "episode 8829, initial_state (2, 5, 2), reward 1528.0, memory_length 2000, epsilon 0.018816054978369336 total_time 722.0\n",
      "episode 8839, initial_state (2, 7, 2), reward 1427.0, memory_length 2000, epsilon 0.01873157340706471 total_time 727.0\n",
      "episode 8849, initial_state (3, 13, 3), reward 1390.0, memory_length 2000, epsilon 0.018647471148736664 total_time 721.0\n",
      "episode 8859, initial_state (2, 9, 2), reward 1476.0, memory_length 2000, epsilon 0.018563746500311607 total_time 727.0\n",
      "episode 8869, initial_state (4, 20, 4), reward 1316.0, memory_length 2000, epsilon 0.018480397766362528 total_time 721.0\n",
      "episode 8879, initial_state (1, 10, 3), reward 1193.0, memory_length 2000, epsilon 0.018397423259074737 total_time 725.0\n",
      "episode 8889, initial_state (3, 7, 4), reward 1333.0, memory_length 2000, epsilon 0.018314821298211603 total_time 722.0\n",
      "episode 8899, initial_state (1, 11, 6), reward 1550.0, memory_length 2000, epsilon 0.018232590211080604 total_time 723.0\n",
      "episode 8909, initial_state (4, 23, 3), reward 1490.0, memory_length 2000, epsilon 0.01815072833249942 total_time 724.0\n",
      "episode 8919, initial_state (4, 11, 4), reward 1379.0, memory_length 2000, epsilon 0.018069234004762222 total_time 722.0\n",
      "episode 8929, initial_state (3, 20, 2), reward 1370.0, memory_length 2000, epsilon 0.017988105577606068 total_time 728.0\n",
      "episode 8939, initial_state (2, 6, 1), reward 1402.0, memory_length 2000, epsilon 0.01790734140817754 total_time 721.0\n",
      "episode 8949, initial_state (4, 8, 5), reward 1473.0, memory_length 2000, epsilon 0.017826939860999452 total_time 725.0\n",
      "episode 8959, initial_state (2, 17, 3), reward 1055.0, memory_length 2000, epsilon 0.01774689930793773 total_time 726.0\n",
      "episode 8969, initial_state (1, 15, 4), reward 1557.0, memory_length 2000, epsilon 0.01766721812816845 total_time 723.0\n",
      "episode 8979, initial_state (4, 9, 0), reward 1504.0, memory_length 2000, epsilon 0.017587894708144966 total_time 731.0\n",
      "episode 8989, initial_state (2, 23, 0), reward 1249.0, memory_length 2000, epsilon 0.017508927441565333 total_time 730.0\n",
      "episode 8999, initial_state (0, 10, 0), reward 1395.0, memory_length 2000, epsilon 0.017430314729339702 total_time 724.0\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Saving Model 9000\n",
      "Saving Model 9000 Weights\n",
      "episode 9009, initial_state (4, 21, 2), reward 1696.0, memory_length 2000, epsilon 0.01735205497955796 total_time 726.0\n",
      "episode 9019, initial_state (3, 3, 4), reward 1328.0, memory_length 2000, epsilon 0.017274146607457505 total_time 726.0\n",
      "episode 9029, initial_state (1, 5, 6), reward 1321.0, memory_length 2000, epsilon 0.017196588035391158 total_time 721.0\n",
      "episode 9039, initial_state (1, 8, 4), reward 1098.0, memory_length 2000, epsilon 0.017119377692795142 total_time 723.0\n",
      "episode 9049, initial_state (1, 4, 4), reward 1402.0, memory_length 2000, epsilon 0.017042514016157407 total_time 721.0\n",
      "episode 9059, initial_state (4, 13, 6), reward 1566.0, memory_length 2000, epsilon 0.016965995448985874 total_time 723.0\n",
      "episode 9069, initial_state (2, 15, 2), reward 1518.0, memory_length 2000, epsilon 0.016889820441776934 total_time 724.0\n",
      "episode 9079, initial_state (0, 6, 6), reward 1352.0, memory_length 2000, epsilon 0.01681398745198411 total_time 725.0\n",
      "episode 9089, initial_state (4, 12, 6), reward 1481.0, memory_length 2000, epsilon 0.01673849494398674 total_time 728.0\n",
      "episode 9099, initial_state (4, 11, 2), reward 1465.0, memory_length 2000, epsilon 0.016663341389058964 total_time 725.0\n",
      "episode 9109, initial_state (0, 10, 0), reward 1521.0, memory_length 2000, epsilon 0.01658852526533873 total_time 725.0\n",
      "episode 9119, initial_state (0, 2, 3), reward 1465.0, memory_length 2000, epsilon 0.01651404505779698 total_time 721.0\n",
      "episode 9129, initial_state (1, 16, 5), reward 1473.0, memory_length 2000, epsilon 0.016439899258206978 total_time 726.0\n",
      "episode 9139, initial_state (1, 21, 4), reward 1575.0, memory_length 2000, epsilon 0.01636608636511372 total_time 730.0\n",
      "episode 9149, initial_state (0, 21, 6), reward 1494.0, memory_length 2000, epsilon 0.01629260488380361 total_time 728.0\n",
      "episode 9159, initial_state (2, 19, 5), reward 1252.0, memory_length 2000, epsilon 0.01621945332627414 total_time 724.0\n",
      "episode 9169, initial_state (3, 4, 4), reward 1508.0, memory_length 2000, epsilon 0.016146630211203782 total_time 728.0\n",
      "episode 9179, initial_state (3, 9, 6), reward 1176.0, memory_length 2000, epsilon 0.016074134063921965 total_time 728.0\n",
      "episode 9189, initial_state (0, 15, 6), reward 1000.0, memory_length 2000, epsilon 0.016001963416379213 total_time 725.0\n",
      "episode 9199, initial_state (4, 6, 6), reward 1059.0, memory_length 2000, epsilon 0.015930116807117457 total_time 724.0\n",
      "episode 9209, initial_state (2, 17, 4), reward 1440.0, memory_length 2000, epsilon 0.0158585927812404 total_time 721.0\n",
      "episode 9219, initial_state (1, 18, 3), reward 1382.0, memory_length 2000, epsilon 0.015787389890384078 total_time 726.0\n",
      "episode 9229, initial_state (2, 23, 6), reward 1185.0, memory_length 2000, epsilon 0.015716506692687537 total_time 721.0\n",
      "episode 9239, initial_state (3, 20, 0), reward 1341.0, memory_length 2000, epsilon 0.015645941752763567 total_time 722.0\n",
      "episode 9249, initial_state (3, 13, 6), reward 1398.0, memory_length 2000, epsilon 0.015575693641669735 total_time 723.0\n",
      "episode 9259, initial_state (0, 15, 3), reward 1397.0, memory_length 2000, epsilon 0.015505760936879396 total_time 728.0\n",
      "episode 9269, initial_state (2, 2, 1), reward 1148.0, memory_length 2000, epsilon 0.015436142222252886 total_time 724.0\n",
      "episode 9279, initial_state (4, 4, 3), reward 1494.0, memory_length 2000, epsilon 0.015366836088008857 total_time 721.0\n",
      "episode 9289, initial_state (4, 2, 6), reward 1577.0, memory_length 2000, epsilon 0.015297841130695733 total_time 722.0\n",
      "episode 9299, initial_state (2, 3, 0), reward 1476.0, memory_length 2000, epsilon 0.015229155953163247 total_time 722.0\n",
      "episode 9309, initial_state (2, 2, 6), reward 1503.0, memory_length 2000, epsilon 0.015160779164534215 total_time 728.0\n",
      "episode 9319, initial_state (3, 7, 4), reward 1245.0, memory_length 2000, epsilon 0.015092709380176338 total_time 726.0\n",
      "episode 9329, initial_state (1, 15, 0), reward 1344.0, memory_length 2000, epsilon 0.015024945221674151 total_time 721.0\n",
      "episode 9339, initial_state (0, 1, 6), reward 1377.0, memory_length 2000, epsilon 0.014957485316801142 total_time 735.0\n",
      "episode 9349, initial_state (3, 9, 5), reward 903.0, memory_length 2000, epsilon 0.01489032829949191 total_time 729.0\n",
      "episode 9359, initial_state (2, 11, 6), reward 1335.0, memory_length 2000, epsilon 0.014823472809814569 total_time 729.0\n",
      "episode 9369, initial_state (0, 19, 0), reward 1311.0, memory_length 2000, epsilon 0.014756917493943169 total_time 721.0\n",
      "episode 9379, initial_state (0, 4, 5), reward 1394.0, memory_length 2000, epsilon 0.014690661004130289 total_time 723.0\n",
      "episode 9389, initial_state (2, 14, 2), reward 1395.0, memory_length 2000, epsilon 0.01462470199867976 total_time 729.0\n",
      "episode 9399, initial_state (3, 15, 4), reward 1518.0, memory_length 2000, epsilon 0.014559039141919443 total_time 727.0\n",
      "episode 9409, initial_state (2, 3, 3), reward 986.0, memory_length 2000, epsilon 0.014493671104174257 total_time 722.0\n",
      "episode 9419, initial_state (0, 15, 3), reward 1487.0, memory_length 2000, epsilon 0.014428596561739202 total_time 721.0\n",
      "episode 9429, initial_state (2, 19, 6), reward 1419.0, memory_length 2000, epsilon 0.014363814196852571 total_time 731.0\n",
      "episode 9439, initial_state (0, 0, 2), reward 1238.0, memory_length 2000, epsilon 0.014299322697669278 total_time 726.0\n",
      "episode 9449, initial_state (4, 14, 4), reward 1693.0, memory_length 2000, epsilon 0.014235120758234228 total_time 724.0\n",
      "episode 9459, initial_state (2, 16, 1), reward 1683.0, memory_length 2000, epsilon 0.01417120707845597 total_time 722.0\n",
      "episode 9469, initial_state (1, 16, 3), reward 1451.0, memory_length 2000, epsilon 0.014107580364080306 total_time 722.0\n",
      "episode 9479, initial_state (4, 0, 6), reward 1359.0, memory_length 2000, epsilon 0.014044239326664092 total_time 726.0\n",
      "episode 9489, initial_state (0, 15, 2), reward 1610.0, memory_length 2000, epsilon 0.013981182683549174 total_time 723.0\n",
      "episode 9499, initial_state (2, 18, 3), reward 1299.0, memory_length 2000, epsilon 0.01391840915783634 total_time 724.0\n",
      "episode 9509, initial_state (2, 17, 0), reward 1538.0, memory_length 2000, epsilon 0.013855917478359573 total_time 721.0\n",
      "episode 9519, initial_state (1, 21, 2), reward 1098.0, memory_length 2000, epsilon 0.01379370637966022 total_time 724.0\n",
      "episode 9529, initial_state (0, 7, 0), reward 1720.0, memory_length 2000, epsilon 0.013731774601961406 total_time 729.0\n",
      "episode 9539, initial_state (0, 19, 3), reward 1297.0, memory_length 2000, epsilon 0.013670120891142524 total_time 725.0\n",
      "episode 9549, initial_state (3, 23, 5), reward 1468.0, memory_length 2000, epsilon 0.013608743998713826 total_time 724.0\n",
      "episode 9559, initial_state (3, 3, 4), reward 1020.0, memory_length 2000, epsilon 0.013547642681791127 total_time 721.0\n",
      "episode 9569, initial_state (1, 18, 6), reward 1459.0, memory_length 2000, epsilon 0.013486815703070677 total_time 732.0\n",
      "episode 9579, initial_state (4, 21, 0), reward 1463.0, memory_length 2000, epsilon 0.01342626183080408 total_time 726.0\n",
      "episode 9589, initial_state (4, 15, 4), reward 1575.0, memory_length 2000, epsilon 0.013365979838773358 total_time 728.0\n",
      "episode 9599, initial_state (3, 18, 2), reward 1445.0, memory_length 2000, epsilon 0.013305968506266117 total_time 731.0\n",
      "episode 9609, initial_state (4, 7, 3), reward 1323.0, memory_length 2000, epsilon 0.013246226618050805 total_time 721.0\n",
      "episode 9619, initial_state (1, 18, 5), reward 1417.0, memory_length 2000, epsilon 0.013186752964352153 total_time 721.0\n",
      "episode 9629, initial_state (2, 23, 0), reward 1073.0, memory_length 2000, epsilon 0.01312754634082664 total_time 726.0\n",
      "episode 9639, initial_state (0, 19, 6), reward 1108.0, memory_length 2000, epsilon 0.013068605548538122 total_time 721.0\n",
      "episode 9649, initial_state (3, 7, 5), reward 1305.0, memory_length 2000, epsilon 0.013009929393933545 total_time 730.0\n",
      "episode 9659, initial_state (1, 18, 3), reward 1375.0, memory_length 2000, epsilon 0.012951516688818759 total_time 721.0\n",
      "episode 9669, initial_state (3, 22, 4), reward 1406.0, memory_length 2000, epsilon 0.012893366250334494 total_time 723.0\n",
      "episode 9679, initial_state (1, 23, 5), reward 1100.0, memory_length 2000, epsilon 0.012835476900932384 total_time 724.0\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "episode 9689, initial_state (3, 10, 1), reward 1597.0, memory_length 2000, epsilon 0.012777847468351129 total_time 721.0\n",
      "episode 9699, initial_state (3, 0, 2), reward 1334.0, memory_length 2000, epsilon 0.012720476785592759 total_time 722.0\n",
      "episode 9709, initial_state (0, 6, 2), reward 1209.0, memory_length 2000, epsilon 0.012663363690898965 total_time 724.0\n",
      "episode 9719, initial_state (1, 5, 0), reward 1212.0, memory_length 2000, epsilon 0.01260650702772764 total_time 723.0\n",
      "episode 9729, initial_state (1, 22, 5), reward 1572.0, memory_length 2000, epsilon 0.012549905644729412 total_time 723.0\n",
      "episode 9739, initial_state (3, 3, 2), reward 1068.0, memory_length 2000, epsilon 0.012493558395724339 total_time 724.0\n",
      "episode 9749, initial_state (3, 4, 0), reward 1401.0, memory_length 2000, epsilon 0.012437464139678717 total_time 721.0\n",
      "episode 9759, initial_state (4, 23, 3), reward 1161.0, memory_length 2000, epsilon 0.01238162174068192 total_time 725.0\n",
      "episode 9769, initial_state (2, 18, 2), reward 1235.0, memory_length 2000, epsilon 0.01232603006792347 total_time 724.0\n",
      "episode 9779, initial_state (0, 9, 6), reward 1635.0, memory_length 2000, epsilon 0.012270687995670099 total_time 724.0\n",
      "episode 9789, initial_state (3, 13, 2), reward 1486.0, memory_length 2000, epsilon 0.01221559440324295 total_time 721.0\n",
      "episode 9799, initial_state (0, 21, 2), reward 1175.0, memory_length 2000, epsilon 0.01216074817499489 total_time 732.0\n",
      "episode 9809, initial_state (4, 0, 2), reward 1216.0, memory_length 2000, epsilon 0.012106148200287938 total_time 723.0\n",
      "episode 9819, initial_state (2, 9, 3), reward 916.0, memory_length 2000, epsilon 0.012051793373470718 total_time 722.0\n",
      "episode 9829, initial_state (4, 13, 0), reward 1580.0, memory_length 2000, epsilon 0.01199768259385614 total_time 725.0\n",
      "episode 9839, initial_state (2, 7, 5), reward 1558.0, memory_length 2000, epsilon 0.011943814765699069 total_time 726.0\n",
      "episode 9849, initial_state (0, 2, 5), reward 1418.0, memory_length 2000, epsilon 0.01189018879817414 total_time 733.0\n",
      "episode 9859, initial_state (2, 20, 3), reward 1294.0, memory_length 2000, epsilon 0.011836803605353696 total_time 725.0\n",
      "episode 9869, initial_state (3, 16, 2), reward 1591.0, memory_length 2000, epsilon 0.011783658106185728 total_time 726.0\n",
      "episode 9879, initial_state (2, 8, 5), reward 1240.0, memory_length 2000, epsilon 0.011730751224472081 total_time 727.0\n",
      "episode 9889, initial_state (1, 15, 4), reward 1162.0, memory_length 2000, epsilon 0.011678081888846585 total_time 726.0\n",
      "episode 9899, initial_state (0, 3, 5), reward 1427.0, memory_length 2000, epsilon 0.011625649032753397 total_time 734.0\n",
      "episode 9909, initial_state (2, 3, 4), reward 1581.0, memory_length 2000, epsilon 0.011573451594425401 total_time 724.0\n",
      "episode 9919, initial_state (1, 22, 2), reward 1505.0, memory_length 2000, epsilon 0.011521488516862665 total_time 723.0\n",
      "episode 9929, initial_state (4, 10, 3), reward 1632.0, memory_length 2000, epsilon 0.011469758747811101 total_time 723.0\n",
      "episode 9939, initial_state (1, 4, 5), reward 1350.0, memory_length 2000, epsilon 0.011418261239741124 total_time 725.0\n",
      "episode 9949, initial_state (0, 14, 6), reward 1420.0, memory_length 2000, epsilon 0.01136699494982643 total_time 724.0\n",
      "episode 9959, initial_state (1, 0, 6), reward 1515.0, memory_length 2000, epsilon 0.011315958839922908 total_time 724.0\n",
      "episode 9969, initial_state (4, 17, 5), reward 1447.0, memory_length 2000, epsilon 0.011265151876547568 total_time 724.0\n",
      "episode 9979, initial_state (1, 13, 1), reward 1644.0, memory_length 2000, epsilon 0.01121457303085768 total_time 721.0\n",
      "episode 9989, initial_state (0, 17, 5), reward 1370.0, memory_length 2000, epsilon 0.011164221278629883 total_time 722.0\n",
      "episode 9999, initial_state (4, 12, 5), reward 1130.0, memory_length 2000, epsilon 0.011114095600239479 total_time 721.0\n",
      "Saving Model 10000\n",
      "Saving Model 10000 Weights\n",
      "episode 10009, initial_state (0, 7, 3), reward 1353.0, memory_length 2000, epsilon 0.011064194980639775 total_time 723.0\n",
      "episode 10019, initial_state (2, 14, 0), reward 1550.0, memory_length 2000, epsilon 0.011014518409341503 total_time 723.0\n",
      "episode 10029, initial_state (0, 15, 5), reward 1562.0, memory_length 2000, epsilon 0.010965064880392401 total_time 724.0\n",
      "episode 10039, initial_state (4, 3, 2), reward 1171.0, memory_length 2000, epsilon 0.01091583339235682 total_time 722.0\n",
      "episode 10049, initial_state (3, 1, 2), reward 1191.0, memory_length 2000, epsilon 0.010866822948295447 total_time 722.0\n",
      "episode 10059, initial_state (3, 3, 5), reward 1712.0, memory_length 2000, epsilon 0.010818032555745115 total_time 724.0\n",
      "episode 10069, initial_state (0, 5, 0), reward 1555.0, memory_length 2000, epsilon 0.010769461226698715 total_time 724.0\n",
      "episode 10079, initial_state (0, 17, 2), reward 1176.0, memory_length 2000, epsilon 0.010721107977585154 total_time 724.0\n",
      "episode 10089, initial_state (4, 16, 3), reward 1274.0, memory_length 2000, epsilon 0.010672971829249495 total_time 729.0\n",
      "episode 10099, initial_state (0, 17, 6), reward 1553.0, memory_length 2000, epsilon 0.010625051806933095 total_time 726.0\n",
      "episode 10109, initial_state (1, 5, 3), reward 1263.0, memory_length 2000, epsilon 0.010577346940253857 total_time 731.0\n",
      "episode 10119, initial_state (1, 7, 5), reward 1553.0, memory_length 2000, epsilon 0.010529856263186614 total_time 728.0\n",
      "episode 10129, initial_state (0, 10, 2), reward 1464.0, memory_length 2000, epsilon 0.010482578814043514 total_time 725.0\n",
      "episode 10139, initial_state (0, 21, 0), reward 1486.0, memory_length 2000, epsilon 0.010435513635454603 total_time 725.0\n",
      "episode 10149, initial_state (4, 14, 4), reward 1292.0, memory_length 2000, epsilon 0.010388659774348408 total_time 727.0\n",
      "episode 10159, initial_state (4, 15, 3), reward 1553.0, memory_length 2000, epsilon 0.010342016281932641 total_time 721.0\n",
      "episode 10169, initial_state (4, 9, 4), reward 1452.0, memory_length 2000, epsilon 0.010295582213674995 total_time 726.0\n",
      "episode 10179, initial_state (4, 19, 6), reward 1231.0, memory_length 2000, epsilon 0.010249356629283984 total_time 725.0\n",
      "episode 10189, initial_state (2, 4, 2), reward 1335.0, memory_length 2000, epsilon 0.01020333859268995 total_time 723.0\n",
      "episode 10199, initial_state (1, 22, 6), reward 1324.0, memory_length 2000, epsilon 0.010157527172026084 total_time 724.0\n",
      "episode 10209, initial_state (0, 12, 6), reward 1136.0, memory_length 2000, epsilon 0.01011192143960955 total_time 725.0\n",
      "episode 10219, initial_state (3, 17, 2), reward 1202.0, memory_length 2000, epsilon 0.010066520471922715 total_time 728.0\n",
      "episode 10229, initial_state (2, 3, 2), reward 1555.0, memory_length 2000, epsilon 0.010021323349594418 total_time 724.0\n",
      "episode 10239, initial_state (3, 6, 6), reward 1172.0, memory_length 2000, epsilon 0.009976329157381395 total_time 726.0\n",
      "episode 10249, initial_state (0, 18, 2), reward 1274.0, memory_length 2000, epsilon 0.009931536984149716 total_time 726.0\n",
      "episode 10259, initial_state (4, 15, 5), reward 1562.0, memory_length 2000, epsilon 0.00988694592285634 total_time 721.0\n",
      "episode 10269, initial_state (0, 6, 0), reward 1395.0, memory_length 2000, epsilon 0.009842555070530765 total_time 727.0\n",
      "episode 10279, initial_state (4, 4, 3), reward 1233.0, memory_length 2000, epsilon 0.009798363528256693 total_time 721.0\n",
      "episode 10289, initial_state (1, 19, 5), reward 1319.0, memory_length 2000, epsilon 0.009754370401153895 total_time 725.0\n",
      "episode 10299, initial_state (0, 5, 0), reward 1347.0, memory_length 2000, epsilon 0.009710574798360043 total_time 722.0\n",
      "episode 10309, initial_state (2, 14, 0), reward 1307.0, memory_length 2000, epsilon 0.009666975833012681 total_time 722.0\n",
      "episode 10319, initial_state (3, 9, 2), reward 1683.0, memory_length 2000, epsilon 0.009623572622231274 total_time 725.0\n",
      "episode 10329, initial_state (3, 19, 6), reward 1347.0, memory_length 2000, epsilon 0.009580364287099328 total_time 730.0\n",
      "episode 10339, initial_state (1, 4, 2), reward 1595.0, memory_length 2000, epsilon 0.009537349952646563 total_time 728.0\n",
      "episode 10349, initial_state (4, 4, 3), reward 1213.0, memory_length 2000, epsilon 0.009494528747831245 total_time 729.0\n",
      "episode 10359, initial_state (4, 4, 4), reward 1275.0, memory_length 2000, epsilon 0.009451899805522513 total_time 721.0\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "episode 10369, initial_state (1, 16, 4), reward 1292.0, memory_length 2000, epsilon 0.00940946226248283 total_time 727.0\n",
      "episode 10379, initial_state (2, 15, 5), reward 1681.0, memory_length 2000, epsilon 0.009367215259350505 total_time 721.0\n",
      "episode 10389, initial_state (4, 16, 6), reward 1574.0, memory_length 2000, epsilon 0.009325157940622266 total_time 722.0\n",
      "episode 10399, initial_state (3, 9, 1), reward 1347.0, memory_length 2000, epsilon 0.009283289454635978 total_time 728.0\n",
      "episode 10409, initial_state (2, 21, 0), reward 1180.0, memory_length 2000, epsilon 0.009241608953553374 total_time 724.0\n",
      "episode 10419, initial_state (3, 10, 4), reward 1054.0, memory_length 2000, epsilon 0.009200115593342877 total_time 721.0\n",
      "episode 10429, initial_state (2, 9, 2), reward 1564.0, memory_length 2000, epsilon 0.009158808533762535 total_time 723.0\n",
      "episode 10439, initial_state (3, 3, 3), reward 1290.0, memory_length 2000, epsilon 0.009117686938342966 total_time 729.0\n",
      "episode 10449, initial_state (2, 21, 4), reward 1302.0, memory_length 2000, epsilon 0.009076749974370463 total_time 727.0\n",
      "episode 10459, initial_state (2, 11, 0), reward 1500.0, memory_length 2000, epsilon 0.009035996812870107 total_time 724.0\n",
      "episode 10469, initial_state (2, 11, 2), reward 1233.0, memory_length 2000, epsilon 0.008995426628588985 total_time 728.0\n",
      "episode 10479, initial_state (4, 6, 0), reward 1418.0, memory_length 2000, epsilon 0.008955038599979486 total_time 726.0\n",
      "episode 10489, initial_state (0, 17, 1), reward 1488.0, memory_length 2000, epsilon 0.00891483190918264 total_time 724.0\n",
      "episode 10499, initial_state (0, 2, 6), reward 1000.0, memory_length 2000, epsilon 0.008874805742011584 total_time 724.0\n",
      "episode 10509, initial_state (1, 14, 2), reward 1095.0, memory_length 2000, epsilon 0.008834959287935071 total_time 725.0\n",
      "episode 10519, initial_state (3, 3, 1), reward 1460.0, memory_length 2000, epsilon 0.008795291740061044 total_time 727.0\n",
      "episode 10529, initial_state (1, 15, 2), reward 1311.0, memory_length 2000, epsilon 0.008755802295120308 total_time 726.0\n",
      "episode 10539, initial_state (2, 7, 0), reward 1503.0, memory_length 2000, epsilon 0.008716490153450239 total_time 733.0\n",
      "episode 10549, initial_state (0, 16, 6), reward 1565.0, memory_length 2000, epsilon 0.008677354518978633 total_time 722.0\n",
      "episode 10559, initial_state (4, 9, 1), reward 1496.0, memory_length 2000, epsilon 0.008638394599207556 total_time 723.0\n",
      "episode 10569, initial_state (2, 4, 1), reward 1442.0, memory_length 2000, epsilon 0.008599609605197298 total_time 723.0\n",
      "episode 10579, initial_state (4, 15, 5), reward 1522.0, memory_length 2000, epsilon 0.008560998751550407 total_time 727.0\n",
      "episode 10589, initial_state (0, 7, 6), reward 1335.0, memory_length 2000, epsilon 0.008522561256395786 total_time 725.0\n",
      "episode 10599, initial_state (4, 15, 5), reward 1605.0, memory_length 2000, epsilon 0.008484296341372828 total_time 724.0\n",
      "episode 10609, initial_state (1, 9, 1), reward 1581.0, memory_length 2000, epsilon 0.008446203231615703 total_time 727.0\n",
      "episode 10619, initial_state (3, 13, 0), reward 1061.0, memory_length 2000, epsilon 0.008408281155737638 total_time 726.0\n",
      "episode 10629, initial_state (1, 12, 1), reward 1783.0, memory_length 2000, epsilon 0.0083705293458153 total_time 728.0\n",
      "episode 10639, initial_state (1, 21, 6), reward 1842.0, memory_length 2000, epsilon 0.008332947037373256 total_time 730.0\n",
      "episode 10649, initial_state (3, 10, 2), reward 1312.0, memory_length 2000, epsilon 0.008295533469368462 total_time 722.0\n",
      "episode 10659, initial_state (4, 7, 5), reward 1234.0, memory_length 2000, epsilon 0.00825828788417489 total_time 723.0\n",
      "episode 10669, initial_state (3, 23, 1), reward 1492.0, memory_length 2000, epsilon 0.008221209527568173 total_time 723.0\n",
      "episode 10679, initial_state (0, 23, 5), reward 1244.0, memory_length 2000, epsilon 0.00818429764871032 total_time 727.0\n",
      "episode 10689, initial_state (3, 0, 1), reward 1404.0, memory_length 2000, epsilon 0.008147551500134529 total_time 726.0\n",
      "episode 10699, initial_state (4, 15, 0), reward 1445.0, memory_length 2000, epsilon 0.008110970337730023 total_time 722.0\n",
      "episode 10709, initial_state (2, 14, 1), reward 1382.0, memory_length 2000, epsilon 0.00807455342072702 total_time 723.0\n",
      "episode 10719, initial_state (4, 14, 3), reward 1314.0, memory_length 2000, epsilon 0.008038300011681706 total_time 730.0\n",
      "episode 10729, initial_state (4, 21, 5), reward 1166.0, memory_length 2000, epsilon 0.00800220937646131 total_time 727.0\n",
      "episode 10739, initial_state (3, 2, 1), reward 1467.0, memory_length 2000, epsilon 0.007966280784229242 total_time 725.0\n",
      "episode 10749, initial_state (0, 3, 3), reward 1590.0, memory_length 2000, epsilon 0.007930513507430265 total_time 728.0\n",
      "episode 10759, initial_state (1, 6, 1), reward 1346.0, memory_length 2000, epsilon 0.007894906821775812 total_time 732.0\n",
      "episode 10769, initial_state (4, 19, 6), reward 1319.0, memory_length 2000, epsilon 0.00785946000622928 total_time 730.0\n",
      "episode 10779, initial_state (2, 1, 2), reward 1410.0, memory_length 2000, epsilon 0.007824172342991443 total_time 725.0\n",
      "episode 10789, initial_state (1, 7, 5), reward 1515.0, memory_length 2000, epsilon 0.007789043117485917 total_time 721.0\n",
      "episode 10799, initial_state (1, 13, 5), reward 1530.0, memory_length 2000, epsilon 0.007754071618344688 total_time 726.0\n",
      "episode 10809, initial_state (3, 9, 1), reward 1335.0, memory_length 2000, epsilon 0.007719257137393692 total_time 731.0\n",
      "episode 10819, initial_state (0, 10, 4), reward 1461.0, memory_length 2000, epsilon 0.007684598969638507 total_time 726.0\n",
      "episode 10829, initial_state (2, 12, 3), reward 1508.0, memory_length 2000, epsilon 0.007650096413250051 total_time 733.0\n",
      "episode 10839, initial_state (0, 8, 3), reward 1523.0, memory_length 2000, epsilon 0.007615748769550378 total_time 729.0\n",
      "episode 10849, initial_state (1, 1, 5), reward 1658.0, memory_length 2000, epsilon 0.007581555342998536 total_time 724.0\n",
      "episode 10859, initial_state (1, 2, 5), reward 1474.0, memory_length 2000, epsilon 0.007547515441176456 total_time 723.0\n",
      "episode 10869, initial_state (0, 17, 3), reward 1191.0, memory_length 2000, epsilon 0.007513628374774969 total_time 728.0\n",
      "episode 10879, initial_state (4, 2, 6), reward 1440.0, memory_length 2000, epsilon 0.007479893457579823 total_time 726.0\n",
      "episode 10889, initial_state (0, 14, 1), reward 987.0, memory_length 2000, epsilon 0.007446310006457792 total_time 727.0\n",
      "episode 10899, initial_state (0, 17, 0), reward 1388.0, memory_length 2000, epsilon 0.007412877341342849 total_time 722.0\n",
      "episode 10909, initial_state (4, 0, 6), reward 1801.0, memory_length 2000, epsilon 0.00737959478522237 total_time 724.0\n",
      "episode 10919, initial_state (4, 0, 5), reward 1045.0, memory_length 2000, epsilon 0.007346461664123464 total_time 733.0\n",
      "episode 10929, initial_state (4, 23, 5), reward 1043.0, memory_length 2000, epsilon 0.007313477307099295 total_time 722.0\n",
      "episode 10939, initial_state (3, 10, 3), reward 1098.0, memory_length 2000, epsilon 0.007280641046215507 total_time 722.0\n",
      "episode 10949, initial_state (4, 23, 5), reward 1594.0, memory_length 2000, epsilon 0.007247952216536702 total_time 724.0\n",
      "episode 10959, initial_state (4, 5, 0), reward 1154.0, memory_length 2000, epsilon 0.007215410156112947 total_time 721.0\n",
      "episode 10969, initial_state (1, 8, 0), reward 1493.0, memory_length 2000, epsilon 0.007183014205966415 total_time 726.0\n",
      "episode 10979, initial_state (2, 3, 3), reward 1617.0, memory_length 2000, epsilon 0.007150763710078007 total_time 728.0\n",
      "episode 10989, initial_state (2, 13, 0), reward 1759.0, memory_length 2000, epsilon 0.00711865801537408 total_time 721.0\n",
      "episode 10999, initial_state (0, 8, 3), reward 1066.0, memory_length 2000, epsilon 0.007086696471713226 total_time 723.0\n",
      "Saving Model 11000\n",
      "Saving Model 11000 Weights\n",
      "episode 11009, initial_state (0, 9, 6), reward 1114.0, memory_length 2000, epsilon 0.0070548784318730795 total_time 721.0\n",
      "episode 11019, initial_state (4, 1, 6), reward 1351.0, memory_length 2000, epsilon 0.007023203251537254 total_time 725.0\n",
      "episode 11029, initial_state (0, 9, 3), reward 1388.0, memory_length 2000, epsilon 0.006991670289282264 total_time 725.0\n",
      "episode 11039, initial_state (2, 10, 5), reward 1089.0, memory_length 2000, epsilon 0.006960278906564548 total_time 735.0\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "episode 11049, initial_state (1, 2, 5), reward 1188.0, memory_length 2000, epsilon 0.006929028467707533 total_time 724.0\n",
      "episode 11059, initial_state (2, 17, 4), reward 1302.0, memory_length 2000, epsilon 0.0068979183398887685 total_time 725.0\n",
      "episode 11069, initial_state (0, 17, 6), reward 1485.0, memory_length 2000, epsilon 0.006866947893127093 total_time 721.0\n",
      "episode 11079, initial_state (3, 20, 3), reward 1275.0, memory_length 2000, epsilon 0.006836116500269907 total_time 726.0\n",
      "episode 11089, initial_state (0, 5, 4), reward 1393.0, memory_length 2000, epsilon 0.00680542353698045 total_time 722.0\n",
      "episode 11099, initial_state (4, 3, 2), reward 1697.0, memory_length 2000, epsilon 0.006774868381725169 total_time 727.0\n",
      "episode 11109, initial_state (3, 13, 5), reward 1481.0, memory_length 2000, epsilon 0.0067444504157611294 total_time 723.0\n",
      "episode 11119, initial_state (1, 4, 1), reward 1501.0, memory_length 2000, epsilon 0.006714169023123471 total_time 723.0\n",
      "episode 11129, initial_state (3, 22, 2), reward 1411.0, memory_length 2000, epsilon 0.006684023590612963 total_time 722.0\n",
      "episode 11139, initial_state (4, 12, 3), reward 1516.0, memory_length 2000, epsilon 0.006654013507783567 total_time 725.0\n",
      "episode 11149, initial_state (4, 23, 4), reward 1160.0, memory_length 2000, epsilon 0.006624138166930081 total_time 724.0\n",
      "episode 11159, initial_state (1, 11, 3), reward 1834.0, memory_length 2000, epsilon 0.006594396963075838 total_time 724.0\n",
      "episode 11169, initial_state (1, 13, 1), reward 1505.0, memory_length 2000, epsilon 0.0065647892939604295 total_time 726.0\n",
      "episode 11179, initial_state (0, 23, 5), reward 1315.0, memory_length 2000, epsilon 0.0065353145600275525 total_time 721.0\n",
      "episode 11189, initial_state (0, 3, 5), reward 1448.0, memory_length 2000, epsilon 0.006505972164412835 total_time 728.0\n",
      "episode 11199, initial_state (1, 3, 3), reward 1372.0, memory_length 2000, epsilon 0.006476761512931767 total_time 726.0\n",
      "episode 11209, initial_state (1, 10, 6), reward 1401.0, memory_length 2000, epsilon 0.00644768201406766 total_time 722.0\n",
      "episode 11219, initial_state (2, 23, 4), reward 1610.0, memory_length 2000, epsilon 0.00641873307895966 total_time 721.0\n",
      "episode 11229, initial_state (0, 16, 2), reward 920.0, memory_length 2000, epsilon 0.0063899141213908444 total_time 725.0\n",
      "episode 11239, initial_state (0, 8, 6), reward 1445.0, memory_length 2000, epsilon 0.006361224557776339 total_time 728.0\n",
      "episode 11249, initial_state (0, 1, 5), reward 1484.0, memory_length 2000, epsilon 0.006332663807151501 total_time 723.0\n",
      "episode 11259, initial_state (3, 9, 2), reward 1630.0, memory_length 2000, epsilon 0.006304231291160159 total_time 725.0\n",
      "episode 11269, initial_state (1, 1, 4), reward 1535.0, memory_length 2000, epsilon 0.006275926434042882 total_time 727.0\n",
      "episode 11279, initial_state (0, 17, 3), reward 1441.0, memory_length 2000, epsilon 0.0062477486626253504 total_time 726.0\n",
      "episode 11289, initial_state (1, 3, 2), reward 1319.0, memory_length 2000, epsilon 0.006219697406306731 total_time 722.0\n",
      "episode 11299, initial_state (3, 14, 6), reward 1157.0, memory_length 2000, epsilon 0.006191772097048125 total_time 723.0\n",
      "episode 11309, initial_state (1, 11, 5), reward 1867.0, memory_length 2000, epsilon 0.006163972169361065 total_time 727.0\n",
      "episode 11319, initial_state (1, 0, 4), reward 1220.0, memory_length 2000, epsilon 0.006136297060296072 total_time 725.0\n",
      "episode 11329, initial_state (3, 6, 2), reward 1513.0, memory_length 2000, epsilon 0.006108746209431229 total_time 725.0\n",
      "episode 11339, initial_state (3, 7, 6), reward 1599.0, memory_length 2000, epsilon 0.006081319058860873 total_time 723.0\n",
      "episode 11349, initial_state (4, 7, 3), reward 1082.0, memory_length 2000, epsilon 0.006054015053184264 total_time 727.0\n",
      "episode 11359, initial_state (4, 8, 6), reward 1284.0, memory_length 2000, epsilon 0.006026833639494356 total_time 723.0\n",
      "episode 11369, initial_state (1, 7, 5), reward 1584.0, memory_length 2000, epsilon 0.005999774267366598 total_time 721.0\n",
      "episode 11379, initial_state (0, 22, 3), reward 1571.0, memory_length 2000, epsilon 0.0059728363888477696 total_time 725.0\n",
      "episode 11389, initial_state (2, 1, 5), reward 1653.0, memory_length 2000, epsilon 0.005946019458444914 total_time 724.0\n",
      "episode 11399, initial_state (2, 20, 0), reward 1672.0, memory_length 2000, epsilon 0.005919322933114276 total_time 721.0\n",
      "episode 11409, initial_state (1, 23, 4), reward 1365.0, memory_length 2000, epsilon 0.0058927462722503056 total_time 721.0\n",
      "episode 11419, initial_state (0, 15, 0), reward 1548.0, memory_length 2000, epsilon 0.0058662889376747155 total_time 721.0\n",
      "episode 11429, initial_state (4, 1, 1), reward 1743.0, memory_length 2000, epsilon 0.005839950393625567 total_time 728.0\n",
      "episode 11439, initial_state (1, 6, 1), reward 1476.0, memory_length 2000, epsilon 0.005813730106746448 total_time 722.0\n",
      "episode 11449, initial_state (1, 5, 6), reward 1433.0, memory_length 2000, epsilon 0.0057876275460756535 total_time 727.0\n",
      "episode 11459, initial_state (0, 8, 6), reward 1773.0, memory_length 2000, epsilon 0.005761642183035438 total_time 727.0\n",
      "episode 11469, initial_state (3, 13, 3), reward 1621.0, memory_length 2000, epsilon 0.005735773491421316 total_time 722.0\n",
      "episode 11479, initial_state (3, 9, 4), reward 1456.0, memory_length 2000, epsilon 0.005710020947391391 total_time 722.0\n",
      "episode 11489, initial_state (3, 7, 3), reward 1630.0, memory_length 2000, epsilon 0.0056843840294557675 total_time 722.0\n",
      "episode 11499, initial_state (1, 13, 5), reward 1188.0, memory_length 2000, epsilon 0.005658862218465985 total_time 729.0\n",
      "episode 11509, initial_state (1, 0, 5), reward 1547.0, memory_length 2000, epsilon 0.005633454997604495 total_time 721.0\n",
      "episode 11519, initial_state (0, 23, 5), reward 1461.0, memory_length 2000, epsilon 0.005608161852374216 total_time 723.0\n",
      "episode 11529, initial_state (2, 3, 4), reward 1478.0, memory_length 2000, epsilon 0.005582982270588079 total_time 725.0\n",
      "episode 11539, initial_state (0, 23, 6), reward 1034.0, memory_length 2000, epsilon 0.0055579157423587 total_time 727.0\n",
      "episode 11549, initial_state (0, 5, 4), reward 1559.0, memory_length 2000, epsilon 0.005532961760088025 total_time 723.0\n",
      "episode 11559, initial_state (4, 5, 3), reward 1368.0, memory_length 2000, epsilon 0.0055081198184570605 total_time 721.0\n",
      "episode 11569, initial_state (0, 3, 6), reward 1247.0, memory_length 2000, epsilon 0.005483389414415639 total_time 725.0\n",
      "episode 11579, initial_state (3, 14, 3), reward 1521.0, memory_length 2000, epsilon 0.005458770047172238 total_time 726.0\n",
      "episode 11589, initial_state (0, 9, 3), reward 1242.0, memory_length 2000, epsilon 0.005434261218183821 total_time 721.0\n",
      "episode 11599, initial_state (4, 21, 2), reward 1604.0, memory_length 2000, epsilon 0.005409862431145768 total_time 727.0\n",
      "episode 11609, initial_state (0, 7, 2), reward 1458.0, memory_length 2000, epsilon 0.0053855731919818075 total_time 726.0\n",
      "episode 11619, initial_state (2, 22, 1), reward 1675.0, memory_length 2000, epsilon 0.005361393008834015 total_time 726.0\n",
      "episode 11629, initial_state (3, 20, 0), reward 1539.0, memory_length 2000, epsilon 0.005337321392052863 total_time 728.0\n",
      "episode 11639, initial_state (2, 20, 6), reward 1796.0, memory_length 2000, epsilon 0.0053133578541872785 total_time 727.0\n",
      "episode 11649, initial_state (3, 5, 4), reward 1692.0, memory_length 2000, epsilon 0.005289501909974805 total_time 725.0\n",
      "episode 11659, initial_state (0, 16, 2), reward 1393.0, memory_length 2000, epsilon 0.005265753076331758 total_time 721.0\n",
      "episode 11669, initial_state (2, 9, 5), reward 1435.0, memory_length 2000, epsilon 0.005242110872343444 total_time 728.0\n",
      "episode 11679, initial_state (3, 19, 1), reward 1436.0, memory_length 2000, epsilon 0.005218574819254428 total_time 732.0\n",
      "episode 11689, initial_state (0, 12, 4), reward 1639.0, memory_length 2000, epsilon 0.0051951444404588236 total_time 722.0\n",
      "episode 11699, initial_state (2, 14, 0), reward 1196.0, memory_length 2000, epsilon 0.005171819261490662 total_time 723.0\n",
      "episode 11709, initial_state (1, 3, 1), reward 1512.0, memory_length 2000, epsilon 0.005148598810014275 total_time 721.0\n",
      "episode 11719, initial_state (2, 12, 5), reward 1667.0, memory_length 2000, epsilon 0.005125482615814723 total_time 723.0\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "episode 11729, initial_state (1, 5, 1), reward 1470.0, memory_length 2000, epsilon 0.00510247021078829 total_time 722.0\n",
      "episode 11739, initial_state (0, 2, 1), reward 1236.0, memory_length 2000, epsilon 0.005079561128932979 total_time 722.0\n",
      "episode 11749, initial_state (2, 11, 3), reward 1544.0, memory_length 2000, epsilon 0.005056754906339104 total_time 724.0\n",
      "episode 11759, initial_state (1, 20, 2), reward 1267.0, memory_length 2000, epsilon 0.005034051081179877 total_time 723.0\n",
      "episode 11769, initial_state (4, 10, 3), reward 1396.0, memory_length 2000, epsilon 0.005011449193702063 total_time 725.0\n",
      "episode 11779, initial_state (2, 19, 4), reward 1265.0, memory_length 2000, epsilon 0.004988948786216675 total_time 727.0\n",
      "episode 11789, initial_state (1, 13, 2), reward 1212.0, memory_length 2000, epsilon 0.004966549403089681 total_time 723.0\n",
      "episode 11799, initial_state (3, 20, 1), reward 1054.0, memory_length 2000, epsilon 0.004944250590732812 total_time 726.0\n",
      "episode 11809, initial_state (3, 10, 2), reward 1717.0, memory_length 2000, epsilon 0.004922051897594358 total_time 722.0\n",
      "episode 11819, initial_state (4, 6, 3), reward 1733.0, memory_length 2000, epsilon 0.004899952874150021 total_time 726.0\n",
      "episode 11829, initial_state (1, 11, 6), reward 952.0, memory_length 2000, epsilon 0.004877953072893825 total_time 724.0\n",
      "episode 11839, initial_state (1, 15, 2), reward 1278.0, memory_length 2000, epsilon 0.004856052048329043 total_time 728.0\n",
      "episode 11849, initial_state (3, 7, 4), reward 1392.0, memory_length 2000, epsilon 0.004834249356959174 total_time 724.0\n",
      "episode 11859, initial_state (3, 8, 3), reward 1306.0, memory_length 2000, epsilon 0.004812544557278973 total_time 729.0\n",
      "episode 11869, initial_state (4, 0, 4), reward 1474.0, memory_length 2000, epsilon 0.004790937209765509 total_time 723.0\n",
      "episode 11879, initial_state (2, 7, 1), reward 1522.0, memory_length 2000, epsilon 0.0047694268768692545 total_time 722.0\n",
      "episode 11889, initial_state (4, 16, 5), reward 1554.0, memory_length 2000, epsilon 0.004748013123005238 total_time 726.0\n",
      "episode 11899, initial_state (4, 18, 1), reward 1755.0, memory_length 2000, epsilon 0.004726695514544204 total_time 729.0\n",
      "episode 11909, initial_state (0, 6, 5), reward 1585.0, memory_length 2000, epsilon 0.004705473619803855 total_time 733.0\n",
      "episode 11919, initial_state (3, 19, 1), reward 1340.0, memory_length 2000, epsilon 0.004684347009040098 total_time 723.0\n",
      "episode 11929, initial_state (2, 20, 6), reward 1262.0, memory_length 2000, epsilon 0.004663315254438345 total_time 730.0\n",
      "episode 11939, initial_state (4, 7, 5), reward 1180.0, memory_length 2000, epsilon 0.004642377930104849 total_time 723.0\n",
      "episode 11949, initial_state (1, 10, 5), reward 1382.0, memory_length 2000, epsilon 0.004621534612058068 total_time 722.0\n",
      "episode 11959, initial_state (2, 13, 0), reward 1300.0, memory_length 2000, epsilon 0.004600784878220106 total_time 723.0\n",
      "episode 11969, initial_state (2, 22, 2), reward 1256.0, memory_length 2000, epsilon 0.004580128308408142 total_time 725.0\n",
      "episode 11979, initial_state (4, 16, 5), reward 1328.0, memory_length 2000, epsilon 0.00455956448432593 total_time 728.0\n",
      "episode 11989, initial_state (0, 20, 6), reward 1404.0, memory_length 2000, epsilon 0.004539092989555337 total_time 723.0\n",
      "episode 11999, initial_state (3, 16, 4), reward 1351.0, memory_length 2000, epsilon 0.004518713409547883 total_time 726.0\n",
      "Saving Model 12000\n",
      "Saving Model 12000 Weights\n",
      "episode 12009, initial_state (1, 13, 3), reward 1494.0, memory_length 2000, epsilon 0.004498425331616382 total_time 722.0\n",
      "episode 12019, initial_state (3, 15, 2), reward 1332.0, memory_length 2000, epsilon 0.004478228344926565 total_time 723.0\n",
      "episode 12029, initial_state (2, 16, 0), reward 1554.0, memory_length 2000, epsilon 0.004458122040488757 total_time 724.0\n",
      "episode 12039, initial_state (0, 4, 5), reward 1262.0, memory_length 2000, epsilon 0.004438106011149611 total_time 721.0\n",
      "episode 12049, initial_state (0, 18, 4), reward 1517.0, memory_length 2000, epsilon 0.004418179851583843 total_time 722.0\n",
      "episode 12059, initial_state (2, 20, 1), reward 1757.0, memory_length 2000, epsilon 0.004398343158286044 total_time 722.0\n",
      "episode 12069, initial_state (1, 0, 6), reward 1377.0, memory_length 2000, epsilon 0.004378595529562495 total_time 726.0\n",
      "episode 12079, initial_state (2, 22, 1), reward 1703.0, memory_length 2000, epsilon 0.004358936565523043 total_time 726.0\n",
      "episode 12089, initial_state (1, 5, 4), reward 1666.0, memory_length 2000, epsilon 0.004339365868072992 total_time 724.0\n",
      "episode 12099, initial_state (0, 16, 6), reward 1410.0, memory_length 2000, epsilon 0.004319883040905055 total_time 724.0\n",
      "episode 12109, initial_state (0, 15, 5), reward 1370.0, memory_length 2000, epsilon 0.0043004876894913065 total_time 721.0\n",
      "episode 12119, initial_state (1, 9, 0), reward 1141.0, memory_length 2000, epsilon 0.004281179421075223 total_time 721.0\n",
      "episode 12129, initial_state (1, 7, 5), reward 1486.0, memory_length 2000, epsilon 0.004261957844663709 total_time 723.0\n",
      "episode 12139, initial_state (2, 22, 4), reward 1508.0, memory_length 2000, epsilon 0.004242822571019185 total_time 724.0\n",
      "episode 12149, initial_state (2, 22, 4), reward 1598.0, memory_length 2000, epsilon 0.00422377321265171 total_time 723.0\n",
      "episode 12159, initial_state (2, 2, 6), reward 1530.0, memory_length 2000, epsilon 0.004204809383811119 total_time 726.0\n",
      "episode 12169, initial_state (2, 18, 5), reward 1505.0, memory_length 2000, epsilon 0.004185930700479232 total_time 721.0\n",
      "episode 12179, initial_state (3, 11, 5), reward 1505.0, memory_length 2000, epsilon 0.004167136780362068 total_time 726.0\n",
      "episode 12189, initial_state (3, 17, 2), reward 1506.0, memory_length 2000, epsilon 0.004148427242882101 total_time 721.0\n",
      "episode 12199, initial_state (0, 12, 3), reward 1527.0, memory_length 2000, epsilon 0.004129801709170563 total_time 728.0\n",
      "episode 12209, initial_state (1, 19, 0), reward 1764.0, memory_length 2000, epsilon 0.004111259802059752 total_time 731.0\n",
      "episode 12219, initial_state (1, 1, 0), reward 1557.0, memory_length 2000, epsilon 0.004092801146075419 total_time 725.0\n",
      "episode 12229, initial_state (1, 9, 0), reward 1500.0, memory_length 2000, epsilon 0.004074425367429149 total_time 726.0\n",
      "episode 12239, initial_state (0, 19, 6), reward 1433.0, memory_length 2000, epsilon 0.004056132094010797 total_time 722.0\n",
      "episode 12249, initial_state (1, 7, 3), reward 1315.0, memory_length 2000, epsilon 0.0040379209553809544 total_time 724.0\n",
      "episode 12259, initial_state (2, 14, 6), reward 1660.0, memory_length 2000, epsilon 0.004019791582763435 total_time 724.0\n",
      "episode 12269, initial_state (4, 3, 1), reward 1368.0, memory_length 2000, epsilon 0.0040017436090378266 total_time 727.0\n",
      "episode 12279, initial_state (4, 12, 4), reward 1470.0, memory_length 2000, epsilon 0.003983776668732045 total_time 724.0\n",
      "episode 12289, initial_state (1, 21, 5), reward 1370.0, memory_length 2000, epsilon 0.003965890398014935 total_time 723.0\n",
      "episode 12299, initial_state (3, 5, 1), reward 1328.0, memory_length 2000, epsilon 0.003948084434688907 total_time 723.0\n",
      "episode 12309, initial_state (3, 22, 3), reward 1265.0, memory_length 2000, epsilon 0.0039303584181825885 total_time 724.0\n",
      "episode 12319, initial_state (2, 4, 2), reward 1339.0, memory_length 2000, epsilon 0.003912711989543543 total_time 721.0\n",
      "episode 12329, initial_state (1, 6, 3), reward 1440.0, memory_length 2000, epsilon 0.0038951447914309855 total_time 722.0\n",
      "episode 12339, initial_state (1, 11, 6), reward 1544.0, memory_length 2000, epsilon 0.003877656468108557 total_time 721.0\n",
      "episode 12349, initial_state (1, 9, 6), reward 1572.0, memory_length 2000, epsilon 0.0038602466654371103 total_time 726.0\n",
      "episode 12359, initial_state (4, 14, 1), reward 1338.0, memory_length 2000, epsilon 0.003842915030867551 total_time 722.0\n",
      "episode 12369, initial_state (3, 13, 4), reward 1117.0, memory_length 2000, epsilon 0.0038256612134336797 total_time 724.0\n",
      "episode 12379, initial_state (1, 19, 5), reward 1584.0, memory_length 2000, epsilon 0.003808484863745107 total_time 726.0\n",
      "episode 12389, initial_state (0, 8, 6), reward 1189.0, memory_length 2000, epsilon 0.0037913856339801652 total_time 725.0\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "episode 12399, initial_state (0, 12, 4), reward 1499.0, memory_length 2000, epsilon 0.0037743631778788668 total_time 723.0\n",
      "episode 12409, initial_state (0, 1, 4), reward 1734.0, memory_length 2000, epsilon 0.003757417150735897 total_time 728.0\n",
      "episode 12419, initial_state (2, 23, 1), reward 1617.0, memory_length 2000, epsilon 0.0037405472093936217 total_time 725.0\n",
      "episode 12429, initial_state (0, 8, 3), reward 1383.0, memory_length 2000, epsilon 0.0037237530122351545 total_time 721.0\n",
      "episode 12439, initial_state (1, 2, 6), reward 1279.0, memory_length 2000, epsilon 0.003707034219177429 total_time 724.0\n",
      "episode 12449, initial_state (4, 16, 1), reward 1486.0, memory_length 2000, epsilon 0.003690390491664315 total_time 722.0\n",
      "episode 12459, initial_state (3, 3, 6), reward 1314.0, memory_length 2000, epsilon 0.0036738214926597653 total_time 723.0\n",
      "episode 12469, initial_state (4, 18, 3), reward 1026.0, memory_length 2000, epsilon 0.003657326886640976 total_time 726.0\n",
      "episode 12479, initial_state (3, 20, 4), reward 1420.0, memory_length 2000, epsilon 0.003640906339591616 total_time 723.0\n",
      "episode 12489, initial_state (4, 13, 3), reward 1350.0, memory_length 2000, epsilon 0.0036245595189950456 total_time 724.0\n",
      "episode 12499, initial_state (2, 20, 6), reward 1644.0, memory_length 2000, epsilon 0.00360828609382759 total_time 724.0\n",
      "episode 12509, initial_state (2, 6, 6), reward 1683.0, memory_length 2000, epsilon 0.0035920857345518357 total_time 729.0\n",
      "episode 12519, initial_state (3, 17, 5), reward 1148.0, memory_length 2000, epsilon 0.0035759581131099482 total_time 728.0\n",
      "episode 12529, initial_state (2, 3, 0), reward 985.0, memory_length 2000, epsilon 0.003559902902917045 total_time 725.0\n",
      "episode 12539, initial_state (4, 20, 4), reward 1268.0, memory_length 2000, epsilon 0.0035439197788545707 total_time 721.0\n",
      "episode 12549, initial_state (0, 7, 4), reward 1603.0, memory_length 2000, epsilon 0.003528008417263718 total_time 722.0\n",
      "episode 12559, initial_state (0, 17, 6), reward 1373.0, memory_length 2000, epsilon 0.003512168495938873 total_time 722.0\n",
      "episode 12569, initial_state (0, 0, 5), reward 1108.0, memory_length 2000, epsilon 0.0034963996941210812 total_time 726.0\n",
      "episode 12579, initial_state (3, 6, 0), reward 1266.0, memory_length 2000, epsilon 0.003480701692491571 total_time 724.0\n",
      "episode 12589, initial_state (1, 14, 3), reward 1286.0, memory_length 2000, epsilon 0.0034650741731652728 total_time 721.0\n",
      "episode 12599, initial_state (1, 4, 2), reward 1298.0, memory_length 2000, epsilon 0.0034495168196843856 total_time 723.0\n",
      "episode 12609, initial_state (4, 4, 0), reward 1302.0, memory_length 2000, epsilon 0.00343402931701197 total_time 722.0\n",
      "episode 12619, initial_state (4, 2, 4), reward 1311.0, memory_length 2000, epsilon 0.0034186113515255706 total_time 721.0\n",
      "episode 12629, initial_state (4, 16, 1), reward 1435.0, memory_length 2000, epsilon 0.003403262611010854 total_time 722.0\n",
      "episode 12639, initial_state (4, 23, 3), reward 1612.0, memory_length 2000, epsilon 0.0033879827846553033 total_time 726.0\n",
      "episode 12649, initial_state (2, 0, 3), reward 1630.0, memory_length 2000, epsilon 0.0033727715630419114 total_time 730.0\n",
      "episode 12659, initial_state (1, 18, 5), reward 1594.0, memory_length 2000, epsilon 0.0033576286381429223 total_time 725.0\n",
      "episode 12669, initial_state (3, 5, 0), reward 1378.0, memory_length 2000, epsilon 0.0033425537033135917 total_time 722.0\n",
      "episode 12679, initial_state (2, 14, 6), reward 1573.0, memory_length 2000, epsilon 0.0033275464532859676 total_time 721.0\n",
      "episode 12689, initial_state (3, 5, 4), reward 1486.0, memory_length 2000, epsilon 0.0033126065841627282 total_time 729.0\n",
      "episode 12699, initial_state (2, 4, 0), reward 1362.0, memory_length 2000, epsilon 0.0032977337934110124 total_time 721.0\n",
      "episode 12709, initial_state (2, 4, 3), reward 1143.0, memory_length 2000, epsilon 0.0032829277798563 total_time 724.0\n",
      "episode 12719, initial_state (4, 14, 0), reward 1223.0, memory_length 2000, epsilon 0.003268188243676313 total_time 724.0\n",
      "episode 12729, initial_state (4, 14, 1), reward 1280.0, memory_length 2000, epsilon 0.0032535148863949343 total_time 726.0\n",
      "episode 12739, initial_state (0, 21, 6), reward 1499.0, memory_length 2000, epsilon 0.0032389074108761804 total_time 726.0\n",
      "episode 12749, initial_state (4, 14, 4), reward 1440.0, memory_length 2000, epsilon 0.003224365521318173 total_time 721.0\n",
      "episode 12759, initial_state (4, 3, 6), reward 1395.0, memory_length 2000, epsilon 0.003209888923247151 total_time 726.0\n",
      "episode 12769, initial_state (3, 22, 5), reward 1600.0, memory_length 2000, epsilon 0.0031954773235115127 total_time 724.0\n",
      "episode 12779, initial_state (4, 3, 4), reward 1586.0, memory_length 2000, epsilon 0.0031811304302758647 total_time 721.0\n",
      "episode 12789, initial_state (0, 16, 1), reward 1458.0, memory_length 2000, epsilon 0.0031668479530151316 total_time 722.0\n",
      "episode 12799, initial_state (4, 14, 6), reward 1186.0, memory_length 2000, epsilon 0.003152629602508661 total_time 735.0\n",
      "episode 12809, initial_state (0, 22, 2), reward 1406.0, memory_length 2000, epsilon 0.0031384750908343695 total_time 726.0\n",
      "episode 12819, initial_state (2, 8, 2), reward 1451.0, memory_length 2000, epsilon 0.0031243841313629136 total_time 723.0\n",
      "episode 12829, initial_state (1, 5, 2), reward 1175.0, memory_length 2000, epsilon 0.003110356438751878 total_time 729.0\n",
      "episode 12839, initial_state (4, 1, 1), reward 1460.0, memory_length 2000, epsilon 0.0030963917289400108 total_time 722.0\n",
      "episode 12849, initial_state (2, 12, 5), reward 1176.0, memory_length 2000, epsilon 0.0030824897191414607 total_time 726.0\n",
      "episode 12859, initial_state (0, 5, 5), reward 1143.0, memory_length 2000, epsilon 0.003068650127840054 total_time 721.0\n",
      "episode 12869, initial_state (4, 18, 4), reward 1355.0, memory_length 2000, epsilon 0.003054872674783594 total_time 721.0\n",
      "episode 12879, initial_state (2, 16, 0), reward 1459.0, memory_length 2000, epsilon 0.0030411570809781895 total_time 722.0\n",
      "episode 12889, initial_state (0, 11, 6), reward 1261.0, memory_length 2000, epsilon 0.00302750306868259 total_time 729.0\n",
      "episode 12899, initial_state (4, 4, 3), reward 882.0, memory_length 2000, epsilon 0.003013910361402584 total_time 725.0\n",
      "episode 12909, initial_state (1, 14, 5), reward 1280.0, memory_length 2000, epsilon 0.0030003786838853834 total_time 726.0\n",
      "episode 12919, initial_state (3, 14, 1), reward 1322.0, memory_length 2000, epsilon 0.002986907762114057 total_time 723.0\n",
      "episode 12929, initial_state (2, 16, 6), reward 1261.0, memory_length 2000, epsilon 0.002973497323301981 total_time 725.0\n",
      "episode 12939, initial_state (3, 21, 1), reward 1180.0, memory_length 2000, epsilon 0.0029601470958873057 total_time 724.0\n",
      "episode 12949, initial_state (3, 17, 0), reward 1206.0, memory_length 2000, epsilon 0.0029468568095274723 total_time 726.0\n",
      "episode 12959, initial_state (3, 11, 3), reward 1550.0, memory_length 2000, epsilon 0.002933626195093729 total_time 724.0\n",
      "episode 12969, initial_state (1, 23, 5), reward 1310.0, memory_length 2000, epsilon 0.0029204549846656796 total_time 725.0\n",
      "episode 12979, initial_state (2, 17, 1), reward 1076.0, memory_length 2000, epsilon 0.0029073429115258668 total_time 726.0\n",
      "episode 12989, initial_state (0, 6, 0), reward 1419.0, memory_length 2000, epsilon 0.002894289710154356 total_time 722.0\n",
      "episode 12999, initial_state (2, 2, 3), reward 1314.0, memory_length 2000, epsilon 0.002881295116223375 total_time 729.0\n",
      "Saving Model 13000\n",
      "Saving Model 13000 Weights\n",
      "episode 13009, initial_state (2, 5, 3), reward 1392.0, memory_length 2000, epsilon 0.002868358866591954 total_time 732.0\n",
      "episode 13019, initial_state (3, 22, 6), reward 1328.0, memory_length 2000, epsilon 0.002855480699300595 total_time 732.0\n",
      "episode 13029, initial_state (3, 20, 2), reward 1452.0, memory_length 2000, epsilon 0.0028426603535659737 total_time 721.0\n",
      "episode 13039, initial_state (1, 2, 4), reward 1800.0, memory_length 2000, epsilon 0.0028298975697756447 total_time 727.0\n",
      "episode 13049, initial_state (2, 9, 2), reward 1167.0, memory_length 2000, epsilon 0.0028171920894828036 total_time 722.0\n",
      "episode 13059, initial_state (4, 18, 4), reward 1332.0, memory_length 2000, epsilon 0.00280454365540104 total_time 728.0\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "episode 13069, initial_state (3, 22, 5), reward 1474.0, memory_length 2000, epsilon 0.002791952011399131 total_time 722.0\n",
      "episode 13079, initial_state (3, 21, 0), reward 1364.0, memory_length 2000, epsilon 0.0027794169024958557 total_time 732.0\n",
      "episode 13089, initial_state (2, 22, 6), reward 1414.0, memory_length 2000, epsilon 0.0027669380748548333 total_time 729.0\n",
      "episode 13099, initial_state (1, 16, 6), reward 1214.0, memory_length 2000, epsilon 0.0027545152757793723 total_time 722.0\n",
      "episode 13109, initial_state (4, 13, 6), reward 1277.0, memory_length 2000, epsilon 0.002742148253707369 total_time 722.0\n",
      "episode 13119, initial_state (3, 18, 0), reward 1413.0, memory_length 2000, epsilon 0.0027298367582062047 total_time 730.0\n",
      "episode 13129, initial_state (1, 8, 6), reward 1144.0, memory_length 2000, epsilon 0.0027175805399676746 total_time 723.0\n",
      "episode 13139, initial_state (3, 11, 1), reward 1134.0, memory_length 2000, epsilon 0.002705379350802942 total_time 725.0\n",
      "episode 13149, initial_state (1, 8, 4), reward 1358.0, memory_length 2000, epsilon 0.002693232943637506 total_time 726.0\n",
      "episode 13159, initial_state (2, 2, 1), reward 1143.0, memory_length 2000, epsilon 0.002681141072506208 total_time 721.0\n",
      "episode 13169, initial_state (3, 0, 2), reward 1491.0, memory_length 2000, epsilon 0.0026691034925482443 total_time 725.0\n",
      "episode 13179, initial_state (1, 12, 2), reward 1575.0, memory_length 2000, epsilon 0.0026571199600022096 total_time 729.0\n",
      "episode 13189, initial_state (1, 6, 1), reward 1523.0, memory_length 2000, epsilon 0.0026451902322011628 total_time 723.0\n",
      "episode 13199, initial_state (0, 12, 2), reward 996.0, memory_length 2000, epsilon 0.0026333140675677035 total_time 723.0\n",
      "episode 13209, initial_state (3, 14, 0), reward 1420.0, memory_length 2000, epsilon 0.0026214912256090944 total_time 726.0\n",
      "episode 13219, initial_state (1, 17, 2), reward 1656.0, memory_length 2000, epsilon 0.0026097214669123818 total_time 724.0\n",
      "episode 13229, initial_state (3, 6, 5), reward 1492.0, memory_length 2000, epsilon 0.0025980045531395495 total_time 722.0\n",
      "episode 13239, initial_state (2, 15, 3), reward 1621.0, memory_length 2000, epsilon 0.0025863402470226965 total_time 722.0\n",
      "episode 13249, initial_state (1, 21, 6), reward 1627.0, memory_length 2000, epsilon 0.0025747283123592197 total_time 722.0\n",
      "episode 13259, initial_state (1, 6, 5), reward 1131.0, memory_length 2000, epsilon 0.0025631685140070485 total_time 728.0\n",
      "episode 13269, initial_state (1, 16, 6), reward 1485.0, memory_length 2000, epsilon 0.0025516606178798704 total_time 727.0\n",
      "episode 13279, initial_state (2, 18, 0), reward 1554.0, memory_length 2000, epsilon 0.0025402043909423964 total_time 723.0\n",
      "episode 13289, initial_state (3, 12, 3), reward 1022.0, memory_length 2000, epsilon 0.0025287996012056412 total_time 727.0\n",
      "episode 13299, initial_state (4, 14, 1), reward 1706.0, memory_length 2000, epsilon 0.002517446017722219 total_time 724.0\n",
      "episode 13309, initial_state (0, 8, 1), reward 1627.0, memory_length 2000, epsilon 0.002506143410581678 total_time 732.0\n",
      "episode 13319, initial_state (2, 12, 4), reward 1729.0, memory_length 2000, epsilon 0.0024948915509058374 total_time 725.0\n",
      "episode 13329, initial_state (4, 0, 1), reward 1360.0, memory_length 2000, epsilon 0.002483690210844154 total_time 726.0\n",
      "episode 13339, initial_state (3, 22, 3), reward 1232.0, memory_length 2000, epsilon 0.00247253916356911 total_time 721.0\n",
      "episode 13349, initial_state (0, 19, 2), reward 1315.0, memory_length 2000, epsilon 0.002461438183271618 total_time 728.0\n",
      "episode 13359, initial_state (3, 19, 0), reward 1260.0, memory_length 2000, epsilon 0.0024503870451564437 total_time 730.0\n",
      "episode 13369, initial_state (2, 21, 5), reward 1575.0, memory_length 2000, epsilon 0.002439385525437665 total_time 726.0\n",
      "episode 13379, initial_state (3, 9, 3), reward 1241.0, memory_length 2000, epsilon 0.002428433401334131 total_time 722.0\n",
      "episode 13389, initial_state (4, 1, 3), reward 1041.0, memory_length 2000, epsilon 0.002417530451064955 total_time 723.0\n",
      "episode 13399, initial_state (2, 21, 5), reward 1307.0, memory_length 2000, epsilon 0.0024066764538450236 total_time 722.0\n",
      "episode 13409, initial_state (3, 22, 2), reward 1446.0, memory_length 2000, epsilon 0.0023958711898805175 total_time 724.0\n",
      "episode 13419, initial_state (4, 13, 6), reward 1121.0, memory_length 2000, epsilon 0.0023851144403644747 total_time 726.0\n",
      "episode 13429, initial_state (3, 2, 4), reward 1446.0, memory_length 2000, epsilon 0.002374405987472349 total_time 722.0\n",
      "episode 13439, initial_state (2, 13, 1), reward 1403.0, memory_length 2000, epsilon 0.002363745614357605 total_time 723.0\n",
      "episode 13449, initial_state (1, 22, 6), reward 1116.0, memory_length 2000, epsilon 0.002353133105147324 total_time 724.0\n",
      "episode 13459, initial_state (1, 8, 1), reward 1510.0, memory_length 2000, epsilon 0.0023425682449378276 total_time 722.0\n",
      "episode 13469, initial_state (2, 14, 5), reward 1360.0, memory_length 2000, epsilon 0.0023320508197903376 total_time 729.0\n",
      "episode 13479, initial_state (2, 14, 1), reward 1593.0, memory_length 2000, epsilon 0.002321580616726636 total_time 726.0\n",
      "episode 13489, initial_state (1, 19, 6), reward 1323.0, memory_length 2000, epsilon 0.002311157423724752 total_time 724.0\n",
      "episode 13499, initial_state (3, 9, 4), reward 1395.0, memory_length 2000, epsilon 0.002300781029714674 total_time 734.0\n",
      "episode 13509, initial_state (0, 6, 5), reward 1060.0, memory_length 2000, epsilon 0.002290451224574064 total_time 727.0\n",
      "episode 13519, initial_state (0, 8, 1), reward 1346.0, memory_length 2000, epsilon 0.002280167799124018 total_time 730.0\n",
      "episode 13529, initial_state (2, 5, 6), reward 1483.0, memory_length 2000, epsilon 0.0022699305451248175 total_time 721.0\n",
      "episode 13539, initial_state (1, 19, 0), reward 1527.0, memory_length 2000, epsilon 0.002259739255271721 total_time 721.0\n",
      "episode 13549, initial_state (2, 0, 0), reward 1314.0, memory_length 2000, epsilon 0.002249593723190762 total_time 723.0\n",
      "episode 13559, initial_state (3, 20, 4), reward 1348.0, memory_length 2000, epsilon 0.0022394937434345643 total_time 723.0\n",
      "episode 13569, initial_state (2, 13, 6), reward 1259.0, memory_length 2000, epsilon 0.0022294391114781958 total_time 722.0\n",
      "episode 13579, initial_state (0, 9, 3), reward 1204.0, memory_length 2000, epsilon 0.0022194296237150156 total_time 726.0\n",
      "episode 13589, initial_state (4, 3, 2), reward 1283.0, memory_length 2000, epsilon 0.0022094650774525542 total_time 729.0\n",
      "episode 13599, initial_state (0, 18, 6), reward 1249.0, memory_length 2000, epsilon 0.0021995452709084097 total_time 722.0\n",
      "episode 13609, initial_state (0, 15, 4), reward 1350.0, memory_length 2000, epsilon 0.0021896700032061622 total_time 728.0\n",
      "episode 13619, initial_state (4, 5, 1), reward 1540.0, memory_length 2000, epsilon 0.0021798390743712993 total_time 727.0\n",
      "episode 13629, initial_state (3, 9, 0), reward 1297.0, memory_length 2000, epsilon 0.002170052285327178 total_time 725.0\n",
      "episode 13639, initial_state (4, 15, 6), reward 1341.0, memory_length 2000, epsilon 0.0021603094378909865 total_time 734.0\n",
      "episode 13649, initial_state (4, 18, 4), reward 1217.0, memory_length 2000, epsilon 0.002150610334769731 total_time 724.0\n",
      "episode 13659, initial_state (1, 4, 5), reward 1432.0, memory_length 2000, epsilon 0.0021409547795562425 total_time 723.0\n",
      "episode 13669, initial_state (2, 18, 4), reward 1515.0, memory_length 2000, epsilon 0.002131342576725196 total_time 722.0\n",
      "episode 13679, initial_state (1, 23, 2), reward 1353.0, memory_length 2000, epsilon 0.0021217735316291565 total_time 722.0\n",
      "episode 13689, initial_state (3, 3, 5), reward 1601.0, memory_length 2000, epsilon 0.002112247450494634 total_time 721.0\n",
      "episode 13699, initial_state (2, 2, 4), reward 1292.0, memory_length 2000, epsilon 0.00210276414041816 total_time 728.0\n",
      "episode 13709, initial_state (3, 3, 2), reward 1371.0, memory_length 2000, epsilon 0.0020933234093623836 total_time 723.0\n",
      "episode 13719, initial_state (4, 4, 1), reward 1281.0, memory_length 2000, epsilon 0.0020839250661521744 total_time 725.0\n",
      "episode 13729, initial_state (2, 9, 4), reward 1528.0, memory_length 2000, epsilon 0.002074568920470763 total_time 724.0\n",
      "episode 13739, initial_state (2, 22, 1), reward 1140.0, memory_length 2000, epsilon 0.00206525478285588 total_time 725.0\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "episode 13749, initial_state (4, 9, 0), reward 1765.0, memory_length 2000, epsilon 0.0020559824646959194 total_time 731.0\n",
      "episode 13759, initial_state (1, 15, 4), reward 1412.0, memory_length 2000, epsilon 0.0020467517782261245 total_time 721.0\n",
      "episode 13769, initial_state (4, 12, 5), reward 1395.0, memory_length 2000, epsilon 0.002037562536524775 total_time 724.0\n",
      "episode 13779, initial_state (1, 9, 3), reward 1341.0, memory_length 2000, epsilon 0.0020284145535094133 total_time 722.0\n",
      "episode 13789, initial_state (1, 13, 1), reward 1655.0, memory_length 2000, epsilon 0.0020193076439330724 total_time 725.0\n",
      "episode 13799, initial_state (3, 19, 5), reward 1702.0, memory_length 2000, epsilon 0.002010241623380521 total_time 722.0\n",
      "episode 13809, initial_state (4, 12, 0), reward 1429.0, memory_length 2000, epsilon 0.002001216308264535 total_time 726.0\n",
      "episode 13819, initial_state (0, 23, 6), reward 1148.0, memory_length 2000, epsilon 0.0019922315158221716 total_time 721.0\n",
      "episode 13829, initial_state (4, 4, 6), reward 1302.0, memory_length 2000, epsilon 0.001983287064111079 total_time 732.0\n",
      "episode 13839, initial_state (1, 13, 5), reward 1269.0, memory_length 2000, epsilon 0.0019743827720058033 total_time 729.0\n",
      "episode 13849, initial_state (4, 21, 4), reward 1317.0, memory_length 2000, epsilon 0.0019655184591941257 total_time 722.0\n",
      "episode 13859, initial_state (3, 5, 5), reward 1511.0, memory_length 2000, epsilon 0.0019566939461734094 total_time 727.0\n",
      "episode 13869, initial_state (2, 5, 3), reward 1321.0, memory_length 2000, epsilon 0.0019479090542469653 total_time 726.0\n",
      "episode 13879, initial_state (4, 2, 3), reward 1480.0, memory_length 2000, epsilon 0.001939163605520428 total_time 722.0\n",
      "episode 13889, initial_state (3, 23, 4), reward 1270.0, memory_length 2000, epsilon 0.0019304574228981642 total_time 726.0\n",
      "episode 13899, initial_state (0, 2, 4), reward 1770.0, memory_length 2000, epsilon 0.0019217903300796778 total_time 727.0\n",
      "episode 13909, initial_state (2, 6, 6), reward 1523.0, memory_length 2000, epsilon 0.0019131621515560435 total_time 722.0\n",
      "episode 13919, initial_state (0, 8, 4), reward 914.0, memory_length 2000, epsilon 0.0019045727126063528 total_time 725.0\n",
      "episode 13929, initial_state (2, 14, 4), reward 1746.0, memory_length 2000, epsilon 0.00189602183929417 total_time 728.0\n",
      "episode 13939, initial_state (0, 12, 6), reward 1326.0, memory_length 2000, epsilon 0.0018875093584640203 total_time 721.0\n",
      "episode 13949, initial_state (2, 4, 6), reward 1644.0, memory_length 2000, epsilon 0.001879035097737876 total_time 723.0\n",
      "episode 13959, initial_state (2, 12, 2), reward 1513.0, memory_length 2000, epsilon 0.0018705988855116675 total_time 726.0\n",
      "episode 13969, initial_state (3, 18, 5), reward 1431.0, memory_length 2000, epsilon 0.001862200550951811 total_time 726.0\n",
      "episode 13979, initial_state (1, 1, 3), reward 1522.0, memory_length 2000, epsilon 0.0018538399239917407 total_time 733.0\n",
      "episode 13989, initial_state (4, 1, 1), reward 1298.0, memory_length 2000, epsilon 0.0018455168353284773 total_time 723.0\n",
      "episode 13999, initial_state (0, 16, 1), reward 1298.0, memory_length 2000, epsilon 0.0018372311164191906 total_time 722.0\n",
      "Saving Model 14000\n",
      "Saving Model 14000 Weights\n",
      "episode 14009, initial_state (1, 19, 1), reward 1401.0, memory_length 2000, epsilon 0.00182898259947779 total_time 721.0\n",
      "episode 14019, initial_state (3, 3, 4), reward 1634.0, memory_length 2000, epsilon 0.0018207711174715263 total_time 726.0\n",
      "episode 14029, initial_state (4, 6, 6), reward 1431.0, memory_length 2000, epsilon 0.0018125965041176058 total_time 729.0\n",
      "episode 14039, initial_state (3, 3, 6), reward 1613.0, memory_length 2000, epsilon 0.0018044585938798302 total_time 721.0\n",
      "episode 14049, initial_state (0, 17, 3), reward 1566.0, memory_length 2000, epsilon 0.001796357221965239 total_time 722.0\n",
      "episode 14059, initial_state (0, 2, 2), reward 1379.0, memory_length 2000, epsilon 0.0017882922243207741 total_time 724.0\n",
      "episode 14069, initial_state (2, 23, 0), reward 1593.0, memory_length 2000, epsilon 0.0017802634376299592 total_time 721.0\n",
      "episode 14079, initial_state (0, 23, 4), reward 1271.0, memory_length 2000, epsilon 0.0017722706993095862 total_time 724.0\n",
      "episode 14089, initial_state (0, 18, 4), reward 1334.0, memory_length 2000, epsilon 0.0017643138475064329 total_time 724.0\n",
      "episode 14099, initial_state (3, 0, 5), reward 1361.0, memory_length 2000, epsilon 0.001756392721093978 total_time 726.0\n",
      "episode 14109, initial_state (0, 9, 0), reward 1613.0, memory_length 2000, epsilon 0.001748507159669141 total_time 725.0\n",
      "episode 14119, initial_state (2, 9, 6), reward 1505.0, memory_length 2000, epsilon 0.0017406570035490337 total_time 722.0\n",
      "episode 14129, initial_state (4, 3, 5), reward 1584.0, memory_length 2000, epsilon 0.0017328420937677282 total_time 721.0\n",
      "episode 14139, initial_state (2, 19, 3), reward 1657.0, memory_length 2000, epsilon 0.0017250622720730309 total_time 723.0\n",
      "episode 14149, initial_state (1, 22, 5), reward 1386.0, memory_length 2000, epsilon 0.0017173173809232882 total_time 732.0\n",
      "episode 14159, initial_state (1, 20, 3), reward 1412.0, memory_length 2000, epsilon 0.00170960726348419 total_time 726.0\n",
      "episode 14169, initial_state (0, 12, 5), reward 1316.0, memory_length 2000, epsilon 0.0017019317636255942 total_time 724.0\n",
      "episode 14179, initial_state (0, 15, 0), reward 1330.0, memory_length 2000, epsilon 0.0016942907259183684 total_time 722.0\n",
      "episode 14189, initial_state (1, 9, 2), reward 1812.0, memory_length 2000, epsilon 0.0016866839956312344 total_time 721.0\n",
      "episode 14199, initial_state (4, 8, 0), reward 1315.0, memory_length 2000, epsilon 0.001679111418727646 total_time 727.0\n",
      "episode 14209, initial_state (4, 15, 6), reward 1350.0, memory_length 2000, epsilon 0.0016715728418626617 total_time 734.0\n",
      "episode 14219, initial_state (3, 22, 5), reward 1571.0, memory_length 2000, epsilon 0.0016640681123798425 total_time 721.0\n",
      "episode 14229, initial_state (3, 17, 0), reward 1319.0, memory_length 2000, epsilon 0.0016565970783081613 total_time 721.0\n",
      "episode 14239, initial_state (4, 23, 3), reward 1581.0, memory_length 2000, epsilon 0.0016491595883589203 total_time 725.0\n",
      "episode 14249, initial_state (4, 10, 4), reward 1205.0, memory_length 2000, epsilon 0.0016417554919226948 total_time 725.0\n",
      "episode 14259, initial_state (1, 2, 1), reward 1358.0, memory_length 2000, epsilon 0.0016343846390662793 total_time 722.0\n",
      "episode 14269, initial_state (2, 9, 3), reward 1505.0, memory_length 2000, epsilon 0.0016270468805296514 total_time 727.0\n",
      "episode 14279, initial_state (3, 15, 0), reward 1537.0, memory_length 2000, epsilon 0.0016197420677229518 total_time 724.0\n",
      "episode 14289, initial_state (1, 22, 2), reward 1173.0, memory_length 2000, epsilon 0.0016124700527234681 total_time 722.0\n",
      "episode 14299, initial_state (3, 15, 0), reward 1333.0, memory_length 2000, epsilon 0.00160523068827265 total_time 724.0\n",
      "episode 14309, initial_state (1, 2, 0), reward 1484.0, memory_length 2000, epsilon 0.0015980238277731199 total_time 727.0\n",
      "episode 14319, initial_state (0, 15, 0), reward 1270.0, memory_length 2000, epsilon 0.001590849325285706 total_time 722.0\n",
      "episode 14329, initial_state (4, 15, 1), reward 1377.0, memory_length 2000, epsilon 0.0015837070355264897 total_time 721.0\n",
      "episode 14339, initial_state (0, 17, 2), reward 1419.0, memory_length 2000, epsilon 0.001576596813863856 total_time 722.0\n",
      "episode 14349, initial_state (2, 20, 5), reward 1182.0, memory_length 2000, epsilon 0.0015695185163155751 total_time 728.0\n",
      "episode 14359, initial_state (1, 4, 1), reward 1464.0, memory_length 2000, epsilon 0.0015624719995458798 total_time 726.0\n",
      "episode 14369, initial_state (4, 10, 3), reward 1587.0, memory_length 2000, epsilon 0.0015554571208625642 total_time 723.0\n",
      "episode 14379, initial_state (2, 16, 3), reward 1446.0, memory_length 2000, epsilon 0.0015484737382140956 total_time 727.0\n",
      "episode 14389, initial_state (4, 20, 1), reward 1328.0, memory_length 2000, epsilon 0.001541521710186738 total_time 728.0\n",
      "episode 14399, initial_state (2, 12, 3), reward 1234.0, memory_length 2000, epsilon 0.001534600896001684 total_time 729.0\n",
      "episode 14409, initial_state (0, 5, 3), reward 1489.0, memory_length 2000, epsilon 0.0015277111555122107 total_time 722.0\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "episode 14419, initial_state (1, 6, 4), reward 1495.0, memory_length 2000, epsilon 0.0015208523492008378 total_time 727.0\n",
      "episode 14429, initial_state (0, 0, 1), reward 1185.0, memory_length 2000, epsilon 0.0015140243381765032 total_time 726.0\n",
      "episode 14439, initial_state (0, 18, 0), reward 1278.0, memory_length 2000, epsilon 0.0015072269841717519 total_time 721.0\n",
      "episode 14449, initial_state (0, 2, 0), reward 1550.0, memory_length 2000, epsilon 0.0015004601495399298 total_time 725.0\n",
      "episode 14459, initial_state (3, 23, 5), reward 1278.0, memory_length 2000, epsilon 0.0014937236972524062 total_time 725.0\n",
      "episode 14469, initial_state (2, 23, 3), reward 1477.0, memory_length 2000, epsilon 0.0014870174908957919 total_time 726.0\n",
      "episode 14479, initial_state (3, 17, 3), reward 1613.0, memory_length 2000, epsilon 0.001480341394669179 total_time 726.0\n",
      "episode 14489, initial_state (3, 10, 5), reward 1491.0, memory_length 2000, epsilon 0.001473695273381392 total_time 726.0\n",
      "episode 14499, initial_state (2, 11, 6), reward 1484.0, memory_length 2000, epsilon 0.0014670789924482457 total_time 722.0\n",
      "episode 14509, initial_state (0, 7, 2), reward 1602.0, memory_length 2000, epsilon 0.0014604924178898257 total_time 728.0\n",
      "episode 14519, initial_state (4, 20, 0), reward 1441.0, memory_length 2000, epsilon 0.0014539354163277721 total_time 722.0\n",
      "episode 14529, initial_state (1, 11, 1), reward 1556.0, memory_length 2000, epsilon 0.0014474078549825797 total_time 726.0\n",
      "episode 14539, initial_state (0, 10, 4), reward 1229.0, memory_length 2000, epsilon 0.0014409096016709088 total_time 721.0\n",
      "episode 14549, initial_state (0, 19, 3), reward 1170.0, memory_length 2000, epsilon 0.0014344405248029064 total_time 725.0\n",
      "episode 14559, initial_state (2, 9, 5), reward 1255.0, memory_length 2000, epsilon 0.0014280004933795447 total_time 724.0\n",
      "episode 14569, initial_state (2, 11, 4), reward 1506.0, memory_length 2000, epsilon 0.0014215893769899683 total_time 721.0\n",
      "episode 14579, initial_state (3, 3, 3), reward 1351.0, memory_length 2000, epsilon 0.001415207045808851 total_time 721.0\n",
      "episode 14589, initial_state (0, 14, 3), reward 1351.0, memory_length 2000, epsilon 0.0014088533705937695 total_time 724.0\n",
      "episode 14599, initial_state (2, 16, 4), reward 1526.0, memory_length 2000, epsilon 0.001402528222682581 total_time 721.0\n",
      "episode 14609, initial_state (0, 3, 5), reward 1459.0, memory_length 2000, epsilon 0.0013962314739908253 total_time 721.0\n",
      "episode 14619, initial_state (1, 5, 6), reward 1697.0, memory_length 2000, epsilon 0.0013899629970091265 total_time 729.0\n",
      "episode 14629, initial_state (1, 9, 6), reward 1387.0, memory_length 2000, epsilon 0.0013837226648006115 total_time 727.0\n",
      "episode 14639, initial_state (3, 8, 0), reward 1277.0, memory_length 2000, epsilon 0.0013775103509983395 total_time 729.0\n",
      "episode 14649, initial_state (0, 10, 4), reward 1845.0, memory_length 2000, epsilon 0.0013713259298027453 total_time 726.0\n",
      "episode 14659, initial_state (2, 0, 4), reward 1365.0, memory_length 2000, epsilon 0.001365169275979086 total_time 725.0\n",
      "episode 14669, initial_state (3, 17, 1), reward 1301.0, memory_length 2000, epsilon 0.001359040264854912 total_time 725.0\n",
      "episode 14679, initial_state (0, 3, 1), reward 1247.0, memory_length 2000, epsilon 0.0013529387723175388 total_time 726.0\n",
      "episode 14689, initial_state (3, 4, 5), reward 1338.0, memory_length 2000, epsilon 0.0013468646748115343 total_time 723.0\n",
      "episode 14699, initial_state (4, 15, 2), reward 1459.0, memory_length 2000, epsilon 0.0013408178493362172 total_time 727.0\n",
      "episode 14709, initial_state (3, 21, 6), reward 1558.0, memory_length 2000, epsilon 0.0013347981734431631 total_time 724.0\n",
      "episode 14719, initial_state (3, 20, 0), reward 1572.0, memory_length 2000, epsilon 0.0013288055252337304 total_time 722.0\n",
      "episode 14729, initial_state (2, 16, 5), reward 1522.0, memory_length 2000, epsilon 0.001322839783356588 total_time 725.0\n",
      "episode 14739, initial_state (1, 3, 6), reward 1472.0, memory_length 2000, epsilon 0.001316900827005259 total_time 734.0\n",
      "episode 14749, initial_state (2, 12, 4), reward 1212.0, memory_length 2000, epsilon 0.001310988535915676 total_time 723.0\n",
      "episode 14759, initial_state (4, 12, 5), reward 1461.0, memory_length 2000, epsilon 0.0013051027903637391 total_time 723.0\n",
      "episode 14769, initial_state (3, 22, 3), reward 1241.0, memory_length 2000, epsilon 0.0012992434711629017 total_time 722.0\n",
      "episode 14779, initial_state (1, 22, 4), reward 1406.0, memory_length 2000, epsilon 0.0012934104596617497 total_time 724.0\n",
      "episode 14789, initial_state (0, 6, 2), reward 1240.0, memory_length 2000, epsilon 0.0012876036377416004 total_time 721.0\n",
      "episode 14799, initial_state (0, 2, 3), reward 1329.0, memory_length 2000, epsilon 0.0012818228878141134 total_time 727.0\n",
      "episode 14809, initial_state (2, 5, 5), reward 1296.0, memory_length 2000, epsilon 0.001276068092818902 total_time 725.0\n",
      "episode 14819, initial_state (1, 23, 4), reward 1325.0, memory_length 2000, epsilon 0.0012703391362211727 total_time 724.0\n",
      "episode 14829, initial_state (1, 19, 4), reward 1661.0, memory_length 2000, epsilon 0.0012646359020093583 total_time 727.0\n",
      "episode 14839, initial_state (0, 18, 4), reward 1694.0, memory_length 2000, epsilon 0.0012589582746927713 total_time 726.0\n",
      "episode 14849, initial_state (2, 9, 3), reward 1476.0, memory_length 2000, epsilon 0.001253306139299265 total_time 722.0\n",
      "episode 14859, initial_state (4, 8, 1), reward 1292.0, memory_length 2000, epsilon 0.0012476793813729035 total_time 730.0\n",
      "episode 14869, initial_state (4, 10, 6), reward 1203.0, memory_length 2000, epsilon 0.0012420778869716466 total_time 728.0\n",
      "episode 14879, initial_state (2, 15, 2), reward 1492.0, memory_length 2000, epsilon 0.0012365015426650417 total_time 725.0\n",
      "episode 14889, initial_state (3, 6, 1), reward 1684.0, memory_length 2000, epsilon 0.001230950235531926 total_time 723.0\n",
      "episode 14899, initial_state (1, 6, 4), reward 1623.0, memory_length 2000, epsilon 0.0012254238531581403 total_time 729.0\n",
      "episode 14909, initial_state (2, 19, 6), reward 1468.0, memory_length 2000, epsilon 0.0012199222836342537 total_time 721.0\n",
      "episode 14919, initial_state (1, 12, 1), reward 1153.0, memory_length 2000, epsilon 0.0012144454155532931 total_time 726.0\n",
      "episode 14929, initial_state (4, 23, 0), reward 1258.0, memory_length 2000, epsilon 0.0012089931380084942 total_time 724.0\n",
      "episode 14939, initial_state (0, 8, 4), reward 1268.0, memory_length 2000, epsilon 0.00120356534059105 total_time 726.0\n",
      "episode 14949, initial_state (2, 11, 4), reward 1490.0, memory_length 2000, epsilon 0.0011981619133878778 total_time 733.0\n",
      "episode 14959, initial_state (4, 13, 5), reward 1267.0, memory_length 2000, epsilon 0.0011927827469793923 total_time 726.0\n",
      "episode 14969, initial_state (3, 16, 6), reward 1751.0, memory_length 2000, epsilon 0.0011874277324372886 total_time 725.0\n",
      "episode 14979, initial_state (0, 19, 2), reward 1240.0, memory_length 2000, epsilon 0.00118209676132234 total_time 724.0\n",
      "episode 14989, initial_state (1, 11, 5), reward 1679.0, memory_length 2000, epsilon 0.0011767897256821991 total_time 721.0\n",
      "episode 14999, initial_state (0, 11, 2), reward 1497.0, memory_length 2000, epsilon 0.0011715065180492132 total_time 723.0\n",
      "Saving Model 15000\n",
      "Saving Model 15000 Weights\n",
      "episode 15009, initial_state (1, 21, 5), reward 1863.0, memory_length 2000, epsilon 0.0011662470314382479 total_time 726.0\n",
      "episode 15019, initial_state (3, 12, 4), reward 1462.0, memory_length 2000, epsilon 0.0011610111593445178 total_time 724.0\n",
      "episode 15029, initial_state (3, 6, 6), reward 1269.0, memory_length 2000, epsilon 0.0011557987957414345 total_time 723.0\n",
      "episode 15039, initial_state (1, 5, 0), reward 1410.0, memory_length 2000, epsilon 0.001150609835078458 total_time 726.0\n",
      "episode 15049, initial_state (1, 2, 0), reward 1382.0, memory_length 2000, epsilon 0.0011454441722789565 total_time 724.0\n",
      "episode 15059, initial_state (3, 2, 2), reward 1424.0, memory_length 2000, epsilon 0.0011403017027380835 total_time 725.0\n",
      "episode 15069, initial_state (2, 22, 0), reward 1733.0, memory_length 2000, epsilon 0.0011351823223206526 total_time 727.0\n",
      "episode 15079, initial_state (2, 3, 6), reward 1341.0, memory_length 2000, epsilon 0.0011300859273590366 total_time 726.0\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "episode 15089, initial_state (3, 9, 0), reward 1640.0, memory_length 2000, epsilon 0.0011250124146510636 total_time 725.0\n",
      "episode 15099, initial_state (1, 6, 1), reward 1290.0, memory_length 2000, epsilon 0.0011199616814579277 total_time 721.0\n",
      "episode 15109, initial_state (4, 1, 0), reward 1035.0, memory_length 2000, epsilon 0.0011149336255021098 total_time 725.0\n",
      "episode 15119, initial_state (3, 15, 2), reward 1523.0, memory_length 2000, epsilon 0.0011099281449653035 total_time 725.0\n",
      "episode 15129, initial_state (4, 3, 0), reward 1535.0, memory_length 2000, epsilon 0.0011049451384863578 total_time 724.0\n",
      "episode 15139, initial_state (2, 9, 4), reward 1521.0, memory_length 2000, epsilon 0.001099984505159221 total_time 721.0\n",
      "episode 15149, initial_state (0, 4, 5), reward 1352.0, memory_length 2000, epsilon 0.0010950461445308987 total_time 721.0\n",
      "episode 15159, initial_state (0, 7, 4), reward 1712.0, memory_length 2000, epsilon 0.0010901299565994197 total_time 723.0\n",
      "episode 15169, initial_state (4, 16, 1), reward 1017.0, memory_length 2000, epsilon 0.001085235841811811 total_time 724.0\n",
      "episode 15179, initial_state (3, 19, 3), reward 1430.0, memory_length 2000, epsilon 0.0010803637010620795 total_time 721.0\n",
      "episode 15189, initial_state (2, 6, 2), reward 1212.0, memory_length 2000, epsilon 0.0010755134356892088 total_time 724.0\n",
      "episode 15199, initial_state (0, 0, 4), reward 1494.0, memory_length 2000, epsilon 0.00107068494747516 total_time 727.0\n",
      "episode 15209, initial_state (3, 9, 5), reward 1113.0, memory_length 2000, epsilon 0.0010658781386428814 total_time 721.0\n",
      "episode 15219, initial_state (4, 3, 4), reward 1113.0, memory_length 2000, epsilon 0.001061092911854331 total_time 724.0\n",
      "episode 15229, initial_state (2, 6, 6), reward 1523.0, memory_length 2000, epsilon 0.0010563291702085007 total_time 722.0\n",
      "episode 15239, initial_state (3, 22, 4), reward 1656.0, memory_length 2000, epsilon 0.0010515868172394603 total_time 728.0\n",
      "episode 15249, initial_state (2, 19, 5), reward 1323.0, memory_length 2000, epsilon 0.0010468657569144006 total_time 722.0\n",
      "episode 15259, initial_state (3, 21, 3), reward 1319.0, memory_length 2000, epsilon 0.0010421658936316885 total_time 721.0\n",
      "episode 15269, initial_state (0, 15, 2), reward 1157.0, memory_length 2000, epsilon 0.0010374871322189325 total_time 723.0\n",
      "episode 15279, initial_state (1, 8, 1), reward 1659.0, memory_length 2000, epsilon 0.0010328293779310524 total_time 728.0\n",
      "episode 15289, initial_state (1, 17, 3), reward 1461.0, memory_length 2000, epsilon 0.001028192536448366 total_time 722.0\n",
      "episode 15299, initial_state (2, 17, 0), reward 1379.0, memory_length 2000, epsilon 0.0010235765138746744 total_time 722.0\n",
      "episode 15309, initial_state (4, 19, 0), reward 1467.0, memory_length 2000, epsilon 0.001018981216735363 total_time 733.0\n",
      "episode 15319, initial_state (2, 21, 2), reward 1530.0, memory_length 2000, epsilon 0.0010144065519755085 total_time 730.0\n",
      "episode 15329, initial_state (0, 7, 2), reward 1581.0, memory_length 2000, epsilon 0.0010098524269579914 total_time 725.0\n",
      "episode 15339, initial_state (0, 7, 6), reward 1465.0, memory_length 2000, epsilon 0.0010053187494616252 total_time 724.0\n",
      "episode 15349, initial_state (4, 1, 0), reward 1196.0, memory_length 2000, epsilon 0.0010008054276792859 total_time 724.0\n",
      "episode 15359, initial_state (1, 15, 6), reward 1296.0, memory_length 2000, epsilon 0.000996312370216053 total_time 722.0\n",
      "episode 15369, initial_state (0, 12, 2), reward 1512.0, memory_length 2000, epsilon 0.0009918394860873596 total_time 729.0\n",
      "episode 15379, initial_state (2, 11, 0), reward 1522.0, memory_length 2000, epsilon 0.0009873866847171502 total_time 727.0\n",
      "episode 15389, initial_state (2, 7, 6), reward 1708.0, memory_length 2000, epsilon 0.0009829538759360424 total_time 721.0\n",
      "episode 15399, initial_state (4, 23, 5), reward 1450.0, memory_length 2000, epsilon 0.0009785409699795086 total_time 724.0\n",
      "episode 15409, initial_state (2, 9, 0), reward 1586.0, memory_length 2000, epsilon 0.0009741478774860519 total_time 724.0\n",
      "episode 15419, initial_state (1, 5, 6), reward 1270.0, memory_length 2000, epsilon 0.0009697745094953994 total_time 724.0\n",
      "episode 15429, initial_state (3, 14, 3), reward 1396.0, memory_length 2000, epsilon 0.0009654207774467008 total_time 724.0\n",
      "episode 15439, initial_state (2, 11, 4), reward 1560.0, memory_length 2000, epsilon 0.0009610865931767315 total_time 721.0\n",
      "episode 15449, initial_state (0, 1, 2), reward 1340.0, memory_length 2000, epsilon 0.0009567718689181126 total_time 721.0\n",
      "episode 15459, initial_state (1, 8, 4), reward 1211.0, memory_length 2000, epsilon 0.0009524765172975306 total_time 723.0\n",
      "episode 15469, initial_state (1, 8, 3), reward 1304.0, memory_length 2000, epsilon 0.0009482004513339684 total_time 724.0\n",
      "episode 15479, initial_state (2, 15, 1), reward 1449.0, memory_length 2000, epsilon 0.0009439435844369451 total_time 722.0\n",
      "episode 15489, initial_state (3, 22, 3), reward 1712.0, memory_length 2000, epsilon 0.0009397058304047586 total_time 726.0\n",
      "episode 15499, initial_state (4, 9, 5), reward 1442.0, memory_length 2000, epsilon 0.0009354871034227459 total_time 721.0\n",
      "episode 15509, initial_state (0, 23, 2), reward 1576.0, memory_length 2000, epsilon 0.0009312873180615416 total_time 724.0\n",
      "episode 15519, initial_state (2, 3, 6), reward 1617.0, memory_length 2000, epsilon 0.0009271063892753483 total_time 723.0\n",
      "episode 15529, initial_state (4, 18, 0), reward 1602.0, memory_length 2000, epsilon 0.0009229442324002163 total_time 723.0\n",
      "episode 15539, initial_state (1, 4, 4), reward 1627.0, memory_length 2000, epsilon 0.000918800763152325 total_time 729.0\n",
      "episode 15549, initial_state (3, 6, 4), reward 1098.0, memory_length 2000, epsilon 0.0009146758976262811 total_time 721.0\n",
      "episode 15559, initial_state (2, 5, 5), reward 1533.0, memory_length 2000, epsilon 0.0009105695522934171 total_time 722.0\n",
      "episode 15569, initial_state (1, 15, 3), reward 1503.0, memory_length 2000, epsilon 0.0009064816440000994 total_time 726.0\n",
      "episode 15579, initial_state (3, 14, 3), reward 1348.0, memory_length 2000, epsilon 0.0009024120899660463 total_time 726.0\n",
      "episode 15589, initial_state (2, 0, 6), reward 1311.0, memory_length 2000, epsilon 0.000898360807782648 total_time 728.0\n",
      "episode 15599, initial_state (4, 0, 4), reward 1350.0, memory_length 2000, epsilon 0.0008943277154113025 total_time 733.0\n",
      "episode 15609, initial_state (1, 3, 0), reward 1585.0, memory_length 2000, epsilon 0.0008903127311817516 total_time 726.0\n",
      "episode 15619, initial_state (1, 17, 3), reward 1504.0, memory_length 2000, epsilon 0.0008863157737904274 total_time 728.0\n",
      "episode 15629, initial_state (0, 9, 4), reward 1684.0, memory_length 2000, epsilon 0.0008823367622988061 total_time 728.0\n",
      "episode 15639, initial_state (3, 6, 6), reward 1390.0, memory_length 2000, epsilon 0.0008783756161317699 total_time 721.0\n",
      "episode 15649, initial_state (3, 19, 6), reward 1595.0, memory_length 2000, epsilon 0.0008744322550759718 total_time 721.0\n",
      "episode 15659, initial_state (3, 19, 4), reward 1062.0, memory_length 2000, epsilon 0.0008705065992782166 total_time 725.0\n",
      "episode 15669, initial_state (0, 16, 0), reward 1552.0, memory_length 2000, epsilon 0.0008665985692438403 total_time 721.0\n",
      "episode 15679, initial_state (4, 7, 2), reward 1618.0, memory_length 2000, epsilon 0.0008627080858351011 total_time 722.0\n",
      "episode 15689, initial_state (1, 0, 2), reward 1609.0, memory_length 2000, epsilon 0.0008588350702695778 total_time 721.0\n",
      "episode 15699, initial_state (4, 17, 0), reward 921.0, memory_length 2000, epsilon 0.0008549794441185712 total_time 724.0\n",
      "episode 15709, initial_state (4, 23, 0), reward 1427.0, memory_length 2000, epsilon 0.0008511411293055209 total_time 727.0\n",
      "episode 15719, initial_state (0, 1, 0), reward 1138.0, memory_length 2000, epsilon 0.0008473200481044207 total_time 726.0\n",
      "episode 15729, initial_state (1, 12, 3), reward 1580.0, memory_length 2000, epsilon 0.0008435161231382458 total_time 725.0\n",
      "episode 15739, initial_state (2, 5, 3), reward 1324.0, memory_length 2000, epsilon 0.0008397292773773862 total_time 722.0\n",
      "episode 15749, initial_state (3, 20, 4), reward 1217.0, memory_length 2000, epsilon 0.0008359594341380847 total_time 721.0\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "episode 15759, initial_state (3, 7, 4), reward 1212.0, memory_length 2000, epsilon 0.0008322065170808873 total_time 726.0\n",
      "episode 15769, initial_state (0, 10, 0), reward 1523.0, memory_length 2000, epsilon 0.0008284704502090953 total_time 723.0\n",
      "episode 15779, initial_state (3, 18, 1), reward 1548.0, memory_length 2000, epsilon 0.0008247511578672272 total_time 728.0\n",
      "episode 15789, initial_state (3, 2, 6), reward 1294.0, memory_length 2000, epsilon 0.0008210485647394864 total_time 721.0\n",
      "episode 15799, initial_state (4, 21, 6), reward 1567.0, memory_length 2000, epsilon 0.0008173625958482344 total_time 729.0\n",
      "episode 15809, initial_state (0, 5, 0), reward 1608.0, memory_length 2000, epsilon 0.0008136931765524756 total_time 722.0\n",
      "episode 15819, initial_state (1, 15, 0), reward 1386.0, memory_length 2000, epsilon 0.000810040232546344 total_time 722.0\n",
      "episode 15829, initial_state (1, 15, 6), reward 1481.0, memory_length 2000, epsilon 0.0008064036898575987 total_time 722.0\n",
      "episode 15839, initial_state (2, 10, 5), reward 1532.0, memory_length 2000, epsilon 0.0008027834748461268 total_time 726.0\n",
      "episode 15849, initial_state (3, 23, 6), reward 1657.0, memory_length 2000, epsilon 0.0007991795142024488 total_time 731.0\n",
      "episode 15859, initial_state (0, 19, 2), reward 1489.0, memory_length 2000, epsilon 0.0007955917349462396 total_time 721.0\n",
      "episode 15869, initial_state (0, 23, 6), reward 1285.0, memory_length 2000, epsilon 0.0007920200644248465 total_time 721.0\n",
      "episode 15879, initial_state (2, 7, 2), reward 1405.0, memory_length 2000, epsilon 0.0007884644303118194 total_time 729.0\n",
      "episode 15889, initial_state (2, 1, 0), reward 1445.0, memory_length 2000, epsilon 0.000784924760605446 total_time 729.0\n",
      "episode 15899, initial_state (3, 3, 0), reward 1297.0, memory_length 2000, epsilon 0.0007814009836272945 total_time 729.0\n",
      "episode 15909, initial_state (0, 2, 6), reward 1455.0, memory_length 2000, epsilon 0.0007778930280207594 total_time 722.0\n",
      "episode 15919, initial_state (1, 10, 6), reward 1239.0, memory_length 2000, epsilon 0.0007744008227496202 total_time 722.0\n",
      "episode 15929, initial_state (4, 10, 1), reward 1457.0, memory_length 2000, epsilon 0.0007709242970966013 total_time 722.0\n",
      "episode 15939, initial_state (0, 16, 0), reward 1275.0, memory_length 2000, epsilon 0.000767463380661939 total_time 724.0\n",
      "episode 15949, initial_state (4, 18, 6), reward 1431.0, memory_length 2000, epsilon 0.0007640180033619579 total_time 733.0\n",
      "episode 15959, initial_state (0, 20, 2), reward 1362.0, memory_length 2000, epsilon 0.0007605880954276488 total_time 721.0\n",
      "episode 15969, initial_state (2, 8, 0), reward 1787.0, memory_length 2000, epsilon 0.0007571735874032595 total_time 723.0\n",
      "episode 15979, initial_state (3, 8, 4), reward 1579.0, memory_length 2000, epsilon 0.0007537744101448857 total_time 721.0\n",
      "episode 15989, initial_state (4, 14, 4), reward 1320.0, memory_length 2000, epsilon 0.0007503904948190719 total_time 721.0\n",
      "episode 15999, initial_state (4, 19, 2), reward 1281.0, memory_length 2000, epsilon 0.0007470217729014176 total_time 723.0\n",
      "Saving Model 16000\n",
      "Saving Model 16000 Weights\n",
      "episode 16009, initial_state (4, 0, 1), reward 1250.0, memory_length 2000, epsilon 0.0007436681761751877 total_time 722.0\n",
      "episode 16019, initial_state (0, 22, 2), reward 1638.0, memory_length 2000, epsilon 0.0007403296367299345 total_time 723.0\n",
      "episode 16029, initial_state (3, 11, 3), reward 1393.0, memory_length 2000, epsilon 0.0007370060869601201 total_time 725.0\n",
      "episode 16039, initial_state (1, 20, 1), reward 1387.0, memory_length 2000, epsilon 0.0007336974595637481 total_time 726.0\n",
      "episode 16049, initial_state (2, 4, 2), reward 1385.0, memory_length 2000, epsilon 0.0007304036875410014 total_time 721.0\n",
      "episode 16059, initial_state (0, 10, 5), reward 984.0, memory_length 2000, epsilon 0.0007271247041928826 total_time 723.0\n",
      "episode 16069, initial_state (2, 17, 0), reward 1632.0, memory_length 2000, epsilon 0.0007238604431198674 total_time 722.0\n",
      "episode 16079, initial_state (4, 10, 6), reward 1197.0, memory_length 2000, epsilon 0.0007206108382205577 total_time 723.0\n",
      "episode 16089, initial_state (3, 9, 3), reward 1348.0, memory_length 2000, epsilon 0.0007173758236903433 total_time 724.0\n",
      "episode 16099, initial_state (3, 18, 4), reward 1109.0, memory_length 2000, epsilon 0.0007141553340200698 total_time 724.0\n",
      "episode 16109, initial_state (4, 11, 2), reward 1474.0, memory_length 2000, epsilon 0.0007109493039947103 total_time 724.0\n",
      "episode 16119, initial_state (1, 12, 4), reward 1243.0, memory_length 2000, epsilon 0.0007077576686920477 total_time 721.0\n",
      "episode 16129, initial_state (1, 21, 6), reward 1451.0, memory_length 2000, epsilon 0.0007045803634813582 total_time 721.0\n",
      "episode 16139, initial_state (4, 19, 6), reward 1113.0, memory_length 2000, epsilon 0.0007014173240221025 total_time 723.0\n",
      "episode 16149, initial_state (3, 7, 0), reward 1684.0, memory_length 2000, epsilon 0.0006982684862626237 total_time 722.0\n",
      "episode 16159, initial_state (0, 6, 5), reward 1188.0, memory_length 2000, epsilon 0.00069513378643885 total_time 721.0\n",
      "episode 16169, initial_state (0, 18, 4), reward 1617.0, memory_length 2000, epsilon 0.0006920131610730018 total_time 725.0\n",
      "episode 16179, initial_state (1, 18, 2), reward 1298.0, memory_length 2000, epsilon 0.0006889065469723092 total_time 725.0\n",
      "episode 16189, initial_state (0, 17, 0), reward 1481.0, memory_length 2000, epsilon 0.0006858138812277307 total_time 723.0\n",
      "episode 16199, initial_state (0, 23, 3), reward 1300.0, memory_length 2000, epsilon 0.0006827351012126793 total_time 722.0\n",
      "episode 16209, initial_state (3, 6, 0), reward 1616.0, memory_length 2000, epsilon 0.0006796701445817548 total_time 730.0\n",
      "episode 16219, initial_state (0, 7, 6), reward 1367.0, memory_length 2000, epsilon 0.0006766189492694798 total_time 728.0\n",
      "episode 16229, initial_state (2, 21, 0), reward 1216.0, memory_length 2000, epsilon 0.0006735814534890454 total_time 727.0\n",
      "episode 16239, initial_state (4, 9, 6), reward 1172.0, memory_length 2000, epsilon 0.0006705575957310582 total_time 731.0\n",
      "episode 16249, initial_state (2, 14, 5), reward 1289.0, memory_length 2000, epsilon 0.0006675473147622955 total_time 728.0\n",
      "episode 16259, initial_state (2, 2, 5), reward 1356.0, memory_length 2000, epsilon 0.0006645505496244653 total_time 725.0\n",
      "episode 16269, initial_state (2, 7, 6), reward 1274.0, memory_length 2000, epsilon 0.0006615672396329699 total_time 725.0\n",
      "episode 16279, initial_state (3, 16, 3), reward 1203.0, memory_length 2000, epsilon 0.0006585973243756806 total_time 725.0\n",
      "episode 16289, initial_state (4, 18, 4), reward 1460.0, memory_length 2000, epsilon 0.000655640743711712 total_time 724.0\n",
      "episode 16299, initial_state (2, 14, 4), reward 1553.0, memory_length 2000, epsilon 0.0006526974377702047 total_time 729.0\n",
      "episode 16309, initial_state (1, 17, 3), reward 1386.0, memory_length 2000, epsilon 0.0006497673469491133 total_time 722.0\n",
      "episode 16319, initial_state (4, 14, 1), reward 1424.0, memory_length 2000, epsilon 0.0006468504119139974 total_time 721.0\n",
      "episode 16329, initial_state (4, 18, 1), reward 1734.0, memory_length 2000, epsilon 0.0006439465735968235 total_time 721.0\n",
      "episode 16339, initial_state (2, 4, 1), reward 982.0, memory_length 2000, epsilon 0.0006410557731947663 total_time 732.0\n",
      "episode 16349, initial_state (0, 23, 0), reward 1639.0, memory_length 2000, epsilon 0.000638177952169019 total_time 724.0\n",
      "episode 16359, initial_state (3, 7, 2), reward 1341.0, memory_length 2000, epsilon 0.000635313052243608 total_time 726.0\n",
      "episode 16369, initial_state (2, 23, 4), reward 1468.0, memory_length 2000, epsilon 0.0006324610154042109 total_time 723.0\n",
      "episode 16379, initial_state (4, 0, 0), reward 1125.0, memory_length 2000, epsilon 0.0006296217838969846 total_time 728.0\n",
      "episode 16389, initial_state (1, 8, 0), reward 1273.0, memory_length 2000, epsilon 0.0006267953002273942 total_time 725.0\n",
      "episode 16399, initial_state (0, 7, 4), reward 1798.0, memory_length 2000, epsilon 0.0006239815071590488 total_time 728.0\n",
      "episode 16409, initial_state (2, 7, 6), reward 1335.0, memory_length 2000, epsilon 0.0006211803477125426 total_time 726.0\n",
      "episode 16419, initial_state (4, 3, 3), reward 1294.0, memory_length 2000, epsilon 0.0006183917651643017 total_time 727.0\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "episode 16429, initial_state (2, 2, 5), reward 1269.0, memory_length 2000, epsilon 0.0006156157030454329 total_time 721.0\n",
      "episode 16439, initial_state (2, 4, 6), reward 1442.0, memory_length 2000, epsilon 0.0006128521051405842 total_time 729.0\n",
      "episode 16449, initial_state (3, 0, 4), reward 1355.0, memory_length 2000, epsilon 0.0006101009154868034 total_time 728.0\n",
      "episode 16459, initial_state (2, 4, 0), reward 1729.0, memory_length 2000, epsilon 0.0006073620783724061 total_time 723.0\n",
      "episode 16469, initial_state (4, 7, 3), reward 1197.0, memory_length 2000, epsilon 0.0006046355383358478 total_time 721.0\n",
      "episode 16479, initial_state (1, 20, 6), reward 1432.0, memory_length 2000, epsilon 0.0006019212401645982 total_time 723.0\n",
      "episode 16489, initial_state (3, 8, 0), reward 1579.0, memory_length 2000, epsilon 0.0005992191288940275 total_time 725.0\n",
      "episode 16499, initial_state (4, 1, 1), reward 1620.0, memory_length 2000, epsilon 0.0005965291498062898 total_time 725.0\n",
      "episode 16509, initial_state (3, 22, 2), reward 1501.0, memory_length 2000, epsilon 0.0005938512484292168 total_time 725.0\n",
      "episode 16519, initial_state (3, 4, 1), reward 1119.0, memory_length 2000, epsilon 0.0005911853705352147 total_time 728.0\n",
      "episode 16529, initial_state (3, 20, 6), reward 1333.0, memory_length 2000, epsilon 0.000588531462140164 total_time 723.0\n",
      "episode 16539, initial_state (1, 19, 2), reward 1557.0, memory_length 2000, epsilon 0.0005858894695023293 total_time 733.0\n",
      "episode 16549, initial_state (2, 13, 4), reward 1347.0, memory_length 2000, epsilon 0.0005832593391212696 total_time 726.0\n",
      "episode 16559, initial_state (0, 10, 2), reward 1564.0, memory_length 2000, epsilon 0.0005806410177367549 total_time 722.0\n",
      "episode 16569, initial_state (2, 17, 4), reward 1190.0, memory_length 2000, epsilon 0.0005780344523276878 total_time 724.0\n",
      "episode 16579, initial_state (4, 12, 1), reward 1467.0, memory_length 2000, epsilon 0.0005754395901110292 total_time 721.0\n",
      "episode 16589, initial_state (2, 21, 0), reward 1235.0, memory_length 2000, epsilon 0.0005728563785407307 total_time 722.0\n",
      "episode 16599, initial_state (1, 17, 2), reward 1554.0, memory_length 2000, epsilon 0.0005702847653066698 total_time 723.0\n",
      "episode 16609, initial_state (4, 18, 2), reward 1308.0, memory_length 2000, epsilon 0.0005677246983335907 total_time 721.0\n",
      "episode 16619, initial_state (3, 18, 2), reward 1417.0, memory_length 2000, epsilon 0.0005651761257800501 total_time 723.0\n",
      "episode 16629, initial_state (2, 8, 6), reward 1173.0, memory_length 2000, epsilon 0.0005626389960373658 total_time 726.0\n",
      "episode 16639, initial_state (1, 14, 4), reward 1266.0, memory_length 2000, epsilon 0.0005601132577285743 total_time 723.0\n",
      "episode 16649, initial_state (0, 14, 6), reward 1352.0, memory_length 2000, epsilon 0.0005575988597073886 total_time 725.0\n",
      "episode 16659, initial_state (2, 3, 5), reward 1407.0, memory_length 2000, epsilon 0.0005550957510571627 total_time 722.0\n",
      "episode 16669, initial_state (0, 3, 3), reward 1602.0, memory_length 2000, epsilon 0.0005526038810898609 total_time 721.0\n",
      "episode 16679, initial_state (3, 12, 3), reward 1512.0, memory_length 2000, epsilon 0.0005501231993450318 total_time 726.0\n",
      "episode 16689, initial_state (3, 18, 3), reward 1182.0, memory_length 2000, epsilon 0.0005476536555887843 total_time 722.0\n",
      "episode 16699, initial_state (3, 15, 6), reward 1400.0, memory_length 2000, epsilon 0.0005451951998127735 total_time 725.0\n",
      "episode 16709, initial_state (4, 7, 1), reward 1459.0, memory_length 2000, epsilon 0.0005427477822331857 total_time 723.0\n",
      "episode 16719, initial_state (4, 23, 1), reward 1523.0, memory_length 2000, epsilon 0.0005403113532897315 total_time 724.0\n",
      "episode 16729, initial_state (1, 0, 3), reward 1527.0, memory_length 2000, epsilon 0.000537885863644642 total_time 723.0\n",
      "episode 16739, initial_state (4, 13, 0), reward 1435.0, memory_length 2000, epsilon 0.0005354712641816679 total_time 721.0\n",
      "episode 16749, initial_state (0, 20, 1), reward 1513.0, memory_length 2000, epsilon 0.0005330675060050881 total_time 726.0\n",
      "episode 16759, initial_state (0, 19, 6), reward 1348.0, memory_length 2000, epsilon 0.0005306745404387175 total_time 721.0\n",
      "episode 16769, initial_state (0, 5, 5), reward 1022.0, memory_length 2000, epsilon 0.0005282923190249216 total_time 724.0\n",
      "episode 16779, initial_state (1, 16, 6), reward 1070.0, memory_length 2000, epsilon 0.0005259207935236355 total_time 724.0\n",
      "episode 16789, initial_state (1, 5, 0), reward 1163.0, memory_length 2000, epsilon 0.0005235599159113864 total_time 721.0\n",
      "episode 16799, initial_state (2, 1, 5), reward 1431.0, memory_length 2000, epsilon 0.0005212096383803219 total_time 728.0\n",
      "episode 16809, initial_state (1, 2, 0), reward 1698.0, memory_length 2000, epsilon 0.0005188699133372419 total_time 728.0\n",
      "episode 16819, initial_state (0, 6, 2), reward 935.0, memory_length 2000, epsilon 0.0005165406934026345 total_time 721.0\n",
      "episode 16829, initial_state (1, 12, 2), reward 1477.0, memory_length 2000, epsilon 0.0005142219314097165 total_time 721.0\n",
      "episode 16839, initial_state (2, 23, 6), reward 1223.0, memory_length 2000, epsilon 0.0005119135804034778 total_time 727.0\n",
      "episode 16849, initial_state (1, 9, 3), reward 1156.0, memory_length 2000, epsilon 0.0005096155936397318 total_time 726.0\n",
      "episode 16859, initial_state (2, 21, 6), reward 1202.0, memory_length 2000, epsilon 0.0005073279245841682 total_time 721.0\n",
      "episode 16869, initial_state (3, 10, 2), reward 1464.0, memory_length 2000, epsilon 0.0005050505269114104 total_time 724.0\n",
      "episode 16879, initial_state (4, 6, 4), reward 1495.0, memory_length 2000, epsilon 0.000502783354504078 total_time 723.0\n",
      "episode 16889, initial_state (3, 0, 6), reward 1436.0, memory_length 2000, epsilon 0.0005005263614518516 total_time 726.0\n",
      "episode 16899, initial_state (3, 21, 1), reward 1311.0, memory_length 2000, epsilon 0.000498279502050545 total_time 725.0\n",
      "episode 16909, initial_state (1, 18, 4), reward 1548.0, memory_length 2000, epsilon 0.0004960427308011788 total_time 730.0\n",
      "episode 16919, initial_state (0, 1, 1), reward 1135.0, memory_length 2000, epsilon 0.0004938160024090586 total_time 726.0\n",
      "episode 16929, initial_state (1, 2, 0), reward 1371.0, memory_length 2000, epsilon 0.0004915992717828584 total_time 729.0\n",
      "episode 16939, initial_state (3, 18, 5), reward 1346.0, memory_length 2000, epsilon 0.0004893924940337076 total_time 725.0\n",
      "episode 16949, initial_state (2, 4, 4), reward 1146.0, memory_length 2000, epsilon 0.00048719562447428077 total_time 723.0\n",
      "episode 16959, initial_state (4, 23, 5), reward 1260.0, memory_length 2000, epsilon 0.0004850086186178944 total_time 728.0\n",
      "episode 16969, initial_state (4, 19, 2), reward 1724.0, memory_length 2000, epsilon 0.00048283143217760536 total_time 722.0\n",
      "episode 16979, initial_state (1, 7, 5), reward 1608.0, memory_length 2000, epsilon 0.0004806640210653137 total_time 722.0\n",
      "episode 16989, initial_state (0, 19, 0), reward 1271.0, memory_length 2000, epsilon 0.0004785063413908709 total_time 722.0\n",
      "episode 16999, initial_state (1, 18, 5), reward 1156.0, memory_length 2000, epsilon 0.0004763583494611888 total_time 722.0\n",
      "Saving Model 17000\n",
      "Saving Model 17000 Weights\n",
      "episode 17009, initial_state (4, 8, 0), reward 1553.0, memory_length 2000, epsilon 0.000474220001779358 total_time 721.0\n",
      "episode 17019, initial_state (0, 12, 4), reward 1264.0, memory_length 2000, epsilon 0.0004720912550437648 total_time 724.0\n",
      "episode 17029, initial_state (1, 17, 0), reward 1548.0, memory_length 2000, epsilon 0.00046997206614721507 total_time 726.0\n",
      "episode 17039, initial_state (4, 20, 0), reward 1512.0, memory_length 2000, epsilon 0.0004678623921760617 total_time 728.0\n",
      "episode 17049, initial_state (2, 14, 2), reward 1457.0, memory_length 2000, epsilon 0.0004657621904093337 total_time 721.0\n",
      "episode 17059, initial_state (0, 19, 4), reward 1197.0, memory_length 2000, epsilon 0.00046367141831787413 total_time 725.0\n",
      "episode 17069, initial_state (1, 3, 2), reward 1344.0, memory_length 2000, epsilon 0.0004615900335634766 total_time 723.0\n",
      "episode 17079, initial_state (3, 23, 5), reward 1259.0, memory_length 2000, epsilon 0.00045951799399802874 total_time 722.0\n",
      "episode 17089, initial_state (4, 23, 3), reward 1316.0, memory_length 2000, epsilon 0.0004574552576626589 total_time 721.0\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "episode 17099, initial_state (2, 10, 4), reward 1320.0, memory_length 2000, epsilon 0.00045540178278688503 total_time 727.0\n",
      "episode 17109, initial_state (3, 8, 4), reward 1413.0, memory_length 2000, epsilon 0.0004533575277877711 total_time 724.0\n",
      "episode 17119, initial_state (1, 1, 6), reward 1224.0, memory_length 2000, epsilon 0.00045132245126908356 total_time 726.0\n",
      "episode 17129, initial_state (3, 13, 0), reward 1500.0, memory_length 2000, epsilon 0.0004492965120204534 total_time 726.0\n",
      "episode 17139, initial_state (4, 20, 6), reward 1305.0, memory_length 2000, epsilon 0.0004472796690165419 total_time 726.0\n",
      "episode 17149, initial_state (4, 19, 0), reward 1433.0, memory_length 2000, epsilon 0.0004452718814162086 total_time 722.0\n",
      "episode 17159, initial_state (1, 5, 2), reward 1607.0, memory_length 2000, epsilon 0.0004432731085616863 total_time 726.0\n",
      "episode 17169, initial_state (2, 14, 5), reward 1405.0, memory_length 2000, epsilon 0.0004412833099777564 total_time 723.0\n",
      "episode 17179, initial_state (3, 22, 0), reward 1656.0, memory_length 2000, epsilon 0.00043930244537092973 total_time 731.0\n",
      "episode 17189, initial_state (1, 13, 5), reward 1311.0, memory_length 2000, epsilon 0.0004373304746286301 total_time 723.0\n",
      "episode 17199, initial_state (0, 21, 6), reward 1734.0, memory_length 2000, epsilon 0.00043536735781838314 total_time 730.0\n",
      "episode 17209, initial_state (2, 5, 0), reward 1627.0, memory_length 2000, epsilon 0.0004334130551870055 total_time 725.0\n",
      "episode 17219, initial_state (2, 4, 5), reward 1510.0, memory_length 2000, epsilon 0.00043146752715980257 total_time 721.0\n",
      "episode 17229, initial_state (3, 4, 1), reward 1404.0, memory_length 2000, epsilon 0.0004295307343397653 total_time 724.0\n",
      "episode 17239, initial_state (0, 5, 6), reward 1357.0, memory_length 2000, epsilon 0.00042760263750677276 total_time 730.0\n",
      "episode 17249, initial_state (0, 16, 0), reward 1101.0, memory_length 2000, epsilon 0.0004256831976167988 total_time 723.0\n",
      "episode 17259, initial_state (0, 5, 2), reward 1333.0, memory_length 2000, epsilon 0.0004237723758011191 total_time 724.0\n",
      "episode 17269, initial_state (3, 3, 1), reward 1470.0, memory_length 2000, epsilon 0.00042187013336552714 total_time 722.0\n",
      "episode 17279, initial_state (1, 10, 5), reward 1500.0, memory_length 2000, epsilon 0.00041997643178954847 total_time 723.0\n",
      "episode 17289, initial_state (3, 9, 1), reward 1230.0, memory_length 2000, epsilon 0.0004180912327256615 total_time 724.0\n",
      "episode 17299, initial_state (3, 0, 6), reward 1428.0, memory_length 2000, epsilon 0.0004162144979985212 total_time 723.0\n",
      "episode 17309, initial_state (0, 0, 6), reward 1539.0, memory_length 2000, epsilon 0.00041434618960418436 total_time 725.0\n",
      "episode 17319, initial_state (3, 0, 0), reward 1359.0, memory_length 2000, epsilon 0.00041248626970934264 total_time 726.0\n",
      "episode 17329, initial_state (4, 21, 0), reward 1535.0, memory_length 2000, epsilon 0.0004106347006505545 total_time 727.0\n",
      "episode 17339, initial_state (3, 6, 3), reward 1602.0, memory_length 2000, epsilon 0.0004087914449334833 total_time 731.0\n",
      "episode 17349, initial_state (3, 9, 6), reward 1285.0, memory_length 2000, epsilon 0.00040695646523213817 total_time 721.0\n",
      "episode 17359, initial_state (4, 10, 3), reward 1401.0, memory_length 2000, epsilon 0.00040512972438811666 total_time 732.0\n",
      "episode 17369, initial_state (2, 22, 0), reward 1549.0, memory_length 2000, epsilon 0.0004033111854098547 total_time 733.0\n",
      "episode 17379, initial_state (3, 16, 5), reward 1488.0, memory_length 2000, epsilon 0.00040150081147187564 total_time 724.0\n",
      "episode 17389, initial_state (1, 16, 3), reward 1635.0, memory_length 2000, epsilon 0.00039969856591404564 total_time 721.0\n",
      "episode 17399, initial_state (4, 20, 2), reward 1233.0, memory_length 2000, epsilon 0.0003979044122408308 total_time 728.0\n",
      "episode 17409, initial_state (2, 13, 1), reward 1421.0, memory_length 2000, epsilon 0.0003961183141205571 total_time 727.0\n",
      "episode 17419, initial_state (2, 21, 4), reward 1226.0, memory_length 2000, epsilon 0.00039434023538467716 total_time 722.0\n",
      "episode 17429, initial_state (0, 14, 4), reward 1575.0, memory_length 2000, epsilon 0.0003925701400270356 total_time 723.0\n",
      "episode 17439, initial_state (1, 2, 2), reward 1413.0, memory_length 2000, epsilon 0.0003908079922031411 total_time 727.0\n",
      "episode 17449, initial_state (4, 1, 2), reward 1294.0, memory_length 2000, epsilon 0.00038905375622943995 total_time 723.0\n",
      "episode 17459, initial_state (3, 22, 4), reward 1271.0, memory_length 2000, epsilon 0.00038730739658259406 total_time 724.0\n",
      "episode 17469, initial_state (4, 6, 2), reward 1228.0, memory_length 2000, epsilon 0.00038556887789876023 total_time 721.0\n",
      "episode 17479, initial_state (3, 5, 6), reward 1609.0, memory_length 2000, epsilon 0.0003838381649728761 total_time 723.0\n",
      "episode 17489, initial_state (0, 11, 6), reward 1639.0, memory_length 2000, epsilon 0.00038211522275794564 total_time 724.0\n",
      "episode 17499, initial_state (3, 2, 1), reward 1480.0, memory_length 2000, epsilon 0.0003804000163643303 total_time 721.0\n",
      "episode 17509, initial_state (2, 4, 2), reward 1368.0, memory_length 2000, epsilon 0.0003786925110590422 total_time 729.0\n",
      "episode 17519, initial_state (1, 10, 0), reward 1482.0, memory_length 2000, epsilon 0.00037699267226503995 total_time 726.0\n",
      "episode 17529, initial_state (1, 2, 5), reward 1366.0, memory_length 2000, epsilon 0.0003753004655605302 total_time 725.0\n",
      "episode 17539, initial_state (1, 7, 6), reward 1555.0, memory_length 2000, epsilon 0.0003736158566782694 total_time 721.0\n",
      "episode 17549, initial_state (2, 20, 5), reward 1364.0, memory_length 2000, epsilon 0.00037193881150487 total_time 730.0\n",
      "episode 17559, initial_state (1, 22, 1), reward 1441.0, memory_length 2000, epsilon 0.0003702692960801104 total_time 726.0\n",
      "episode 17569, initial_state (2, 18, 2), reward 1259.0, memory_length 2000, epsilon 0.00036860727659624544 total_time 725.0\n",
      "episode 17579, initial_state (0, 3, 1), reward 1629.0, memory_length 2000, epsilon 0.0003669527193973242 total_time 721.0\n",
      "episode 17589, initial_state (2, 22, 1), reward 1710.0, memory_length 2000, epsilon 0.0003653055909785068 total_time 729.0\n",
      "episode 17599, initial_state (4, 2, 0), reward 1568.0, memory_length 2000, epsilon 0.00036366585798538645 total_time 723.0\n",
      "episode 17609, initial_state (3, 11, 3), reward 1196.0, memory_length 2000, epsilon 0.0003620334872133144 total_time 724.0\n",
      "episode 17619, initial_state (3, 4, 6), reward 1136.0, memory_length 2000, epsilon 0.00036040844560672605 total_time 722.0\n",
      "episode 17629, initial_state (4, 8, 1), reward 1512.0, memory_length 2000, epsilon 0.0003587907002584736 total_time 726.0\n",
      "episode 17639, initial_state (0, 13, 4), reward 1149.0, memory_length 2000, epsilon 0.00035718021840915856 total_time 724.0\n",
      "episode 17649, initial_state (2, 0, 1), reward 1475.0, memory_length 2000, epsilon 0.0003555769674464685 total_time 722.0\n",
      "episode 17659, initial_state (4, 9, 2), reward 1056.0, memory_length 2000, epsilon 0.0003539809149045164 total_time 721.0\n",
      "episode 17669, initial_state (1, 1, 6), reward 1472.0, memory_length 2000, epsilon 0.0003523920284631844 total_time 721.0\n",
      "episode 17679, initial_state (1, 10, 1), reward 1288.0, memory_length 2000, epsilon 0.00035081027594746676 total_time 724.0\n",
      "episode 17689, initial_state (1, 1, 0), reward 1392.0, memory_length 2000, epsilon 0.00034923562532682156 total_time 728.0\n",
      "episode 17699, initial_state (1, 19, 4), reward 1585.0, memory_length 2000, epsilon 0.0003476680447145199 total_time 729.0\n",
      "episode 17709, initial_state (3, 2, 3), reward 1359.0, memory_length 2000, epsilon 0.0003461075023670007 total_time 721.0\n",
      "episode 17719, initial_state (0, 5, 5), reward 1173.0, memory_length 2000, epsilon 0.0003445539666832286 total_time 721.0\n",
      "episode 17729, initial_state (4, 4, 6), reward 1334.0, memory_length 2000, epsilon 0.00034300740620405205 total_time 731.0\n",
      "episode 17739, initial_state (1, 23, 2), reward 1571.0, memory_length 2000, epsilon 0.0003414677896115689 total_time 725.0\n",
      "episode 17749, initial_state (2, 23, 2), reward 1458.0, memory_length 2000, epsilon 0.0003399350857284907 total_time 728.0\n",
      "episode 17759, initial_state (1, 7, 3), reward 1611.0, memory_length 2000, epsilon 0.00033840926351751127 total_time 723.0\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "episode 17769, initial_state (2, 16, 4), reward 1716.0, memory_length 2000, epsilon 0.000336890292080679 total_time 725.0\n",
      "episode 17779, initial_state (1, 4, 1), reward 1189.0, memory_length 2000, epsilon 0.0003353781406587695 total_time 722.0\n",
      "episode 17789, initial_state (2, 14, 0), reward 1569.0, memory_length 2000, epsilon 0.00033387277863066583 total_time 729.0\n",
      "episode 17799, initial_state (0, 21, 0), reward 1680.0, memory_length 2000, epsilon 0.0003323741755127356 total_time 723.0\n",
      "episode 17809, initial_state (4, 10, 2), reward 1218.0, memory_length 2000, epsilon 0.0003308823009582134 total_time 721.0\n",
      "episode 17819, initial_state (0, 15, 3), reward 1265.0, memory_length 2000, epsilon 0.000329397124756589 total_time 725.0\n",
      "episode 17829, initial_state (3, 5, 2), reward 1619.0, memory_length 2000, epsilon 0.00032791861683299366 total_time 721.0\n",
      "episode 17839, initial_state (3, 22, 0), reward 1482.0, memory_length 2000, epsilon 0.0003264467472475914 total_time 723.0\n",
      "episode 17849, initial_state (0, 8, 1), reward 1540.0, memory_length 2000, epsilon 0.00032498148619497275 total_time 725.0\n",
      "episode 17859, initial_state (4, 7, 4), reward 1522.0, memory_length 2000, epsilon 0.00032352280400355137 total_time 724.0\n",
      "episode 17869, initial_state (4, 1, 0), reward 1359.0, memory_length 2000, epsilon 0.00032207067113496307 total_time 724.0\n",
      "episode 17879, initial_state (3, 21, 1), reward 1401.0, memory_length 2000, epsilon 0.00032062505818346754 total_time 725.0\n",
      "episode 17889, initial_state (1, 20, 1), reward 1210.0, memory_length 2000, epsilon 0.0003191859358753533 total_time 722.0\n",
      "episode 17899, initial_state (3, 9, 1), reward 1665.0, memory_length 2000, epsilon 0.0003177532750683447 total_time 730.0\n",
      "episode 17909, initial_state (2, 4, 2), reward 1736.0, memory_length 2000, epsilon 0.0003163270467510106 total_time 722.0\n",
      "episode 17919, initial_state (3, 13, 4), reward 1224.0, memory_length 2000, epsilon 0.00031490722204217925 total_time 723.0\n",
      "episode 17929, initial_state (4, 4, 6), reward 1700.0, memory_length 2000, epsilon 0.00031349377219035184 total_time 725.0\n",
      "episode 17939, initial_state (2, 21, 1), reward 1634.0, memory_length 2000, epsilon 0.0003120866685731205 total_time 731.0\n",
      "episode 17949, initial_state (3, 10, 2), reward 1352.0, memory_length 2000, epsilon 0.000310685882696589 total_time 722.0\n",
      "episode 17959, initial_state (2, 18, 0), reward 1553.0, memory_length 2000, epsilon 0.0003092913861947954 total_time 725.0\n",
      "episode 17969, initial_state (0, 1, 4), reward 1383.0, memory_length 2000, epsilon 0.00030790315082913786 total_time 722.0\n",
      "episode 17979, initial_state (4, 15, 5), reward 1404.0, memory_length 2000, epsilon 0.00030652114848780287 total_time 722.0\n",
      "episode 17989, initial_state (3, 3, 0), reward 1370.0, memory_length 2000, epsilon 0.0003051453511851957 total_time 723.0\n",
      "episode 17999, initial_state (0, 22, 2), reward 1371.0, memory_length 2000, epsilon 0.00030377573106137463 total_time 723.0\n",
      "Saving Model 18000\n",
      "Saving Model 18000 Weights\n",
      "episode 18009, initial_state (3, 18, 3), reward 1346.0, memory_length 2000, epsilon 0.0003024122603814841 total_time 728.0\n",
      "episode 18019, initial_state (0, 9, 6), reward 1366.0, memory_length 2000, epsilon 0.00030105491153519697 total_time 721.0\n",
      "episode 18029, initial_state (3, 10, 5), reward 1630.0, memory_length 2000, epsilon 0.00029970365703615255 total_time 727.0\n",
      "episode 18039, initial_state (4, 23, 6), reward 1498.0, memory_length 2000, epsilon 0.0002983584695214012 total_time 721.0\n",
      "episode 18049, initial_state (3, 23, 2), reward 1701.0, memory_length 2000, epsilon 0.0002970193217508497 total_time 723.0\n",
      "episode 18059, initial_state (3, 11, 1), reward 1497.0, memory_length 2000, epsilon 0.0002956861866067099 total_time 726.0\n",
      "episode 18069, initial_state (1, 1, 1), reward 1344.0, memory_length 2000, epsilon 0.00029435903709294965 total_time 722.0\n",
      "episode 18079, initial_state (0, 12, 2), reward 1504.0, memory_length 2000, epsilon 0.00029303784633474594 total_time 721.0\n",
      "episode 18089, initial_state (3, 3, 0), reward 1133.0, memory_length 2000, epsilon 0.0002917225875779408 total_time 726.0\n",
      "episode 18099, initial_state (2, 16, 0), reward 1584.0, memory_length 2000, epsilon 0.00029041323418849927 total_time 725.0\n",
      "episode 18109, initial_state (2, 5, 4), reward 1230.0, memory_length 2000, epsilon 0.0002891097596519713 total_time 730.0\n",
      "episode 18119, initial_state (2, 14, 4), reward 1342.0, memory_length 2000, epsilon 0.00028781213757295165 total_time 723.0\n",
      "episode 18129, initial_state (2, 16, 3), reward 1561.0, memory_length 2000, epsilon 0.00028652034167454956 total_time 723.0\n",
      "episode 18139, initial_state (3, 18, 5), reward 1521.0, memory_length 2000, epsilon 0.0002852343457978539 total_time 725.0\n",
      "episode 18149, initial_state (4, 9, 0), reward 1311.0, memory_length 2000, epsilon 0.00028395412390140427 total_time 726.0\n",
      "episode 18159, initial_state (4, 9, 2), reward 1147.0, memory_length 2000, epsilon 0.00028267965006066343 total_time 727.0\n",
      "episode 18169, initial_state (0, 1, 3), reward 1644.0, memory_length 2000, epsilon 0.00028141089846749263 total_time 731.0\n",
      "episode 18179, initial_state (3, 8, 1), reward 1472.0, memory_length 2000, epsilon 0.0002801478434296287 total_time 723.0\n",
      "episode 18189, initial_state (2, 15, 6), reward 1467.0, memory_length 2000, epsilon 0.000278890459370164 total_time 734.0\n",
      "episode 18199, initial_state (3, 16, 4), reward 1278.0, memory_length 2000, epsilon 0.00027763872082702836 total_time 722.0\n",
      "episode 18209, initial_state (1, 11, 1), reward 1464.0, memory_length 2000, epsilon 0.000276392602452474 total_time 722.0\n",
      "episode 18219, initial_state (0, 1, 0), reward 1449.0, memory_length 2000, epsilon 0.00027515207901256025 total_time 728.0\n",
      "episode 18229, initial_state (0, 7, 0), reward 1520.0, memory_length 2000, epsilon 0.00027391712538664544 total_time 721.0\n",
      "episode 18239, initial_state (4, 5, 2), reward 1511.0, memory_length 2000, epsilon 0.00027268771656687666 total_time 723.0\n",
      "episode 18249, initial_state (1, 11, 2), reward 858.0, memory_length 2000, epsilon 0.0002714638276576832 total_time 721.0\n",
      "episode 18259, initial_state (1, 23, 2), reward 1346.0, memory_length 2000, epsilon 0.00027024543387527285 total_time 722.0\n",
      "episode 18269, initial_state (1, 8, 6), reward 1416.0, memory_length 2000, epsilon 0.0002690325105471298 total_time 727.0\n",
      "episode 18279, initial_state (3, 19, 0), reward 1626.0, memory_length 2000, epsilon 0.0002678250331115153 total_time 721.0\n",
      "episode 18289, initial_state (4, 13, 4), reward 1579.0, memory_length 2000, epsilon 0.00026662297711696996 total_time 721.0\n",
      "episode 18299, initial_state (3, 2, 1), reward 1467.0, memory_length 2000, epsilon 0.0002654263182218188 total_time 722.0\n",
      "episode 18309, initial_state (2, 2, 6), reward 1420.0, memory_length 2000, epsilon 0.0002642350321936785 total_time 721.0\n",
      "episode 18319, initial_state (3, 6, 1), reward 1279.0, memory_length 2000, epsilon 0.0002630490949089664 total_time 735.0\n",
      "episode 18329, initial_state (2, 12, 1), reward 1590.0, memory_length 2000, epsilon 0.0002618684823524113 total_time 727.0\n",
      "episode 18339, initial_state (3, 18, 4), reward 1424.0, memory_length 2000, epsilon 0.00026069317061656893 total_time 724.0\n",
      "episode 18349, initial_state (1, 8, 6), reward 1427.0, memory_length 2000, epsilon 0.0002595231359013365 total_time 730.0\n",
      "episode 18359, initial_state (3, 18, 4), reward 1242.0, memory_length 2000, epsilon 0.0002583583545134711 total_time 721.0\n",
      "episode 18369, initial_state (4, 22, 3), reward 1446.0, memory_length 2000, epsilon 0.0002571988028661097 total_time 722.0\n",
      "episode 18379, initial_state (0, 7, 4), reward 1172.0, memory_length 2000, epsilon 0.000256044457478292 total_time 729.0\n",
      "episode 18389, initial_state (4, 2, 4), reward 1096.0, memory_length 2000, epsilon 0.0002548952949744842 total_time 721.0\n",
      "episode 18399, initial_state (3, 23, 6), reward 1526.0, memory_length 2000, epsilon 0.0002537512920841066 total_time 726.0\n",
      "episode 18409, initial_state (2, 12, 5), reward 1562.0, memory_length 2000, epsilon 0.0002526124256410614 total_time 721.0\n",
      "episode 18419, initial_state (3, 10, 4), reward 1778.0, memory_length 2000, epsilon 0.0002514786725832647 total_time 728.0\n",
      "episode 18429, initial_state (2, 23, 2), reward 1725.0, memory_length 2000, epsilon 0.0002503500099521774 total_time 721.0\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "episode 18439, initial_state (3, 9, 6), reward 1287.0, memory_length 2000, epsilon 0.00024922641489234326 total_time 727.0\n",
      "episode 18449, initial_state (0, 12, 4), reward 1593.0, memory_length 2000, epsilon 0.0002481078646509238 total_time 728.0\n",
      "episode 18459, initial_state (0, 2, 2), reward 1236.0, memory_length 2000, epsilon 0.00024699433657723836 total_time 729.0\n",
      "episode 18469, initial_state (2, 2, 0), reward 1375.0, memory_length 2000, epsilon 0.00024588580812230545 total_time 726.0\n",
      "episode 18479, initial_state (1, 7, 6), reward 1566.0, memory_length 2000, epsilon 0.00024478225683838604 total_time 725.0\n",
      "episode 18489, initial_state (2, 19, 1), reward 1520.0, memory_length 2000, epsilon 0.00024368366037852884 total_time 721.0\n",
      "episode 18499, initial_state (3, 6, 1), reward 1427.0, memory_length 2000, epsilon 0.00024258999649611802 total_time 732.0\n",
      "episode 18509, initial_state (4, 17, 1), reward 1676.0, memory_length 2000, epsilon 0.0002415012430444226 total_time 722.0\n",
      "episode 18519, initial_state (1, 20, 6), reward 1657.0, memory_length 2000, epsilon 0.00024041737797614836 total_time 722.0\n",
      "episode 18529, initial_state (1, 10, 0), reward 1602.0, memory_length 2000, epsilon 0.00023933837934298985 total_time 729.0\n",
      "episode 18539, initial_state (1, 3, 5), reward 1134.0, memory_length 2000, epsilon 0.00023826422529518823 total_time 734.0\n",
      "episode 18549, initial_state (3, 2, 2), reward 1661.0, memory_length 2000, epsilon 0.0002371948940810874 total_time 723.0\n",
      "episode 18559, initial_state (0, 8, 6), reward 1181.0, memory_length 2000, epsilon 0.0002361303640466937 total_time 723.0\n",
      "episode 18569, initial_state (4, 22, 0), reward 1638.0, memory_length 2000, epsilon 0.00023507061363523752 total_time 722.0\n",
      "episode 18579, initial_state (2, 7, 5), reward 1290.0, memory_length 2000, epsilon 0.00023401562138673687 total_time 726.0\n",
      "episode 18589, initial_state (2, 17, 6), reward 1300.0, memory_length 2000, epsilon 0.0002329653659375626 total_time 721.0\n",
      "episode 18599, initial_state (4, 6, 4), reward 1564.0, memory_length 2000, epsilon 0.00023191982602000606 total_time 722.0\n",
      "episode 18609, initial_state (4, 6, 4), reward 1350.0, memory_length 2000, epsilon 0.0002308789804618481 total_time 734.0\n",
      "episode 18619, initial_state (2, 12, 3), reward 1287.0, memory_length 2000, epsilon 0.00022984280818593073 total_time 726.0\n",
      "episode 18629, initial_state (2, 16, 3), reward 1744.0, memory_length 2000, epsilon 0.0002288112882097302 total_time 726.0\n",
      "episode 18639, initial_state (3, 20, 0), reward 1175.0, memory_length 2000, epsilon 0.00022778439964493105 total_time 727.0\n",
      "episode 18649, initial_state (2, 8, 0), reward 1757.0, memory_length 2000, epsilon 0.00022676212169700513 total_time 725.0\n",
      "episode 18659, initial_state (0, 9, 1), reward 1206.0, memory_length 2000, epsilon 0.000225744433664789 total_time 723.0\n",
      "episode 18669, initial_state (2, 8, 4), reward 1519.0, memory_length 2000, epsilon 0.00022473131494006533 total_time 721.0\n",
      "episode 18679, initial_state (2, 20, 2), reward 1565.0, memory_length 2000, epsilon 0.00022372274500714526 total_time 722.0\n",
      "episode 18689, initial_state (3, 0, 1), reward 1666.0, memory_length 2000, epsilon 0.0002227187034424532 total_time 725.0\n",
      "episode 18699, initial_state (3, 4, 6), reward 1275.0, memory_length 2000, epsilon 0.00022171916991411313 total_time 732.0\n",
      "episode 18709, initial_state (4, 20, 6), reward 968.0, memory_length 2000, epsilon 0.00022072412418153698 total_time 721.0\n",
      "episode 18719, initial_state (3, 9, 5), reward 1179.0, memory_length 2000, epsilon 0.00021973354609501465 total_time 731.0\n",
      "episode 18729, initial_state (0, 3, 0), reward 1526.0, memory_length 2000, epsilon 0.00021874741559530644 total_time 724.0\n",
      "episode 18739, initial_state (3, 6, 0), reward 1643.0, memory_length 2000, epsilon 0.00021776571271323524 total_time 732.0\n",
      "episode 18749, initial_state (3, 9, 4), reward 1522.0, memory_length 2000, epsilon 0.00021678841756928455 total_time 726.0\n",
      "episode 18759, initial_state (2, 22, 3), reward 1199.0, memory_length 2000, epsilon 0.00021581551037319426 total_time 723.0\n",
      "episode 18769, initial_state (0, 13, 5), reward 1455.0, memory_length 2000, epsilon 0.00021484697142356047 total_time 727.0\n",
      "episode 18779, initial_state (0, 6, 6), reward 1440.0, memory_length 2000, epsilon 0.00021388278110743628 total_time 729.0\n",
      "episode 18789, initial_state (0, 16, 0), reward 1541.0, memory_length 2000, epsilon 0.00021292291989993492 total_time 722.0\n",
      "episode 18799, initial_state (4, 11, 0), reward 1315.0, memory_length 2000, epsilon 0.00021196736836383406 total_time 725.0\n",
      "episode 18809, initial_state (3, 13, 3), reward 1531.0, memory_length 2000, epsilon 0.0002110161071491825 total_time 729.0\n",
      "episode 18819, initial_state (0, 9, 3), reward 1665.0, memory_length 2000, epsilon 0.0002100691169929081 total_time 728.0\n",
      "episode 18829, initial_state (2, 0, 1), reward 1496.0, memory_length 2000, epsilon 0.0002091263787184279 total_time 727.0\n",
      "episode 18839, initial_state (1, 15, 0), reward 1474.0, memory_length 2000, epsilon 0.00020818787323525988 total_time 724.0\n",
      "episode 18849, initial_state (2, 14, 4), reward 1461.0, memory_length 2000, epsilon 0.00020725358153863527 total_time 721.0\n",
      "episode 18859, initial_state (3, 12, 3), reward 1512.0, memory_length 2000, epsilon 0.00020632348470911566 total_time 722.0\n",
      "episode 18869, initial_state (0, 17, 1), reward 1395.0, memory_length 2000, epsilon 0.0002053975639122085 total_time 724.0\n",
      "episode 18879, initial_state (4, 23, 3), reward 1475.0, memory_length 2000, epsilon 0.00020447580039798592 total_time 723.0\n",
      "episode 18889, initial_state (3, 10, 5), reward 1235.0, memory_length 2000, epsilon 0.00020355817550070534 total_time 724.0\n",
      "episode 18899, initial_state (3, 23, 5), reward 1448.0, memory_length 2000, epsilon 0.00020264467063843116 total_time 722.0\n",
      "episode 18909, initial_state (4, 18, 0), reward 1391.0, memory_length 2000, epsilon 0.00020173526731265878 total_time 727.0\n",
      "episode 18919, initial_state (3, 22, 1), reward 1256.0, memory_length 2000, epsilon 0.0002008299471079397 total_time 726.0\n",
      "episode 18929, initial_state (0, 1, 4), reward 1133.0, memory_length 2000, epsilon 0.00019992869169150893 total_time 721.0\n",
      "episode 18939, initial_state (3, 21, 3), reward 1341.0, memory_length 2000, epsilon 0.00019903148281291377 total_time 728.0\n",
      "episode 18949, initial_state (4, 6, 1), reward 1549.0, memory_length 2000, epsilon 0.00019813830230364308 total_time 725.0\n",
      "episode 18959, initial_state (2, 9, 2), reward 1342.0, memory_length 2000, epsilon 0.0001972491320767614 total_time 727.0\n",
      "episode 18969, initial_state (1, 11, 5), reward 1366.0, memory_length 2000, epsilon 0.0001963639541265412 total_time 725.0\n",
      "episode 18979, initial_state (4, 11, 6), reward 1484.0, memory_length 2000, epsilon 0.0001954827505280988 total_time 726.0\n",
      "episode 18989, initial_state (0, 0, 2), reward 1378.0, memory_length 2000, epsilon 0.0001946055034370312 total_time 726.0\n",
      "episode 18999, initial_state (4, 13, 6), reward 1153.0, memory_length 2000, epsilon 0.0001937321950890548 total_time 722.0\n",
      "Saving Model 19000\n",
      "Saving Model 19000 Weights\n",
      "episode 19009, initial_state (4, 23, 2), reward 1586.0, memory_length 2000, epsilon 0.00019286280779964578 total_time 722.0\n",
      "episode 19019, initial_state (1, 17, 6), reward 1722.0, memory_length 2000, epsilon 0.0001919973239636817 total_time 721.0\n",
      "episode 19029, initial_state (0, 11, 4), reward 1588.0, memory_length 2000, epsilon 0.00019113572605508543 total_time 721.0\n",
      "episode 19039, initial_state (4, 10, 4), reward 1271.0, memory_length 2000, epsilon 0.00019027799662647016 total_time 723.0\n",
      "episode 19049, initial_state (3, 20, 1), reward 1473.0, memory_length 2000, epsilon 0.000189424118308785 total_time 723.0\n",
      "episode 19059, initial_state (1, 5, 0), reward 1552.0, memory_length 2000, epsilon 0.00018857407381096513 total_time 721.0\n",
      "episode 19069, initial_state (1, 7, 2), reward 1369.0, memory_length 2000, epsilon 0.00018772784591958045 total_time 727.0\n",
      "episode 19079, initial_state (4, 11, 5), reward 1324.0, memory_length 2000, epsilon 0.00018688541749848725 total_time 727.0\n",
      "episode 19089, initial_state (4, 23, 4), reward 1206.0, memory_length 2000, epsilon 0.0001860467714884812 total_time 728.0\n",
      "episode 19099, initial_state (0, 19, 6), reward 1617.0, memory_length 2000, epsilon 0.00018521189090695196 total_time 722.0\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "episode 19109, initial_state (3, 20, 6), reward 1539.0, memory_length 2000, epsilon 0.0001843807588475392 total_time 729.0\n",
      "episode 19119, initial_state (4, 15, 6), reward 1300.0, memory_length 2000, epsilon 0.00018355335847979032 total_time 721.0\n",
      "episode 19129, initial_state (1, 17, 0), reward 1423.0, memory_length 2000, epsilon 0.0001827296730488196 total_time 722.0\n",
      "episode 19139, initial_state (4, 18, 4), reward 1595.0, memory_length 2000, epsilon 0.00018190968587496894 total_time 724.0\n",
      "episode 19149, initial_state (4, 19, 4), reward 1284.0, memory_length 2000, epsilon 0.00018109338035347033 total_time 727.0\n",
      "episode 19159, initial_state (1, 7, 6), reward 1452.0, memory_length 2000, epsilon 0.00018028073995410846 total_time 723.0\n",
      "episode 19169, initial_state (4, 11, 0), reward 1342.0, memory_length 2000, epsilon 0.00017947174822088778 total_time 722.0\n",
      "episode 19179, initial_state (0, 2, 5), reward 1187.0, memory_length 2000, epsilon 0.00017866638877169806 total_time 721.0\n",
      "episode 19189, initial_state (2, 1, 6), reward 1557.0, memory_length 2000, epsilon 0.0001778646452979829 total_time 726.0\n",
      "episode 19199, initial_state (2, 17, 4), reward 1297.0, memory_length 2000, epsilon 0.00017706650156440958 total_time 726.0\n",
      "episode 19209, initial_state (1, 12, 3), reward 1236.0, memory_length 2000, epsilon 0.00017627194140854021 total_time 722.0\n",
      "episode 19219, initial_state (4, 5, 1), reward 1026.0, memory_length 2000, epsilon 0.0001754809487405045 total_time 725.0\n",
      "episode 19229, initial_state (4, 6, 0), reward 1392.0, memory_length 2000, epsilon 0.00017469350754267392 total_time 723.0\n",
      "episode 19239, initial_state (1, 5, 6), reward 1564.0, memory_length 2000, epsilon 0.00017390960186933725 total_time 721.0\n",
      "episode 19249, initial_state (0, 8, 5), reward 1604.0, memory_length 2000, epsilon 0.00017312921584637813 total_time 721.0\n",
      "episode 19259, initial_state (4, 22, 5), reward 1563.0, memory_length 2000, epsilon 0.00017235233367095232 total_time 724.0\n",
      "episode 19269, initial_state (4, 9, 1), reward 1472.0, memory_length 2000, epsilon 0.00017157893961116955 total_time 729.0\n",
      "episode 19279, initial_state (3, 20, 0), reward 1488.0, memory_length 2000, epsilon 0.00017080901800577367 total_time 721.0\n",
      "episode 19289, initial_state (3, 18, 3), reward 1590.0, memory_length 2000, epsilon 0.00017004255326382586 total_time 723.0\n",
      "episode 19299, initial_state (3, 0, 6), reward 1954.0, memory_length 2000, epsilon 0.0001692795298643889 total_time 722.0\n",
      "episode 19309, initial_state (1, 14, 2), reward 1423.0, memory_length 2000, epsilon 0.00016851993235621288 total_time 724.0\n",
      "episode 19319, initial_state (3, 5, 1), reward 1296.0, memory_length 2000, epsilon 0.0001677637453574223 total_time 727.0\n",
      "episode 19329, initial_state (3, 5, 6), reward 1576.0, memory_length 2000, epsilon 0.0001670109535552046 total_time 733.0\n",
      "episode 19339, initial_state (4, 3, 3), reward 1409.0, memory_length 2000, epsilon 0.00016626154170550005 total_time 728.0\n",
      "episode 19349, initial_state (4, 7, 4), reward 1578.0, memory_length 2000, epsilon 0.0001655154946326931 total_time 725.0\n",
      "episode 19359, initial_state (0, 9, 4), reward 1270.0, memory_length 2000, epsilon 0.00016477279722930532 total_time 723.0\n",
      "episode 19369, initial_state (1, 9, 0), reward 1415.0, memory_length 2000, epsilon 0.00016403343445568834 total_time 723.0\n",
      "episode 19379, initial_state (3, 4, 0), reward 1370.0, memory_length 2000, epsilon 0.00016329739133972096 total_time 724.0\n",
      "episode 19389, initial_state (2, 8, 2), reward 1519.0, memory_length 2000, epsilon 0.00016256465297650498 total_time 722.0\n",
      "episode 19399, initial_state (3, 5, 6), reward 1658.0, memory_length 2000, epsilon 0.00016183520452806355 total_time 725.0\n",
      "episode 19409, initial_state (4, 21, 6), reward 1092.0, memory_length 2000, epsilon 0.00016110903122304053 total_time 721.0\n",
      "episode 19419, initial_state (4, 3, 0), reward 1464.0, memory_length 2000, epsilon 0.00016038611835640182 total_time 730.0\n",
      "episode 19429, initial_state (4, 22, 3), reward 1328.0, memory_length 2000, epsilon 0.00015966645128913712 total_time 724.0\n",
      "episode 19439, initial_state (1, 22, 0), reward 1670.0, memory_length 2000, epsilon 0.00015895001544796367 total_time 728.0\n",
      "episode 19449, initial_state (0, 12, 3), reward 1692.0, memory_length 2000, epsilon 0.00015823679632503127 total_time 730.0\n",
      "episode 19459, initial_state (1, 11, 5), reward 1446.0, memory_length 2000, epsilon 0.00015752677947762858 total_time 728.0\n",
      "episode 19469, initial_state (1, 2, 3), reward 1328.0, memory_length 2000, epsilon 0.0001568199505278896 total_time 721.0\n",
      "episode 19479, initial_state (1, 0, 0), reward 1368.0, memory_length 2000, epsilon 0.0001561162951625042 total_time 725.0\n",
      "episode 19489, initial_state (0, 8, 0), reward 1564.0, memory_length 2000, epsilon 0.00015541579913242723 total_time 722.0\n",
      "episode 19499, initial_state (4, 18, 5), reward 1584.0, memory_length 2000, epsilon 0.0001547184482525901 total_time 730.0\n",
      "episode 19509, initial_state (3, 23, 5), reward 1443.0, memory_length 2000, epsilon 0.00015402422840161373 total_time 722.0\n",
      "episode 19519, initial_state (3, 3, 5), reward 1469.0, memory_length 2000, epsilon 0.00015333312552152237 total_time 724.0\n",
      "episode 19529, initial_state (0, 4, 4), reward 1547.0, memory_length 2000, epsilon 0.0001526451256174591 total_time 724.0\n",
      "episode 19539, initial_state (2, 10, 1), reward 1428.0, memory_length 2000, epsilon 0.00015196021475740232 total_time 725.0\n",
      "episode 19549, initial_state (3, 13, 4), reward 1337.0, memory_length 2000, epsilon 0.00015127837907188372 total_time 721.0\n",
      "episode 19559, initial_state (0, 8, 3), reward 1314.0, memory_length 2000, epsilon 0.00015059960475370765 total_time 722.0\n",
      "episode 19569, initial_state (0, 2, 2), reward 1678.0, memory_length 2000, epsilon 0.00014992387805767043 total_time 721.0\n",
      "episode 19579, initial_state (2, 13, 2), reward 1371.0, memory_length 2000, epsilon 0.00014925118530028367 total_time 721.0\n",
      "episode 19589, initial_state (4, 22, 3), reward 1584.0, memory_length 2000, epsilon 0.000148581512859496 total_time 721.0\n",
      "episode 19599, initial_state (2, 9, 4), reward 1584.0, memory_length 2000, epsilon 0.00014791484717441763 total_time 724.0\n",
      "episode 19609, initial_state (3, 12, 2), reward 1770.0, memory_length 2000, epsilon 0.00014725117474504565 total_time 725.0\n",
      "episode 19619, initial_state (4, 10, 1), reward 1188.0, memory_length 2000, epsilon 0.00014659048213199073 total_time 725.0\n",
      "episode 19629, initial_state (0, 10, 0), reward 1127.0, memory_length 2000, epsilon 0.0001459327559562048 total_time 723.0\n",
      "episode 19639, initial_state (2, 16, 0), reward 1584.0, memory_length 2000, epsilon 0.0001452779828987104 total_time 724.0\n",
      "episode 19649, initial_state (0, 16, 3), reward 1293.0, memory_length 2000, epsilon 0.00014462614970033066 total_time 723.0\n",
      "episode 19659, initial_state (0, 21, 4), reward 1385.0, memory_length 2000, epsilon 0.00014397724316142111 total_time 721.0\n",
      "episode 19669, initial_state (1, 7, 3), reward 1671.0, memory_length 2000, epsilon 0.00014333125014160238 total_time 725.0\n",
      "episode 19679, initial_state (3, 19, 0), reward 1733.0, memory_length 2000, epsilon 0.00014268815755949328 total_time 727.0\n",
      "episode 19689, initial_state (2, 21, 6), reward 1320.0, memory_length 2000, epsilon 0.00014204795239244723 total_time 731.0\n",
      "episode 19699, initial_state (0, 18, 5), reward 1477.0, memory_length 2000, epsilon 0.00014141062167628778 total_time 724.0\n",
      "episode 19709, initial_state (3, 8, 4), reward 1233.0, memory_length 2000, epsilon 0.00014077615250504612 total_time 724.0\n",
      "episode 19719, initial_state (4, 5, 0), reward 1235.0, memory_length 2000, epsilon 0.00014014453203069983 total_time 724.0\n",
      "episode 19729, initial_state (2, 11, 3), reward 1360.0, memory_length 2000, epsilon 0.0001395157474629128 total_time 727.0\n",
      "episode 19739, initial_state (3, 6, 2), reward 1370.0, memory_length 2000, epsilon 0.00013888978606877597 total_time 722.0\n",
      "episode 19749, initial_state (0, 1, 3), reward 1406.0, memory_length 2000, epsilon 0.00013826663517254976 total_time 726.0\n",
      "episode 19759, initial_state (4, 15, 5), reward 1309.0, memory_length 2000, epsilon 0.00013764628215540717 total_time 722.0\n",
      "episode 19769, initial_state (1, 2, 0), reward 1404.0, memory_length 2000, epsilon 0.00013702871445517872 total_time 728.0\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "episode 19779, initial_state (4, 14, 2), reward 1539.0, memory_length 2000, epsilon 0.00013641391956609685 total_time 730.0\n",
      "episode 19789, initial_state (4, 19, 6), reward 1350.0, memory_length 2000, epsilon 0.0001358018850385443 total_time 731.0\n",
      "episode 19799, initial_state (2, 2, 0), reward 1586.0, memory_length 2000, epsilon 0.00013519259847880098 total_time 723.0\n",
      "episode 19809, initial_state (4, 18, 6), reward 1689.0, memory_length 2000, epsilon 0.00013458604754879323 total_time 723.0\n",
      "episode 19819, initial_state (3, 7, 6), reward 1464.0, memory_length 2000, epsilon 0.00013398221996584395 total_time 723.0\n",
      "episode 19829, initial_state (2, 2, 0), reward 1489.0, memory_length 2000, epsilon 0.00013338110350242404 total_time 723.0\n",
      "episode 19839, initial_state (2, 0, 0), reward 1121.0, memory_length 2000, epsilon 0.0001327826859859045 total_time 722.0\n",
      "episode 19849, initial_state (4, 4, 1), reward 1287.0, memory_length 2000, epsilon 0.0001321869552983102 total_time 729.0\n",
      "episode 19859, initial_state (3, 13, 6), reward 1709.0, memory_length 2000, epsilon 0.00013159389937607432 total_time 722.0\n",
      "episode 19869, initial_state (3, 7, 2), reward 1699.0, memory_length 2000, epsilon 0.00013100350620979422 total_time 721.0\n",
      "episode 19879, initial_state (4, 20, 6), reward 1762.0, memory_length 2000, epsilon 0.00013041576384398832 total_time 729.0\n",
      "episode 19889, initial_state (4, 19, 1), reward 1579.0, memory_length 2000, epsilon 0.00012983066037685319 total_time 721.0\n",
      "episode 19899, initial_state (3, 4, 2), reward 1208.0, memory_length 2000, epsilon 0.0001292481839600238 total_time 728.0\n",
      "episode 19909, initial_state (4, 13, 6), reward 1653.0, memory_length 2000, epsilon 0.00012866832279833285 total_time 724.0\n",
      "episode 19919, initial_state (3, 22, 0), reward 1567.0, memory_length 2000, epsilon 0.00012809106514957203 total_time 728.0\n",
      "episode 19929, initial_state (2, 23, 4), reward 1377.0, memory_length 2000, epsilon 0.00012751639932425413 total_time 723.0\n",
      "episode 19939, initial_state (0, 20, 3), reward 1602.0, memory_length 2000, epsilon 0.00012694431368537662 total_time 726.0\n",
      "episode 19949, initial_state (1, 22, 0), reward 1522.0, memory_length 2000, epsilon 0.00012637479664818577 total_time 726.0\n",
      "episode 19959, initial_state (4, 14, 3), reward 1506.0, memory_length 2000, epsilon 0.0001258078366799421 total_time 722.0\n",
      "episode 19969, initial_state (4, 6, 3), reward 1617.0, memory_length 2000, epsilon 0.00012524342229968684 total_time 726.0\n",
      "episode 19979, initial_state (0, 9, 2), reward 1692.0, memory_length 2000, epsilon 0.00012468154207800978 total_time 726.0\n",
      "episode 19989, initial_state (1, 14, 0), reward 1435.0, memory_length 2000, epsilon 0.00012412218463681677 total_time 723.0\n",
      "episode 19999, initial_state (4, 20, 3), reward 1461.0, memory_length 2000, epsilon 0.00012356533864910072 total_time 722.0\n",
      "11725.39568567276\n"
     ]
    }
   ],
   "source": [
    "start_time = time.time()\n",
    "score_tracked = []\n",
    "\n",
    "for episode in range(Episodes):\n",
    "\n",
    "    done = False\n",
    "    score = 0\n",
    "    track_reward = False\n",
    "\n",
    "    # Reset at the start of each episode\n",
    "    env = CabDriver()\n",
    "    action_space, state_space, state = env.reset()\n",
    "    # State Initialization \n",
    "    initial_state = state\n",
    "\n",
    "\n",
    "    total_time = 0  # Total time driver rode in this episode\n",
    "    while done!= True:\n",
    "        # 1. Pick epsilon-greedy action from possible actions for the current state.\n",
    "        action= agent.get_action(state)\n",
    "        \n",
    "        # 2. Evaluate your reward and next state\n",
    "        next_state, reward, step_time = env.step(state, env.action_space[action], Time_matrix)\n",
    "        \n",
    "        # 3. Total time driver rode in this episode\n",
    "        total_time += step_time\n",
    "        if (total_time > episode_time):\n",
    "            # The cab driver accepts the last ride prior to the end of total time limit (720 hours). \n",
    "            # So the last trip begins before the end of 720 hrs but the total time of episode might cross 720 hrs.\n",
    "            done = True\n",
    "        else:\n",
    "            # 4. Append the experience to the memory\n",
    "            agent.append_sample(state, action, reward, next_state, done)   ## Note: Here action is action index\n",
    "            # 5. Train the model by calling function agent.train_model\n",
    "            agent.train_model()\n",
    "            # 6. Keep a track of rewards, Q-values, loss, etc (Note: Loss is taken care of in the model loss='mse')\n",
    "            score += reward\n",
    "            state = next_state\n",
    "\n",
    "    # Store total reward obtained in this episode\n",
    "    rewards_per_episode.append(score)\n",
    "    episodes.append(episode)\n",
    "    \n",
    "\n",
    "    # Epsilon-decay\n",
    "    agent.epsilon = agent.epsilon_min + (agent.epsilon_max - agent.epsilon_min) * np.exp(agent.epsilon_decay * episode)\n",
    "\n",
    "    # Every 10 episodes:\n",
    "    if ((episode + 1) % 10 == 0):\n",
    "        print(\"episode {0}, initial_state {1}, reward {2}, memory_length {3}, epsilon {4} total_time {5}\".format(episode, \n",
    "                                                                         initial_state,\n",
    "                                                                         score,\n",
    "                                                                         len(agent.memory),\n",
    "                                                                         agent.epsilon, total_time))\n",
    "    # Save the Q_value of the state-action pair we are tracking (every 5 episodes)\n",
    "    if ((episode + 1) % 5 == 0):\n",
    "        agent.save_tracking_states()\n",
    "\n",
    "    # Total rewards per episode\n",
    "    score_tracked.append(score)\n",
    "\n",
    "    ## Saving the 'DQN_model' and 'model_weights' every 1000th episode.\n",
    "    if(episode % 1000 == 0):\n",
    "        print(\"Saving Model {}\".format(episode))\n",
    "        agent.save(name=\"DQN_model.h5\")                     ## Saves DQN model in Keras H5 format\n",
    "        print(\"Saving Model {} Weights\".format(episode))\n",
    "        agent.save_weights_numpy(name=\"model_weights.pkl\")  ## Saves model_weights in pkl file (model_weights pickle file has a list of numpy arrays)   \n",
    "    \n",
    "elapsed_time = time.time() - start_time\n",
    "print(elapsed_time)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "The episode total time limit is `720 i.e. 24*30 hours`. However, we see the total time at the end of each episode crosses 720 hours limit. This is because the cab driver accepts the last ride prior to the end of total time (720 hours). Hence, the cab driver has to complete the last trip in order to complete the episode."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Saving DQN_Model and Model Weights"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [],
   "source": [
    "## Saves the complete DQN model in Keras h5 format\n",
    "agent.save(name=\"DQN_model.h5\")\n",
    "\n",
    "## Saves the model_weights as list of Numpy arrays\n",
    "agent.save_weights_numpy(name=\"model_weights.pkl\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "[array([[ 9.25591469e+00,  7.18056202e-01,  1.26659727e+00,\n",
       "         -3.70899057e+00, -2.30337930e+00, -4.11312771e+00,\n",
       "         -2.69715905e+00, -4.63396597e+00, -3.86690497e+00,\n",
       "         -3.00464845e+00,  4.56464261e-01, -3.21279621e+00,\n",
       "         -1.94325662e+00, -3.52571034e+00,  5.16652703e-01,\n",
       "         -2.82427621e+00, -7.05090523e+00,  1.27863562e+00,\n",
       "         -4.77406549e+00, -8.73321772e-01, -5.37756252e+00,\n",
       "          2.43566918e+00, -8.23643112e+00, -2.97834754e+00,\n",
       "         -1.94208062e+00,  1.48330140e+00, -3.78979778e+00,\n",
       "         -1.56868562e-01, -2.35054278e+00, -1.49977601e+00,\n",
       "         -1.59945107e+00, -3.76700425e+00],\n",
       "        [-5.69953918e+00, -2.75832462e+00, -5.49629688e-01,\n",
       "         -3.70714188e+00, -5.50291824e+00,  2.20942068e+00,\n",
       "         -3.00161028e+00, -4.30807638e+00, -1.29523981e+00,\n",
       "         -3.87663746e+00, -6.07029438e+00,  4.21938286e+01,\n",
       "         -1.29401755e+00, -5.01755047e+00, -3.08870268e+00,\n",
       "         -2.88319373e+00, -7.22745705e+00, -1.27353680e+00,\n",
       "          5.92504501e-01, -7.41126716e-01,  1.22084606e+00,\n",
       "         -3.54859638e+00, -5.30733824e+00,  8.29949188e+00,\n",
       "         -4.84683084e+00, -3.40854001e+00, -4.05702257e+00,\n",
       "         -4.79611969e+00, -4.83609486e+00, -3.87685800e+00,\n",
       "         -4.56436920e+00, -7.86749220e+00],\n",
       "        [-4.96280861e+00, -2.81346369e+00, -2.55869126e+00,\n",
       "         -3.82194543e+00, -5.33552885e+00,  2.31084538e+00,\n",
       "         -7.54936504e+00,  2.71563888e+00,  1.27907646e+00,\n",
       "          5.47897458e-01, -5.44859505e+00, -4.74577665e+00,\n",
       "         -6.08853292e+00,  1.87013447e+00, -2.60691857e+00,\n",
       "         -5.86375475e+00, -3.79000235e+00, -2.25769782e+00,\n",
       "         -4.55046326e-01, -7.44872153e-01, -5.53614974e-01,\n",
       "         -4.89958048e+00, -4.58069038e+00, -4.24922276e+00,\n",
       "          4.64863253e+00, -5.45354319e+00, -3.22260165e+00,\n",
       "         -4.82439470e+00,  4.87570107e-01, -3.30382442e+00,\n",
       "         -7.20774126e+00, -6.20445919e+00],\n",
       "        [-3.81770992e+00, -3.51739955e+00, -1.86483884e+00,\n",
       "         -3.34840989e+00, -2.89897132e+00, -4.34420443e+00,\n",
       "         -5.30048490e-01, -1.81191659e+00, -1.51405320e-01,\n",
       "         -3.38883471e+00,  5.68067551e-01, -2.95386291e+00,\n",
       "         -4.99234152e+00, -1.10803878e+00, -4.71215391e+00,\n",
       "          6.02592230e-01, -7.36361361e+00, -2.55471015e+00,\n",
       "          7.36453652e-01, -2.39177868e-01,  1.79831266e+00,\n",
       "         -4.85193157e+00, -5.40538359e+00,  1.20485754e+01,\n",
       "         -2.58185601e+00, -3.75144625e+00, -1.64711207e-01,\n",
       "         -4.90549231e+00, -2.83562660e+00, -4.41370821e+00,\n",
       "         -2.73682523e+00, -5.95624161e+00],\n",
       "        [ 8.81303501e+00, -3.08845949e+00, -2.77211308e+00,\n",
       "          2.76439898e-02, -5.00307465e+00,  1.78240299e+00,\n",
       "         -7.51850128e+00, -4.53553867e+00, -2.99988079e+00,\n",
       "         -2.57826471e+00, -5.82580853e+00, -3.07967687e+00,\n",
       "         -4.33826494e+00, -2.69260120e+00, -9.95677292e-01,\n",
       "         -1.07556677e+00, -6.72160530e+00, -1.17503250e+00,\n",
       "         -1.72320247e+00, -8.63504171e-01,  1.37556720e+00,\n",
       "         -4.18580580e+00, -4.32233524e+00,  4.32453424e-01,\n",
       "         -4.86780834e+00, -4.06400156e+00, -3.39371276e+00,\n",
       "         -1.96575522e+00, -4.35927057e+00,  4.53983635e-01,\n",
       "         -4.99161339e+00, -5.20954418e+00],\n",
       "        [-9.95260525e+00, -1.89592516e+00, -2.08087102e-01,\n",
       "         -2.72763491e+00, -9.51435471e+00, -4.12933445e+00,\n",
       "         -5.26836491e+00, -3.10734820e+00, -2.81744814e+00,\n",
       "          4.37424421e+00, -5.51724815e+00, -4.96503983e+01,\n",
       "         -2.07292140e-01, -2.78680992e+00,  1.81390691e+00,\n",
       "         -2.04545712e+00, -4.89039898e-01,  2.37664390e+00,\n",
       "         -1.58513355e+00, -5.40924835e+00, -7.38962889e+00,\n",
       "         -3.07469583e+00, -2.70210242e+00, -1.91139908e+01,\n",
       "         -1.40976162e+01, -5.27805662e+00, -3.48842573e+00,\n",
       "         -2.96602774e+00, -2.67218304e+00, -9.77757633e-01,\n",
       "          3.59502602e+00,  9.58322644e-01],\n",
       "        [-1.00759659e+01, -2.58922815e+00,  1.78658843e-01,\n",
       "         -4.73483771e-01, -9.91822147e+00, -4.06466293e+00,\n",
       "         -5.32377911e+00, -3.00897670e+00, -2.13878417e+00,\n",
       "         -1.38109601e+00, -5.33769417e+00, -5.10953712e+01,\n",
       "         -2.49918401e-01, -3.32973862e+00, -2.11295485e+00,\n",
       "         -2.01613808e+00, -5.52221596e-01,  2.25000167e+00,\n",
       "         -1.24831629e+00, -4.60445595e+00, -7.82188034e+00,\n",
       "         -3.00096583e+00, -2.54034281e+00, -1.92820263e+01,\n",
       "         -1.49482803e+01, -4.01657677e+00, -3.99512261e-01,\n",
       "         -1.99393141e+00, -1.90814483e+00, -2.24262452e+00,\n",
       "          3.25832558e+00,  9.86498833e-01],\n",
       "        [-1.02542934e+01,  5.04562187e+00, -7.10145593e-01,\n",
       "         -1.95307434e-02, -9.85008049e+00, -4.06510019e+00,\n",
       "         -5.35824537e+00, -3.00647807e+00, -2.07467771e+00,\n",
       "          1.61656010e+00, -5.30935001e+00, -5.00785484e+01,\n",
       "         -2.41519332e+00, -2.59770322e+00, -2.28354311e+00,\n",
       "         -2.48067975e+00, -4.34377313e-01,  2.49414206e+00,\n",
       "          2.17707828e-02, -5.46612215e+00, -8.41596127e+00,\n",
       "         -2.64962792e+00, -3.30760932e+00, -1.94541664e+01,\n",
       "         -1.27110882e+01, -4.72354507e+00,  4.17507458e+00,\n",
       "         -2.05369377e+00, -2.26516271e+00, -7.25948930e-01,\n",
       "          5.02644777e+00,  9.25110102e-01],\n",
       "        [-1.00550613e+01, -8.40999782e-01,  9.61495042e-01,\n",
       "         -2.40132406e-01, -9.90637589e+00, -4.12170553e+00,\n",
       "         -5.15608311e+00, -2.41307712e+00, -2.25238705e+00,\n",
       "          4.01151752e+00, -4.93304873e+00, -4.91248665e+01,\n",
       "         -4.97213507e+00, -3.26673722e+00,  2.12449372e-01,\n",
       "         -2.71214771e+00, -7.83975899e-01,  1.39091277e+00,\n",
       "          1.04812574e+00, -5.03269148e+00, -7.35416412e+00,\n",
       "         -2.65052223e+00, -2.46702504e+00, -1.95247974e+01,\n",
       "         -1.04003010e+01, -5.36802197e+00,  3.79823804e+00,\n",
       "         -2.41358542e+00, -1.55731213e+00, -8.34869683e-01,\n",
       "          9.52527344e-01,  1.06655157e+00],\n",
       "        [-1.02496252e+01, -7.65914470e-02,  3.15459943e+00,\n",
       "         -4.06598598e-01, -9.63832855e+00, -4.05679083e+00,\n",
       "         -5.13956785e+00, -2.64526057e+00, -2.53864408e+00,\n",
       "          3.76719952e+00, -5.18496180e+00, -5.01724243e+01,\n",
       "         -2.56965303e+00, -1.70105898e+00, -1.49184048e+00,\n",
       "         -8.62702668e-01, -1.28397655e+00, -3.96611392e-01,\n",
       "          4.97106934e+00, -5.07345724e+00, -7.15066147e+00,\n",
       "         -3.09703112e+00, -2.55472136e+00, -1.96128445e+01,\n",
       "         -1.41171322e+01, -5.34569645e+00,  5.83527660e+00,\n",
       "         -2.31684780e+00, -2.70984936e+00, -2.27462435e+00,\n",
       "          1.27536631e+00,  1.04220688e+00],\n",
       "        [-1.01806793e+01, -1.53868115e+00, -1.05489480e+00,\n",
       "         -3.13986450e-01, -9.86408043e+00, -4.17832232e+00,\n",
       "         -5.03055191e+00, -2.52660513e+00, -2.62483478e+00,\n",
       "         -8.92109752e-01, -5.70082617e+00, -5.15672112e+01,\n",
       "         -6.67003512e-01, -2.53770542e+00, -2.17744184e+00,\n",
       "         -1.06923461e+00, -1.45279205e+00, -2.31097102e+00,\n",
       "          3.31162786e+00, -4.19729948e+00, -7.85497570e+00,\n",
       "         -2.60017920e+00, -2.95798850e+00, -1.89995251e+01,\n",
       "         -1.24364357e+01, -4.73920822e+00, -1.11347091e+00,\n",
       "         -2.12907362e+00, -2.06374812e+00,  8.96883130e-01,\n",
       "          6.26606083e+00,  9.91351962e-01],\n",
       "        [-6.35070181e+00, -2.19105506e+00, -1.79721522e+00,\n",
       "         -2.15133524e+00, -6.00509453e+00, -4.17202330e+00,\n",
       "         -5.08165693e+00, -6.31161833e+00,  8.31874371e-01,\n",
       "          5.78906679e+00, -4.75047302e+00, -5.04387817e+01,\n",
       "         -3.75432491e+00, -1.52832663e+00,  4.85181153e-01,\n",
       "         -4.16621774e-01, -2.26638341e+00, -1.32757640e+00,\n",
       "         -1.24467027e+00, -5.66837645e+00, -5.85257590e-01,\n",
       "         -1.42771423e+00,  1.60304868e+00, -1.98229046e+01,\n",
       "         -1.43138819e+01, -2.04234052e+00,  3.82286119e+00,\n",
       "          2.81307411e+00, -2.22237730e+00,  2.19194269e+00,\n",
       "         -7.37912655e-01, -8.70482922e+00],\n",
       "        [-6.41560364e+00, -2.99061489e+00, -8.59068453e-01,\n",
       "         -3.57476711e-01, -9.88094997e+00, -4.13490057e+00,\n",
       "         -5.23936844e+00, -6.73317432e+00,  9.54916596e-01,\n",
       "         -2.06481647e+00, -5.08189917e+00, -4.95049896e+01,\n",
       "         -3.52416372e+00, -1.35055614e+00, -1.12728716e-03,\n",
       "          1.45509863e+00, -3.30250931e+00,  7.83768594e-01,\n",
       "         -1.42304409e+00, -6.48847532e+00, -3.64488244e-01,\n",
       "         -2.93723083e+00,  1.43615651e+00, -2.00963268e+01,\n",
       "         -1.46058044e+01, -1.29909444e+00,  1.10026729e+00,\n",
       "          4.78888083e+00, -2.07950902e+00,  9.00351480e-02,\n",
       "         -2.89810717e-01, -5.30783224e+00],\n",
       "        [-5.75241613e+00, -2.48434424e+00, -1.61517239e+00,\n",
       "         -1.85702050e+00, -7.76436758e+00, -4.12818289e+00,\n",
       "         -5.00331306e+00, -6.48865986e+00, -2.04421425e+00,\n",
       "         -1.75662899e+00, -5.35998726e+00, -4.95140114e+01,\n",
       "         -4.23392725e+00, -1.70098114e+00,  2.85998917e+00,\n",
       "         -6.10773563e-01,  5.60012293e+00,  1.56053126e+00,\n",
       "         -4.98112738e-02, -6.99740934e+00, -4.24725860e-01,\n",
       "         -3.10188222e+00,  8.98892701e-01, -2.00294304e+01,\n",
       "         -1.40576696e+01, -5.01367426e+00,  2.89522290e+00,\n",
       "          6.45706415e-01, -1.98553228e+00,  5.21620321e+00,\n",
       "         -1.75182986e+00, -6.46154690e+00],\n",
       "        [-6.21800852e+00, -1.75289917e+00, -8.77958953e-01,\n",
       "         -3.44114572e-01, -5.47982407e+00, -4.04989004e+00,\n",
       "         -4.85591221e+00, -6.28684711e+00,  3.76171613e+00,\n",
       "         -1.53126132e+00, -5.35653782e+00, -4.92178307e+01,\n",
       "         -3.99296141e+00,  1.75003159e+00,  4.46225941e-01,\n",
       "          3.57285714e+00,  3.10627174e+00,  1.58926117e+00,\n",
       "         -7.39914119e-01, -5.84628344e+00, -1.51724949e-01,\n",
       "         -2.64924812e+00,  7.33193398e-01, -1.95002346e+01,\n",
       "         -1.46774492e+01, -5.16196585e+00, -3.41015697e-01,\n",
       "          5.32265472e+00, -1.02413106e+00, -1.00167763e+00,\n",
       "         -7.88312793e-01, -9.00614357e+00],\n",
       "        [-5.88377380e+00, -1.71377933e+00, -1.42619872e+00,\n",
       "         -5.60187101e-01, -6.03529787e+00, -4.20526600e+00,\n",
       "         -4.72860479e+00, -6.41563702e+00,  3.67093325e+00,\n",
       "         -7.61137664e-01, -5.39474344e+00, -4.84954300e+01,\n",
       "         -4.33917236e+00,  2.06734872e+00, -2.03490570e-01,\n",
       "          9.70608592e-01, -2.47965121e+00,  1.52243650e+00,\n",
       "         -1.71485427e-03, -6.13079977e+00, -7.51958191e-01,\n",
       "         -2.41400051e+00,  3.31337988e-01, -1.87410736e+01,\n",
       "         -1.47610455e+01, -3.61983633e+00,  1.29459098e-01,\n",
       "          1.28750765e+00, -7.74010956e-01, -1.97484636e+00,\n",
       "          2.74084002e-01, -4.29235840e+00],\n",
       "        [-6.50307178e+00, -1.45556104e+00, -2.23698139e-01,\n",
       "         -8.91564041e-02, -9.61987019e+00, -3.90389943e+00,\n",
       "         -5.10767031e+00, -6.48327780e+00,  5.91239357e+00,\n",
       "         -1.70326567e+00, -5.15552139e+00, -4.91370697e+01,\n",
       "         -3.99871039e+00,  1.27038193e+00, -1.96711493e+00,\n",
       "          6.44868279e+00, -3.30983162e+00, -1.60125792e-01,\n",
       "         -1.46741128e+00, -5.75623655e+00, -3.68694961e-01,\n",
       "         -2.45293331e+00,  3.36758912e-01, -1.84763851e+01,\n",
       "         -1.27305994e+01, -4.56961918e+00, -9.09323812e-01,\n",
       "          5.56069994e+00, -9.06049252e-01, -1.13780117e+00,\n",
       "         -1.02421951e+00, -6.35605478e+00],\n",
       "        [-1.00828743e+01, -1.79389906e+00, -2.17289791e-01,\n",
       "          6.07602119e-01, -2.87935424e+00,  7.62444675e-01,\n",
       "         -4.18403244e+00, -6.33913755e+00,  3.15403223e+00,\n",
       "         -9.01632369e-01, -5.09686804e+00, -5.50543671e+01,\n",
       "         -4.95423698e+00, -2.40184236e+00, -9.00177419e-01,\n",
       "         -2.80085158e+00, -3.16240764e+00, -2.37859368e+00,\n",
       "         -4.37239170e-01, -5.71737814e+00, -1.21272251e-01,\n",
       "         -3.33816028e+00, -2.39581037e+00, -1.92745380e+01,\n",
       "         -1.24070902e+01, -3.59227562e+00, -1.80176079e+00,\n",
       "         -1.24695692e-02, -2.29679060e+00, -1.74396074e+00,\n",
       "         -8.60365272e-01, -6.77055979e+00],\n",
       "        [-1.04295321e+01, -1.72715414e+00, -5.73342502e-01,\n",
       "         -9.90103364e-01, -2.03812671e+00,  6.54925644e-01,\n",
       "         -4.21790409e+00, -6.45867491e+00,  3.28214717e+00,\n",
       "         -7.07011223e-01, -5.93106747e+00, -5.51056671e+01,\n",
       "         -4.66860342e+00, -1.96899092e+00, -1.86694467e+00,\n",
       "         -3.11553335e+00, -1.60221589e+00, -2.29960752e+00,\n",
       "         -1.52535534e+00, -5.57262182e+00, -2.29600325e-01,\n",
       "         -3.43014836e+00, -2.69177532e+00, -1.95840092e+01,\n",
       "         -1.42711821e+01, -3.67013097e+00, -1.00025332e+00,\n",
       "         -1.80133462e+00, -1.45851481e+00, -1.99070263e+00,\n",
       "         -1.04795134e+00, -9.12407398e+00],\n",
       "        [-1.05302181e+01,  2.66084164e-01,  1.60118616e+00,\n",
       "         -1.25473523e+00, -2.24523139e+00,  8.64281654e-01,\n",
       "         -4.40119123e+00, -6.53056097e+00,  1.84674335e+00,\n",
       "         -1.25752878e+00, -5.75950480e+00, -5.53845406e+01,\n",
       "         -5.33525419e+00, -2.45909119e+00,  1.56772602e+00,\n",
       "         -2.44313049e+00, -3.21168089e+00, -2.92651939e+00,\n",
       "         -1.40663528e+00, -5.71350241e+00, -5.94144464e-01,\n",
       "         -3.17105103e+00, -3.00557256e+00, -1.92409420e+01,\n",
       "         -1.34555235e+01, -3.88696098e+00, -2.26225525e-01,\n",
       "         -6.67644799e-01, -7.52041936e-01, -1.87227452e+00,\n",
       "         -6.33023143e-01, -9.20871258e+00],\n",
       "        [-1.04178724e+01, -6.01003766e-01, -1.69753158e+00,\n",
       "         -2.07408130e-01, -3.04124570e+00,  7.33789325e-01,\n",
       "         -4.21302986e+00, -6.63249731e+00,  1.65153348e+00,\n",
       "          1.17706287e+00, -5.75931454e+00, -5.42012253e+01,\n",
       "         -4.88843203e+00, -1.82595718e+00, -1.81154478e+00,\n",
       "         -2.38006711e+00, -3.26351666e+00, -2.32746053e+00,\n",
       "         -1.64280701e+00, -5.51614571e+00, -7.34625220e-01,\n",
       "         -3.46920323e+00, -2.65934324e+00, -1.92208977e+01,\n",
       "         -1.26150999e+01, -9.56570506e-01,  6.78936839e-02,\n",
       "         -1.50510979e+00, -7.52483845e-01, -2.01815677e+00,\n",
       "         -2.02528644e+00, -8.07137775e+00],\n",
       "        [-9.56182098e+00,  4.91094887e-01, -1.56711996e+00,\n",
       "         -5.17987251e-01, -3.90820050e+00,  7.49984741e-01,\n",
       "         -4.24025011e+00, -6.56631374e+00,  4.11363840e+00,\n",
       "         -1.55023909e+00, -5.29534054e+00, -5.37399254e+01,\n",
       "         -4.83611345e+00, -9.61362839e-01, -2.01566577e+00,\n",
       "         -1.22454560e+00, -1.95474160e+00, -2.21345520e+00,\n",
       "         -1.43128562e+00, -4.84540987e+00, -8.76315594e-01,\n",
       "         -3.47335672e+00, -3.04174924e+00, -1.99441128e+01,\n",
       "         -1.22122841e+01, -1.18892944e+00, -1.09322250e+00,\n",
       "          9.67968345e-01, -7.30816901e-01, -1.71560541e-01,\n",
       "         -1.94060624e+00, -2.85320711e+00],\n",
       "        [-9.75172138e+00, -5.62885284e-01, -1.05030370e+00,\n",
       "         -9.80011463e-01, -3.80291390e+00,  5.97800791e-01,\n",
       "         -4.46511841e+00, -6.81210184e+00, -2.39955783e+00,\n",
       "         -3.14967453e-01, -5.85670900e+00, -5.31122932e+01,\n",
       "         -4.88532162e+00, -2.61810851e+00, -3.00510073e+00,\n",
       "         -2.70371866e+00, -3.09900093e+00, -1.59382284e+00,\n",
       "         -1.20302415e+00, -5.37368822e+00, -8.07888699e+00,\n",
       "         -3.26658916e+00, -2.53813434e+00, -2.03683815e+01,\n",
       "         -1.22541037e+01, -3.48716825e-02, -1.13496065e+00,\n",
       "          3.91589880e-01, -8.04199338e-01, -8.16832185e-01,\n",
       "         -1.73757684e+00, -7.38448381e+00],\n",
       "        [-1.07621613e+01,  2.66668630e+00, -2.37602010e-01,\n",
       "          4.57344580e+00, -3.03248072e+00, -4.08320189e+00,\n",
       "          8.01445723e-01, -6.30893040e+00, -1.13250279e+00,\n",
       "          2.48411202e+00, -9.75234985e-01, -5.15671425e+01,\n",
       "         -5.23454857e+00, -1.93991542e+00, -1.38963306e+00,\n",
       "         -4.73030001e-01, -3.39557147e+00,  4.42872375e-01,\n",
       "         -1.49565125e+00, -5.45355463e+00, -7.25693655e+00,\n",
       "          2.07769704e+00, -3.15586925e+00, -2.21761131e+01,\n",
       "         -1.07953587e+01, -4.66205883e+00, -1.96360111e+00,\n",
       "         -2.47703338e+00,  3.37966979e-01, -1.01551378e+00,\n",
       "         -1.65549505e+00, -6.94654655e+00],\n",
       "        [-9.94261551e+00,  2.50498557e+00, -1.78470576e+00,\n",
       "         -1.42405942e-01, -3.38550162e+00, -4.06209469e+00,\n",
       "         -1.17341173e+00, -6.40591621e+00, -1.28669739e+00,\n",
       "          2.46999812e+00, -7.76090384e-01, -5.13924599e+01,\n",
       "         -3.47008204e+00, -1.84012771e+00,  1.08707488e+00,\n",
       "          1.14751148e+00, -3.26278305e+00,  1.81904447e+00,\n",
       "         -1.34180307e+00, -5.45409536e+00, -8.18505478e+00,\n",
       "          1.97355199e+00, -3.19933558e+00, -2.16441689e+01,\n",
       "         -1.06248798e+01, -4.64194536e+00, -2.01862478e+00,\n",
       "         -2.77365065e+00,  3.63925552e+00, -9.82797623e-01,\n",
       "         -2.45691991e+00, -6.87517881e+00],\n",
       "        [-9.92126656e+00,  2.12676573e+00, -4.56821948e-01,\n",
       "          5.43979025e+00, -3.03684568e+00, -4.01503134e+00,\n",
       "         -2.30066180e+00, -6.40198946e+00, -2.33805943e+00,\n",
       "         -5.56189835e-01, -1.05797207e+00, -5.04392128e+01,\n",
       "         -4.90353775e+00, -1.83427417e+00, -1.42518389e+00,\n",
       "          1.32064540e-02, -3.24186087e+00, -1.39986682e+00,\n",
       "         -7.44982958e-01, -5.52087116e+00, -7.43360329e+00,\n",
       "          1.12394822e+00, -2.62912989e+00, -2.15961132e+01,\n",
       "         -9.70853233e+00, -5.19978905e+00, -3.02019978e+00,\n",
       "         -1.80506206e+00,  5.91105700e+00, -1.88491249e+00,\n",
       "         -3.03129292e+00, -7.19507742e+00],\n",
       "        [-1.00252924e+01,  2.67066836e+00,  1.09249055e+00,\n",
       "         -6.47043049e-01, -3.09841371e+00, -4.11863327e+00,\n",
       "         -2.54485583e+00, -6.55610085e+00, -1.02014506e+00,\n",
       "         -2.01778221e+00, -6.34515464e-01, -5.07573166e+01,\n",
       "         -4.69752741e+00, -1.54607916e+00,  2.05432820e+00,\n",
       "         -7.55788505e-01, -3.19786859e+00,  1.57272017e+00,\n",
       "         -1.53013659e+00, -5.37746429e+00, -6.42463160e+00,\n",
       "         -1.23718715e+00, -2.23327637e+00, -2.15428848e+01,\n",
       "         -8.34588909e+00, -5.38092661e+00, -2.68740511e+00,\n",
       "         -2.95825243e+00,  2.52072287e+00, -7.43485332e-01,\n",
       "         -3.12871766e+00, -8.32860661e+00],\n",
       "        [-9.45811081e+00,  3.16288280e+00, -8.44378710e-01,\n",
       "          1.94583356e+00, -3.31365275e+00, -4.07466316e+00,\n",
       "          2.70395517e+00, -6.37109661e+00,  8.69134665e-01,\n",
       "         -2.51154041e+00, -8.80934596e-01, -5.17175941e+01,\n",
       "         -4.74458170e+00, -2.47488379e+00,  4.66453266e+00,\n",
       "          9.19061244e-01, -3.30828452e+00,  4.68110895e+00,\n",
       "         -1.30805922e+00, -5.61180592e+00, -2.23171258e+00,\n",
       "         -1.98851562e+00, -2.97081566e+00, -2.15795631e+01,\n",
       "         -1.46411657e+01, -5.34953976e+00, -2.11317039e+00,\n",
       "         -2.54061651e+00, -2.61095548e+00, -6.12061977e-01,\n",
       "         -3.52543831e+00, -8.70907879e+00],\n",
       "        [-9.85300446e+00, -2.19389582e+00, -5.41682839e-01,\n",
       "         -2.05435610e+00, -3.36702132e+00, -4.13842392e+00,\n",
       "          2.57097006e-01, -6.33366251e+00, -1.52260864e+00,\n",
       "         -2.11678624e+00, -5.63868463e-01, -5.04632530e+01,\n",
       "         -2.36581111e+00, -2.01699233e+00,  3.67822981e+00,\n",
       "         -3.93874317e-01, -3.16422486e+00, -7.52996087e-01,\n",
       "         -3.86830568e-01, -5.19480181e+00, -2.11458325e+00,\n",
       "          4.23937023e-01, -3.10565281e+00, -2.15453091e+01,\n",
       "         -1.20661869e+01, -4.22993541e-01, -3.33202004e+00,\n",
       "         -2.18275619e+00,  2.05579829e+00,  2.14497566e-01,\n",
       "         -2.91464305e+00, -8.96332264e+00],\n",
       "        [-1.46662407e+01, -3.70643902e+00, -2.14668846e+00,\n",
       "         -3.70603418e+00, -7.50219584e+00, -6.53613949e+00,\n",
       "          4.35751009e+00, -5.52865791e+00, -1.98060668e+00,\n",
       "         -3.45076632e+00, -4.77437353e+00, -2.45489540e+01,\n",
       "         -4.08950329e+00, -2.80552983e+00, -3.51898384e+00,\n",
       "         -9.99072909e-01, -4.43364978e-01, -3.79210663e+00,\n",
       "         -2.38859940e+00, -2.99570870e+00, -3.86679626e+00,\n",
       "         -4.56139994e+00, -3.40752745e+00, -7.64508820e+00,\n",
       "         -9.33339405e+00, -4.10982180e+00, -2.20447516e+00,\n",
       "         -3.89155060e-01, -3.37157845e+00, -3.02444386e+00,\n",
       "         -3.71769142e+00, -5.16907501e+00],\n",
       "        [-1.47840109e+01,  9.74598050e-01, -1.96929026e+00,\n",
       "         -1.04442966e+00, -9.32472420e+00, -8.37097073e+00,\n",
       "         -3.00987482e+00, -5.37725592e+00, -3.65002441e+00,\n",
       "         -3.23361254e+00, -5.18804932e+00, -2.55265865e+01,\n",
       "         -4.43000746e+00, -3.75423646e+00, -1.74234712e+00,\n",
       "         -1.88626158e+00, -3.91463113e+00, -3.48157382e+00,\n",
       "         -3.08643889e+00, -3.14677072e+00, -2.39217997e+00,\n",
       "          1.87844694e+00,  3.77941765e-02, -7.11713457e+00,\n",
       "         -5.41789246e+00, -3.45069289e+00, -3.58333063e+00,\n",
       "         -3.38526344e+00,  1.19150364e+00, -3.46931529e+00,\n",
       "         -3.41355109e+00, -5.75952196e+00],\n",
       "        [-1.53403988e+01, -2.56102300e+00, -3.05368352e+00,\n",
       "         -3.71391225e+00, -7.78669024e+00, -8.05584431e+00,\n",
       "         -2.56382585e+00, -5.59561014e+00, -3.01976871e+00,\n",
       "         -3.71326661e+00, -4.50809121e-01, -2.52159863e+01,\n",
       "         -3.33032179e+00, -5.20846653e+00, -7.57755220e-01,\n",
       "          2.09837984e-02, -5.77978134e+00,  6.51019156e-01,\n",
       "         -3.09953737e+00,  1.05349541e+00,  1.84442429e-03,\n",
       "         -4.28401423e+00,  2.77111381e-01, -7.94697618e+00,\n",
       "         -9.89808559e+00, -3.50584459e+00,  5.02403438e-01,\n",
       "         -6.89549863e-01, -3.99442029e+00, -1.85328901e+00,\n",
       "         -4.66829014e+00, -2.42567992e+00],\n",
       "        [-1.44546432e+01, -3.33704686e+00, -3.02722716e+00,\n",
       "         -3.52422643e+00, -8.43362236e+00, -6.82475233e+00,\n",
       "         -1.14190996e+00, -7.26371193e+00, -2.93432975e+00,\n",
       "          8.75381827e-01, -1.08622873e+00, -2.50627117e+01,\n",
       "         -4.42747402e+00, -3.94500947e+00, -3.27721930e+00,\n",
       "         -1.83094966e+00, -2.19264030e+00, -2.76157761e+00,\n",
       "         -1.29629111e+00,  1.31449628e+00, -3.68119645e+00,\n",
       "         -1.76091778e+00, -3.22440600e+00, -7.38109064e+00,\n",
       "         -1.08854084e+01, -3.59036016e+00, -3.57788473e-01,\n",
       "          2.11187744e+00, -1.58244789e+00,  5.54560304e-01,\n",
       "         -3.59864068e+00, -5.19197798e+00],\n",
       "        [-1.47032452e+01, -8.97884816e-02, -1.87104678e+00,\n",
       "          1.28806043e+00, -9.21219444e+00, -6.55057526e+00,\n",
       "         -1.59727752e+00, -6.42139769e+00,  8.18925261e-01,\n",
       "         -3.11671877e+00, -7.41586304e+00, -2.41254730e+01,\n",
       "         -3.96727872e+00,  1.45235884e+00, -3.16178203e+00,\n",
       "         -3.51112175e+00, -4.29320669e+00, -3.69269466e+00,\n",
       "         -6.10783100e-02,  2.73308158e+00, -8.10968113e+00,\n",
       "         -5.98416615e+00,  3.27076107e-01, -5.81917858e+00,\n",
       "         -2.51098537e+00, -4.55951154e-01, -1.27686787e+00,\n",
       "         -1.74872708e+00, -2.20103121e+00,  7.45507061e-01,\n",
       "         -2.86267948e+00, -5.27430773e+00],\n",
       "        [-1.44406300e+01, -3.54947662e+00, -4.45357822e-02,\n",
       "         -3.58746815e+00, -7.88891029e+00, -6.17420340e+00,\n",
       "         -4.31537485e+00, -1.82970154e+00, -2.42769480e+00,\n",
       "         -2.01412010e+00, -6.80666590e+00, -2.58067017e+01,\n",
       "          4.47620678e+00, -1.54957330e+00, -1.50651336e+00,\n",
       "          1.63831651e-01,  1.21249604e+00, -2.79740834e+00,\n",
       "          1.45890141e+00, -2.82113528e+00, -5.01842880e+00,\n",
       "         -4.49400091e+00, -3.37262011e+00, -5.46384478e+00,\n",
       "         -8.19738483e+00, -3.92689419e+00, -1.64475167e+00,\n",
       "         -3.96053934e+00, -3.33805013e+00, -3.13673878e+00,\n",
       "          2.06683278e+00, -6.78590822e+00],\n",
       "        [-1.48979378e+01, -3.17519450e+00, -3.04455090e+00,\n",
       "         -3.59064722e+00, -8.73259926e+00, -5.96050215e+00,\n",
       "         -3.39318657e+00,  2.64892392e-02, -1.69007254e+00,\n",
       "         -2.32658195e+00, -7.69422114e-01, -2.34566879e+01,\n",
       "         -2.77193642e+00, -4.48224354e+00,  1.17130518e+00,\n",
       "         -4.42415857e+00, -2.41289854e+00, -2.50595427e+00,\n",
       "         -3.05805779e+00, -3.28200197e+00, -9.21304417e+00,\n",
       "         -2.62871051e+00,  9.59738064e+00, -4.72383356e+00,\n",
       "         -4.84683132e+00,  3.47929090e-01,  2.16464758e+00,\n",
       "         -1.15624928e+00, -3.62678623e+00, -2.97107053e+00,\n",
       "         -2.58519268e+00, -3.16348505e+00]], dtype=float32),\n",
       " array([13.701671 , -6.9089193, -4.4684916, -7.0397863, 18.135963 ,\n",
       "         6.638889 , -6.535851 ,  2.808883 , -8.5051   , -7.2386627,\n",
       "         2.8063731, 35.910282 , -3.0899706, -5.4986105, -6.9278393,\n",
       "        -7.3664284, -3.129531 , -6.8780503, -7.1784105,  5.7487   ,\n",
       "         4.0494924, -6.4103756, -3.964335 , 18.000381 ,  5.714109 ,\n",
       "        -1.8557907, -7.9778132, -7.5949807, -8.410666 , -6.6592665,\n",
       "        -7.4422116, 11.429976 ], dtype=float32),\n",
       " array([[ 5.27182341e-01, -1.12251139e+01, -1.57085314e+01,\n",
       "         -4.32084122e+01, -2.42278500e+01,  3.56753254e+00,\n",
       "         -3.05243671e-01, -3.65298309e+01, -3.98047523e+01,\n",
       "         -8.54745102e+00, -1.28328886e+01, -2.10765660e-01,\n",
       "         -1.32910356e-01, -1.72332287e-01,  2.95735955e+00,\n",
       "         -2.65299153e+00, -1.44778595e+01, -2.01280746e+01,\n",
       "         -2.55482227e-01, -1.52162209e-01, -1.50323896e+01,\n",
       "         -1.44751530e+01, -9.26201418e-02, -2.74771023e+01,\n",
       "          2.62594676e+00, -1.72346573e+01, -2.25493031e+01,\n",
       "         -1.08561838e+00, -2.90018253e+01, -1.87520400e-01,\n",
       "         -2.31057682e+01, -1.65374339e+00],\n",
       "        [-7.38775909e-01,  1.97773802e+00,  4.41921115e-01,\n",
       "         -3.99524625e-03, -5.64070177e+00, -1.41993654e+00,\n",
       "         -7.85894990e-02, -1.47053404e+01, -8.86684239e-01,\n",
       "         -4.84600592e+00, -8.30618516e-02, -1.27141726e+00,\n",
       "          1.11089617e-01, -2.78122604e-01,  1.80187434e-01,\n",
       "          1.39327276e+00, -7.57146168e+00, -8.47168255e+00,\n",
       "          2.02498651e+00, -1.02935762e+01,  2.95824677e-01,\n",
       "          1.57003868e+00, -1.62638444e-02, -5.89414167e+00,\n",
       "         -1.58104286e-01, -2.26844549e-01, -7.67510128e+00,\n",
       "         -5.62546730e-01, -2.50806308e+00, -5.95854461e-01,\n",
       "         -6.86503458e+00, -1.04169881e+00],\n",
       "        [-1.52476823e+00,  2.97676414e-01, -5.40120125e-01,\n",
       "         -3.04252839e+00, -6.38548708e+00, -3.04561615e-01,\n",
       "         -3.04495096e-01,  5.72262764e-01, -1.48906422e+00,\n",
       "         -3.33042789e+00,  3.54664803e-01, -2.63357162e-01,\n",
       "          2.96563059e-01, -3.56281400e-02,  1.89132094e-01,\n",
       "         -2.33082557e+00, -5.28676605e+00, -5.91733217e-01,\n",
       "          1.40462911e+00,  1.03039598e+00,  1.70264542e-01,\n",
       "         -1.34803212e+00, -4.16632503e-01,  3.89664483e+00,\n",
       "          7.15805963e-02,  1.99481934e-01, -8.78760099e-01,\n",
       "         -1.20157361e+00,  3.20796680e+00, -1.00218499e+00,\n",
       "         -2.37973261e+00, -5.74553490e-01],\n",
       "        [-1.57076824e+00, -7.31336355e-01,  3.79003972e-01,\n",
       "         -2.33158321e+01, -3.22324204e+00, -1.31605244e+00,\n",
       "         -3.61130506e-01,  4.09899443e-01, -8.33001232e+00,\n",
       "          1.20680499e+00, -3.69302511e+00, -2.66386032e-01,\n",
       "         -5.63177347e-01, -4.32621837e-01,  4.88577455e-01,\n",
       "         -6.07978916e+00, -4.19069481e+00, -6.14262152e+00,\n",
       "          1.66571665e+00, -4.37056732e+00, -3.57426554e-01,\n",
       "         -2.87795395e-01,  3.33297104e-01,  7.44492710e-01,\n",
       "          1.27764031e-01,  1.78589448e-01, -4.17504120e+00,\n",
       "          2.24226475e+00, -4.02091980e+00,  2.08993722e-02,\n",
       "          2.52802181e+00, -1.16301692e+00],\n",
       "        [ 4.17535454e-01, -9.67509537e+01, -5.20259514e+01,\n",
       "         -8.78157020e-01, -3.94355469e+01, -6.83091655e-02,\n",
       "         -6.86725140e-01, -6.57076569e+01, -5.00178604e+01,\n",
       "         -7.72550888e+01,  1.84427679e-01,  1.00762315e-01,\n",
       "         -8.51198137e-02, -1.85144633e-01,  1.89514548e-01,\n",
       "         -2.26743832e+01, -2.17141652e+00, -3.40468979e+01,\n",
       "          3.36590782e-02, -2.32668858e+01, -2.17497158e+01,\n",
       "         -1.31415039e-01, -1.51656732e-01, -5.71036053e+00,\n",
       "          2.47342624e-02, -3.09256721e+00, -3.50711212e+01,\n",
       "         -1.30584812e+00, -1.92739506e+01,  1.40146136e-01,\n",
       "         -1.43441749e+00, -2.44964808e-01],\n",
       "        [ 8.67855370e-01, -3.15753531e+00, -3.90490007e+00,\n",
       "         -4.37351456e+01, -2.61554146e+01,  1.42407000e+00,\n",
       "          7.26602797e-04, -9.04871845e+00, -4.62875671e+01,\n",
       "         -1.01086855e+00, -2.35181160e+01, -4.72566038e-01,\n",
       "          2.17439607e-01, -2.77960807e-01, -1.27206316e+01,\n",
       "         -9.94266319e+00, -2.02899570e+01, -1.76342888e+01,\n",
       "          7.35011399e-01, -9.26823044e+00, -2.16344700e+01,\n",
       "          3.19710612e+00, -8.45298395e-02, -3.44276886e+01,\n",
       "         -4.62508202e+01, -1.75632076e+01, -5.84266233e+00,\n",
       "         -3.67033601e+00, -2.41767044e+01, -7.55771935e-01,\n",
       "         -3.68177071e+01, -1.06266665e+00],\n",
       "        [-1.15429544e+00, -1.02459612e+01, -7.10918713e+00,\n",
       "         -1.95617714e+01, -8.60497475e+00, -3.16590786e+00,\n",
       "         -4.71889973e-04,  7.21453249e-01, -1.21047487e+01,\n",
       "         -7.62298632e+00, -1.08683558e+01, -9.57967639e-01,\n",
       "          2.00111747e-01, -2.62018055e-01,  1.13608611e+00,\n",
       "          7.01362431e-01, -1.06438389e+01, -5.69479609e+00,\n",
       "         -1.86418152e+00, -4.73321104e+00, -1.21965485e+01,\n",
       "         -1.77758503e+01, -3.05176079e-01, -9.80205059e+00,\n",
       "         -2.04951262e+00, -4.20976543e+00, -2.47232189e+01,\n",
       "          2.95760345e+00, -3.03450222e+01, -9.29631770e-01,\n",
       "         -1.69903069e+01, -2.98697877e+00],\n",
       "        [ 3.31149149e+00, -2.27427826e+01, -1.06036198e+00,\n",
       "         -1.83906155e+01, -2.92874527e+01, -2.77058163e+01,\n",
       "         -3.26034844e-01, -2.39600067e+01, -1.73455887e+01,\n",
       "          1.22604859e+00, -3.04214740e+00, -1.10698736e+00,\n",
       "         -3.02416831e-01, -3.45020860e-01, -4.63128128e+01,\n",
       "          2.21708074e-01, -1.18590727e+01, -3.26134491e+01,\n",
       "         -9.06073675e-02, -4.21999817e+01, -4.21465635e-02,\n",
       "          4.52562380e+00, -3.92617106e-01,  7.65522599e-01,\n",
       "         -2.20275097e+01, -2.32158947e+01, -2.81437435e+01,\n",
       "         -5.72931099e+00, -4.66732836e+00, -1.08431554e+00,\n",
       "         -1.21986923e+01, -1.29563475e+00],\n",
       "        [ 2.13088751e+00,  5.53800392e+00,  3.37277961e+00,\n",
       "          1.16930103e+00, -2.35259247e+00, -7.35904789e+00,\n",
       "         -3.15807573e-02,  1.63127375e+00, -3.74139965e-01,\n",
       "         -1.11850615e+01,  4.10224885e-01, -4.10188198e-01,\n",
       "         -3.30948494e-02,  1.22019770e-02, -8.88284862e-01,\n",
       "         -4.61279202e+00, -3.64628291e+00, -1.42731180e+01,\n",
       "          7.50026286e-01, -4.58378172e+00,  1.82004645e-01,\n",
       "         -2.76446533e+00, -4.68159348e-01, -2.34994006e+00,\n",
       "         -1.58163741e-01,  1.27987623e+00, -7.82430410e-01,\n",
       "         -2.02226448e+00, -8.18660796e-01,  4.55495864e-01,\n",
       "          2.48186350e+00, -6.47183239e-01],\n",
       "        [-9.67323244e-01,  5.21141469e-01, -1.64741084e-01,\n",
       "          4.76367760e+00, -1.70336950e+00, -3.89339089e+00,\n",
       "          2.52328485e-01, -1.87384951e+00, -1.22465162e+01,\n",
       "          9.37105864e-02,  2.18480897e+00, -5.86086750e-01,\n",
       "          1.04599699e-01, -2.36964837e-01, -4.59770160e-03,\n",
       "         -1.46475661e+00, -1.20451822e+01, -7.95465469e+00,\n",
       "         -2.22855616e+00, -5.03228092e+00,  2.24657208e-02,\n",
       "          2.49573541e+00, -4.58493948e-01, -3.99814057e+00,\n",
       "         -8.07063058e-02,  1.67091918e+00, -2.88836932e+00,\n",
       "          2.69752622e+00, -5.27106285e+00, -3.14456850e-01,\n",
       "         -1.57949102e+00, -6.75876379e-01],\n",
       "        [-2.61314034e-01, -1.13701630e+00,  4.11115026e+00,\n",
       "         -1.86348553e+01, -9.86940479e+00, -3.71524658e+01,\n",
       "         -3.78943980e-01, -2.65959549e+01, -1.74577808e+01,\n",
       "         -4.59264984e+01, -2.58997002e+01, -9.82005715e-01,\n",
       "         -5.38010895e-01, -7.18477368e-02,  4.62441444e+00,\n",
       "         -9.02891827e+00, -6.44223261e+00, -2.33628330e+01,\n",
       "         -6.37096047e-01, -8.00527763e+00, -9.98493958e+00,\n",
       "         -8.90377104e-01, -3.60250890e-01, -1.93685913e+01,\n",
       "         -3.45469742e+01, -4.52485580e+01, -1.18430967e+01,\n",
       "          2.86268544e+00, -6.69163227e+00, -6.34644985e-01,\n",
       "         -4.14598083e+01, -7.01823652e-01],\n",
       "        [-2.96976638e+00, -1.62996941e+01,  8.01161170e-01,\n",
       "         -9.18941574e+01, -1.50445919e+01, -1.11988103e+00,\n",
       "         -4.20231700e-01, -2.79881191e+01, -3.39742317e+01,\n",
       "         -1.33679304e+01, -1.41267443e+01, -5.25726616e-01,\n",
       "         -7.71153152e-01, -7.70728067e-02,  1.56416917e+00,\n",
       "         -7.68760979e-01, -2.14682846e+01, -2.71173058e+01,\n",
       "          1.51621795e+00, -4.19081688e+00, -1.32840748e+01,\n",
       "         -7.69060745e+01, -1.15161985e-01, -2.44380836e+01,\n",
       "         -1.07907619e+01, -6.27484398e+01, -2.03537679e+00,\n",
       "         -1.92750168e+00,  1.46660948e+00, -4.73964810e-01,\n",
       "         -4.22027740e+01, -1.21466708e+00],\n",
       "        [ 1.50247681e+00, -9.60122490e+00,  4.20250368e+00,\n",
       "         -3.66666527e+01, -1.08625898e+01,  2.98666859e+00,\n",
       "         -3.15837637e-02, -2.45095444e+00, -1.96710167e+01,\n",
       "          3.71393003e-02, -2.59686985e+01, -1.28123850e-01,\n",
       "         -2.77165294e-01, -1.17065482e-01,  2.51063609e+00,\n",
       "         -2.00753098e+01, -1.34927845e+01, -1.67325649e+01,\n",
       "          1.09329915e+00, -1.14188662e+01, -9.60410786e+00,\n",
       "         -2.25462360e+01, -2.45357350e-01, -2.03131161e+01,\n",
       "         -5.16425085e+00, -6.37987947e+00, -7.41040277e+00,\n",
       "         -7.79874265e-01, -2.71125565e+01, -1.00541639e+00,\n",
       "         -5.04050598e+01, -6.70228362e-01],\n",
       "        [ 4.93282974e-01, -1.02584906e+01, -1.72414374e+00,\n",
       "         -7.94524765e+00, -1.34584417e+01, -1.74526386e+01,\n",
       "         -7.50849247e-02, -4.44180298e+00,  4.66923982e-01,\n",
       "         -1.50251322e+01, -1.76765049e+00, -1.19028121e-01,\n",
       "          1.52756378e-01,  2.25479319e-03, -1.46834087e+01,\n",
       "         -8.60606384e+00, -1.41449137e+01, -1.86113491e+01,\n",
       "          2.28662252e+00, -1.26252422e+01, -1.20584269e+01,\n",
       "         -2.10921383e+00, -9.35500413e-02, -1.68395603e+00,\n",
       "         -5.54974794e+00, -5.84790421e+00, -1.97630024e+01,\n",
       "         -2.82969928e+00,  1.02746964e+00, -6.08971715e-01,\n",
       "         -9.08927155e+00, -2.95117259e-01],\n",
       "        [ 1.18050539e+00, -3.09982628e-01, -8.05079937e-02,\n",
       "         -2.73783898e+00, -1.29730606e+00,  7.04415202e-01,\n",
       "          2.41289943e-01, -1.39870727e+00, -4.48390055e+00,\n",
       "         -1.07382259e+01, -8.18240643e-02, -3.30140293e-01,\n",
       "         -2.08852798e-01, -3.89200330e-01,  1.03417136e-01,\n",
       "          3.78058028e+00, -3.59418702e+00, -5.87371588e+00,\n",
       "         -7.71001279e-01, -8.00361729e+00, -2.88114876e-01,\n",
       "         -6.79809740e-03, -2.77476162e-01, -1.56672847e+00,\n",
       "         -6.87282020e-03,  1.36042878e-01, -2.13557765e-01,\n",
       "          2.08534741e+00, -3.68315315e+00,  2.48346105e-01,\n",
       "         -2.73971319e+00, -9.94864345e-01],\n",
       "        [-4.16415453e-01, -3.66427517e+00,  1.32022297e+00,\n",
       "         -2.42013121e+00, -4.81237173e+00, -1.37360230e-01,\n",
       "         -3.65135431e-01, -1.18355894e+01, -3.05353832e+00,\n",
       "         -5.76666403e+00, -1.70447338e+00, -1.16162458e-02,\n",
       "         -8.28651011e-01,  3.36130261e-02, -3.31006497e-01,\n",
       "          2.16850686e+00, -2.80016017e+00, -1.19875154e+01,\n",
       "          5.97356319e-01,  2.77210736e+00, -5.27585208e-01,\n",
       "          4.52219629e+00, -2.71381080e-01, -2.77243996e+00,\n",
       "         -4.35862482e-01,  1.34274036e-01, -1.82826674e+00,\n",
       "         -1.74010825e+00, -3.97201443e+00, -1.12345601e-02,\n",
       "         -3.19793057e+00, -1.77227902e+00],\n",
       "        [-5.32955647e+00, -1.89887695e+01, -1.41821022e+01,\n",
       "         -5.81715250e+00, -1.42902298e+01, -1.43586130e+01,\n",
       "         -1.66907698e-01, -2.76930332e+00, -2.62485905e+01,\n",
       "         -1.08619509e+01, -8.11700153e+00, -1.12968612e+00,\n",
       "          3.88006791e-02, -4.43779945e-01, -2.66597295e+00,\n",
       "         -1.10435257e+01, -1.77770061e+01, -7.13905764e+00,\n",
       "          2.32153606e+00, -9.04866886e+00, -1.86350174e+01,\n",
       "         -8.31098557e+00, -2.47286767e-01, -2.38087296e+00,\n",
       "         -5.89693642e+00, -1.33891258e+01, -1.97478981e+01,\n",
       "         -1.11657310e+00, -8.40993023e+00, -1.62900373e-01,\n",
       "          3.26464629e+00, -1.51322886e-01],\n",
       "        [-2.20629513e-01,  2.07732290e-01,  6.60218000e-02,\n",
       "         -2.99539971e+00,  2.33594918e+00, -7.69112051e-01,\n",
       "         -3.73018295e-01, -4.62607956e+00, -5.67018318e+00,\n",
       "         -2.43324232e+00, -2.16208652e-01, -6.40168071e-01,\n",
       "         -4.20167685e-01, -1.23904914e-01, -8.60503092e-02,\n",
       "         -5.35515594e+00, -1.58366048e+00, -8.24045277e+00,\n",
       "          4.41650480e-01, -7.39399338e+00,  4.01190192e-01,\n",
       "          2.99646437e-01, -1.74821183e-01, -9.36613846e+00,\n",
       "          2.27912530e-01, -3.17513019e-01, -6.23884583e+00,\n",
       "          1.36745608e+00,  1.43324375e+00, -2.91211993e-01,\n",
       "          1.51418817e+00, -5.86113691e-01],\n",
       "        [ 1.57283068e+00, -1.33957946e+00,  1.70603466e+00,\n",
       "          3.35267276e-01, -7.76207626e-01, -3.75518918e+00,\n",
       "          1.55830309e-02, -4.08476734e+00, -6.85361385e-01,\n",
       "          3.49952906e-01, -4.37551737e+00, -2.73382068e-01,\n",
       "         -4.09257054e-01, -3.68505299e-01, -5.72769880e-01,\n",
       "         -1.03921299e+01, -7.09872341e+00, -7.61000299e+00,\n",
       "         -2.21074557e+00, -4.05688584e-02, -5.09867296e-02,\n",
       "         -9.44882035e-01, -1.54267654e-01,  3.29928136e+00,\n",
       "          2.22114787e-01, -1.79328665e-01, -2.70309210e+00,\n",
       "          9.11009252e-01,  1.77872634e+00,  4.33320105e-01,\n",
       "         -2.16582608e+00, -3.23413610e-01],\n",
       "        [-4.15564217e-02,  7.62985587e-01,  2.63311481e+00,\n",
       "         -8.63530517e-01, -3.98411865e+01, -5.93128383e-01,\n",
       "         -7.39357471e-01, -5.19048080e+01, -1.54294434e+01,\n",
       "          2.10424647e-01, -3.25381546e+01, -7.06545174e-01,\n",
       "         -1.02789319e+00, -4.16477740e-01, -3.42933941e+00,\n",
       "          1.82492638e+00, -2.19148273e+01, -8.74151707e-01,\n",
       "          6.65252507e-01, -4.89704590e+01, -1.00387192e+01,\n",
       "          6.93321645e-01, -7.36044705e-01,  3.77559811e-01,\n",
       "          2.18217087e+00,  2.61851335e+00, -5.40102081e+01,\n",
       "         -1.02225125e+00, -1.83166099e+00, -1.18840337e+00,\n",
       "         -1.74279251e+01, -2.12913489e+00],\n",
       "        [-1.20885086e+00, -2.50232220e+01, -5.13991623e+01,\n",
       "         -3.30860291e+01, -2.14963455e+01,  2.09331751e+00,\n",
       "         -6.64330363e-01, -1.46095979e+00, -2.48245373e+01,\n",
       "         -7.77800140e+01, -1.90740013e+01, -8.65673840e-01,\n",
       "         -1.92308217e-01, -5.27727842e-01, -8.01750958e-01,\n",
       "          1.89682758e+00, -2.10649471e+01,  4.84043032e-01,\n",
       "          9.34045672e-01, -2.86714153e+01, -4.17548561e+01,\n",
       "         -1.06391134e+01, -4.49133009e-01, -3.63039169e+01,\n",
       "          4.22694206e-01,  1.82450569e+00, -2.99112282e+01,\n",
       "         -1.75689065e+00, -7.54801750e+00, -7.23564327e-01,\n",
       "         -2.71853695e+01, -8.66526723e-01],\n",
       "        [-2.85482836e+00, -9.71690559e+00, -6.45956087e+00,\n",
       "         -1.90363083e+01, -7.93609619e+00, -1.40033026e+01,\n",
       "         -6.13735691e-02, -9.40316391e+00, -1.19343309e+01,\n",
       "         -1.32243109e+01, -1.34664166e+00, -9.38365757e-02,\n",
       "         -3.25803459e-01, -2.22981304e-01, -4.36423540e+00,\n",
       "         -7.49543524e+00, -2.46605778e+01, -5.86069250e+00,\n",
       "          1.75008547e+00, -1.34924841e+01, -8.24981594e+00,\n",
       "         -1.25566053e+01, -1.00605071e-01, -6.56541395e+00,\n",
       "         -1.46454477e+00, -3.88064194e+00, -8.28383827e+00,\n",
       "          1.82832098e+00, -1.06253939e+01,  4.56803739e-02,\n",
       "         -1.20514975e+01, -1.09111212e-01],\n",
       "        [ 1.14126742e+00, -3.43965411e+00,  2.91232085e+00,\n",
       "         -2.27326260e+01, -1.48974552e+01, -2.96046913e-01,\n",
       "         -2.54470825e-01, -5.09917557e-01, -1.39580879e+01,\n",
       "         -9.52727985e+00, -4.42436066e+01, -1.16066301e+00,\n",
       "         -6.48435056e-01, -1.62841395e-01, -4.84073944e+01,\n",
       "         -4.02492142e+00, -1.06346245e+01, -1.14690657e+01,\n",
       "          9.85204637e-01, -1.97981319e+01, -9.23282814e+00,\n",
       "          4.00907469e+00,  4.37270501e-04, -1.91471252e+01,\n",
       "          2.31514573e-01, -6.58508360e-01, -9.80889988e+00,\n",
       "         -3.78393888e+00,  2.27429196e-01,  2.46339634e-01,\n",
       "         -1.00689859e+01, -1.72578478e+00],\n",
       "        [-7.37704098e-01, -2.29381542e+01,  7.10540041e-02,\n",
       "         -6.32568884e+00, -3.47101440e+01, -3.07591176e+00,\n",
       "         -6.97440803e-01, -2.82382164e+01, -5.31873817e+01,\n",
       "          1.41500688e+00,  5.55927277e-01, -4.20258254e-01,\n",
       "         -4.43710417e-01, -6.00734293e-01, -7.40550137e+00,\n",
       "         -7.15057325e+00, -2.87151623e+01, -1.79331398e+01,\n",
       "          7.48550713e-01, -3.27630310e+01, -2.55666370e+01,\n",
       "         -6.61358261e+01, -7.73442447e-01, -7.61807861e+01,\n",
       "         -2.91546936e+01,  2.26776910e+00, -3.94562408e-02,\n",
       "         -2.06117845e+00, -1.44711053e+00, -1.00754750e+00,\n",
       "         -6.83655090e+01, -7.03923106e-01],\n",
       "        [ 6.30306768e+00, -2.18512630e+01, -8.66162777e+00,\n",
       "         -3.00335732e+01, -2.41894112e+01, -1.46732388e+01,\n",
       "         -2.09908009e-01, -3.62155533e+00, -2.89931927e+01,\n",
       "         -8.88812256e+00, -6.99077129e+00, -1.36188078e+00,\n",
       "         -7.84866810e-01, -4.38984215e-01, -1.32362204e+01,\n",
       "         -5.10136175e+00, -2.27397003e+01, -1.87478561e+01,\n",
       "         -3.16128802e+00, -8.31334114e+00, -1.81481438e+01,\n",
       "         -1.26573973e+01, -4.06500787e-01, -1.08471632e+01,\n",
       "          6.51176929e-01, -2.88054028e+01, -1.12726946e+01,\n",
       "         -1.56303000e+00, -2.78382607e+01, -1.47468626e-01,\n",
       "         -3.99874115e+01, -1.72155634e-01],\n",
       "        [-1.77317393e+00, -7.37412167e+00, -5.40596056e+00,\n",
       "         -1.23377523e+01, -9.64566708e+00, -3.65250516e+00,\n",
       "         -2.13100761e-02, -6.45207310e+00, -1.52879429e+01,\n",
       "         -1.60096340e+01, -2.34705296e+01, -9.81809855e-01,\n",
       "         -5.78670382e-01, -3.41245830e-01, -5.75562239e-01,\n",
       "         -1.37445955e+01, -1.09607439e+01, -2.87808456e+01,\n",
       "          1.23642480e+00, -2.86620617e+00, -4.10905123e+00,\n",
       "         -3.06193790e+01,  6.57635555e-02, -8.06668949e+00,\n",
       "         -1.82166803e+00, -1.03673697e+01, -4.81878519e+00,\n",
       "          2.21858907e+00, -5.15214682e-01, -6.12322211e-01,\n",
       "         -2.00949745e+01, -1.46969473e+00],\n",
       "        [ 2.63928175e+00, -2.31815904e-01, -2.10689157e-01,\n",
       "         -3.33817816e+00, -1.20843321e-01, -3.36351514e+00,\n",
       "         -3.14527154e-01,  2.08125114e+00, -6.70716238e+00,\n",
       "         -5.08751988e-01, -7.61939287e-01, -9.41401124e-01,\n",
       "         -7.39678979e-01, -7.74339810e-02,  2.80245304e-01,\n",
       "          1.69439983e+00, -1.33235371e+00, -8.84426403e+00,\n",
       "         -9.89872277e-01, -1.06733961e+01, -1.41207695e-01,\n",
       "         -8.25695042e-03,  1.21002331e-01, -1.91974366e+00,\n",
       "         -3.71953607e-01, -2.16915655e+00, -5.11398077e+00,\n",
       "          1.70023739e+00,  3.00921869e+00, -2.81949520e-01,\n",
       "         -8.41301382e-01, -1.79234731e+00],\n",
       "        [ 1.27493978e+00, -4.27103853e+00, -8.02878320e-01,\n",
       "         -1.21697092e+00, -2.30079293e+00, -6.33117390e+00,\n",
       "         -4.24079359e-01, -1.09868469e+01, -3.73929358e+00,\n",
       "         -3.65747285e+00,  5.67733908e+00, -4.61712182e-01,\n",
       "         -2.51142979e-02, -5.86924404e-02, -3.50162238e-01,\n",
       "         -1.22458067e+01, -7.46452332e+00, -2.30437107e-02,\n",
       "         -7.60286689e-01, -6.26394987e+00, -1.53775185e-01,\n",
       "          1.70508549e-01, -1.02272056e-01, -6.09607983e+00,\n",
       "         -4.81805176e-01,  4.13931385e-02, -1.91585875e+00,\n",
       "         -4.47679460e-01, -2.68733287e+00,  1.48808673e-01,\n",
       "         -2.26707768e+00, -2.60240436e-01],\n",
       "        [-2.86682987e+00, -1.03021688e+01, -4.47985935e+00,\n",
       "         -4.58248758e+00, -1.52808881e+00, -8.24011040e+00,\n",
       "         -1.34746641e-01, -4.17766905e+00,  5.65955788e-02,\n",
       "          3.12793398e+00, -1.83438587e+00, -5.08633077e-01,\n",
       "          3.72359082e-02, -4.98369277e-01, -5.16491127e+00,\n",
       "         -1.05382376e+01, -1.65520420e+01,  1.55782461e+00,\n",
       "          2.08402410e-01, -8.27303982e+00,  1.55722833e+00,\n",
       "         -1.80852890e+00,  4.93824221e-02, -1.40955186e+00,\n",
       "         -7.12011099e-01, -4.70689011e+00, -1.02344990e+01,\n",
       "          2.50077367e+00, -1.12845058e+01, -9.27273214e-01,\n",
       "         -4.47521448e-01, -4.96425450e-01],\n",
       "        [ 3.66013336e+00, -1.69540182e-01, -3.26638430e-01,\n",
       "         -1.22955799e+00, -1.89915383e+00, -1.50679052e+00,\n",
       "          1.58753961e-01,  1.13587868e+00, -2.91289616e+00,\n",
       "          4.82004166e+00, -3.87407720e-01,  7.81961158e-02,\n",
       "         -2.07520425e-02, -8.13512355e-02, -2.81560302e-01,\n",
       "         -4.98248005e+00, -7.87813902e+00, -4.74546003e+00,\n",
       "         -2.35614491e+00,  2.48499346e+00, -5.93186498e-01,\n",
       "          1.64429700e+00, -3.45514953e-01, -2.81508684e+00,\n",
       "          1.72759101e-01, -1.61697716e-02, -4.53314924e+00,\n",
       "         -1.00350046e+00, -2.38853741e+00,  1.53717265e-01,\n",
       "         -2.68542439e-01,  2.65735120e-01],\n",
       "        [-1.25711751e+00, -4.13684845e+00, -4.11554098e+00,\n",
       "         -8.66607189e+00, -2.32180476e+00, -7.54449248e-01,\n",
       "         -2.90715583e-02,  1.49221706e+00, -1.06794252e+01,\n",
       "          4.41983414e+00, -1.20703039e+01, -3.10262948e-01,\n",
       "         -4.61811692e-01,  2.48405486e-02, -5.05470991e-01,\n",
       "         -1.04180632e+01, -5.54088926e+00, -2.44476175e+00,\n",
       "         -6.53707922e-01,  5.56965828e-01, -4.47314709e-01,\n",
       "         -1.50817289e+01, -2.93371230e-01, -7.30237293e+00,\n",
       "         -3.75562638e-01,  1.44916344e+00,  1.86115921e+00,\n",
       "          7.45362639e-01, -1.07913322e+01, -7.67741054e-02,\n",
       "         -7.64286470e+00, -1.15028417e+00],\n",
       "        [ 5.86024940e-01, -4.17377777e+01,  2.80204320e+00,\n",
       "         -3.55919304e+01, -3.73169994e+00,  1.62659466e-01,\n",
       "         -1.97385922e-01, -4.69908066e+01, -4.06333275e+01,\n",
       "          3.36412430e-01, -1.61288567e+01, -1.56854063e-01,\n",
       "         -4.57984328e-01, -5.75669289e-01,  7.09266901e-01,\n",
       "         -8.94674969e+00, -1.89257030e+01, -1.89189472e+01,\n",
       "          1.06153697e-01, -1.35653667e+01, -7.22801018e+00,\n",
       "          1.43903041e+00,  5.20989858e-02, -2.75856948e+00,\n",
       "         -5.08330488e+00, -1.06883440e+01, -3.27799950e+01,\n",
       "         -1.09170020e+00, -2.30410862e+00, -3.61943305e-01,\n",
       "         -2.17797899e+00, -5.72359681e-01]], dtype=float32),\n",
       " array([ 16.027225  ,   1.9522132 ,  -4.646764  ,   5.7658796 ,\n",
       "          3.4043674 ,   3.3100433 ,  -0.8062032 ,   4.030086  ,\n",
       "          4.1225185 ,   0.8704881 ,   1.945571  ,  -1.1996932 ,\n",
       "         -1.2886304 ,  -0.4037747 ,   1.3383483 ,  -1.3998606 ,\n",
       "          2.7909632 ,   2.091795  ,  38.727833  ,  -0.05536124,\n",
       "          1.4249934 ,  -7.3037825 ,  -0.7187314 ,   4.8434377 ,\n",
       "         -0.8497978 , -12.678458  ,  -1.1142735 ,  19.53675   ,\n",
       "          0.8466482 ,  -1.6083117 ,   3.6843789 ,  -0.813405  ],\n",
       "       dtype=float32),\n",
       " array([[-1.22170210e-01, -3.45258415e-01,  5.49397373e+00,\n",
       "          5.25823641e+00,  5.88858843e+00,  2.09900594e+00,\n",
       "          1.10295081e+00,  2.96716785e+00,  2.67553902e+00,\n",
       "         -1.47513956e-01,  6.00843382e+00,  7.22467947e+00,\n",
       "          6.09393406e+00, -1.27579987e-01,  3.64197969e+00,\n",
       "          4.64153576e+00, -2.26941779e-01,  5.65022230e+00,\n",
       "          5.36857033e+00,  5.12193155e+00,  6.35269070e+00],\n",
       "        [ 2.52402276e-01, -3.01338881e-01,  2.72627378e+00,\n",
       "          9.05724907e+00,  1.02930641e+01,  3.81130362e+00,\n",
       "          1.25908661e+01,  9.66998863e+00,  8.20285702e+00,\n",
       "         -1.66072786e-01, -5.60139465e+00, -6.04636812e+00,\n",
       "          1.89404106e+00, -3.35190713e-01,  5.27027941e+00,\n",
       "          3.18577933e+00, -2.08098099e-01, -1.20521336e+01,\n",
       "         -1.18104963e+01, -1.06549902e+01, -9.85383129e+00],\n",
       "        [-2.43787467e-01, -3.95429432e-01, -8.21135223e-01,\n",
       "          5.68383217e-01, -2.19082904e+00,  1.16395660e-01,\n",
       "         -2.94177341e+00, -7.27525175e-01,  7.78257310e-01,\n",
       "          8.20811093e-02, -8.89257789e-01, -1.71816289e+00,\n",
       "         -1.71078265e+00,  3.22553188e-01, -2.16229367e+00,\n",
       "         -3.74203801e+00,  1.86291307e-01, -2.70320684e-01,\n",
       "          3.00648898e-01,  2.58420348e-01, -2.05045462e+00],\n",
       "        [ 3.75459820e-01, -3.33497792e-01, -4.45473289e+00,\n",
       "         -3.88273168e+00, -8.63748741e+00, -2.63231468e+00,\n",
       "         -8.84743309e+00, -3.94426990e+00, -4.58479357e+00,\n",
       "          2.80361205e-01,  1.30757391e+00,  2.01672697e+00,\n",
       "          1.38206267e+00, -1.43881530e-01, -1.60705373e-01,\n",
       "          3.05982685e+00, -2.28341788e-01,  6.67972326e+00,\n",
       "          1.94650471e+00,  1.03368044e+00,  5.37424803e+00],\n",
       "        [-1.27590954e-01,  3.03102434e-02, -6.69435310e+00,\n",
       "         -2.87395096e+01, -4.29974079e+00,  3.43074131e+00,\n",
       "          7.86181450e+00, -6.55457830e+00, -2.36905909e+00,\n",
       "         -2.65609145e-01, -1.54223084e+00, -1.10786676e+01,\n",
       "         -3.11895704e+00, -2.40359753e-01,  1.06778612e+01,\n",
       "          4.07118082e+00,  4.22025472e-01,  1.12214289e+01,\n",
       "         -5.49497890e+00, -1.71866751e+00,  6.60969400e+00],\n",
       "        [ 1.15235448e-02,  1.51223034e-01, -1.06003249e+00,\n",
       "         -2.42141873e-01, -2.18356133e+00,  1.76148689e+00,\n",
       "         -5.15925884e-01, -8.70250817e-03,  2.77535498e-01,\n",
       "         -2.09105924e-01, -1.97115648e+00,  7.21786678e-01,\n",
       "         -1.83901295e-01, -1.21887237e-01, -1.77549100e+00,\n",
       "         -4.35530710e+00, -2.30744556e-01,  1.44576216e+00,\n",
       "          2.58273578e+00,  2.47557402e+00,  2.63907838e+00],\n",
       "        [ 1.72080845e-01, -1.12865865e-01, -9.61541161e-02,\n",
       "         -4.06199664e-01,  6.97094023e-01, -2.61786640e-01,\n",
       "          6.47670984e-01,  2.71471232e-01,  2.31466845e-01,\n",
       "         -1.52212352e-01,  2.94741631e-01,  1.98251888e-01,\n",
       "         -3.67151320e-01,  1.03152603e-01, -4.90051471e-02,\n",
       "          4.41258922e-02,  2.74948180e-02, -3.23403269e-01,\n",
       "         -2.47182980e-01,  2.10684747e-01,  1.56493112e-01],\n",
       "        [ 2.42709547e-01, -4.36693728e-01, -8.67244148e+00,\n",
       "         -1.35778122e+01, -1.09064283e+01, -1.33240528e+01,\n",
       "         -9.80501842e+00, -7.06974411e+00, -7.06254625e+00,\n",
       "          1.62762552e-01, -5.71355343e-01,  4.19238186e+00,\n",
       "          4.99902010e+00,  1.36107713e-01, -7.07122564e+00,\n",
       "         -1.49355960e+00,  3.69463116e-01, -2.73781133e+00,\n",
       "         -5.34681845e+00, -4.23805046e+00, -5.62669039e-01],\n",
       "        [-1.71237499e-01, -6.44386411e-02,  1.22048092e+01,\n",
       "          2.42880363e+01,  9.03525543e+00,  2.62715197e+00,\n",
       "          8.94002378e-01,  8.00251484e+00,  2.93529892e+00,\n",
       "         -2.37274066e-01,  3.75409484e-01,  5.19415331e+00,\n",
       "         -2.87902141e+00,  1.35602146e-01,  6.08225918e+00,\n",
       "          8.37749958e+00,  1.80997223e-01, -5.95460558e+00,\n",
       "          9.59982872e+00,  7.61249542e-01,  6.37225533e+00],\n",
       "        [ 4.15538579e-01,  2.60544568e-01, -1.56992090e+00,\n",
       "         -1.59554958e+00, -6.85049677e+00, -1.71451703e-01,\n",
       "         -2.38333821e+00, -2.57162833e+00, -1.94348991e+00,\n",
       "          2.13025540e-01, -1.98741794e+00, -6.76515281e-01,\n",
       "          4.51593250e-01, -3.98336709e-01,  1.78659630e+00,\n",
       "          2.53077698e+00, -2.74646014e-01,  2.00580406e+00,\n",
       "          3.66729450e+00,  3.59337282e+00,  1.13866460e+00],\n",
       "        [-2.10597813e-01,  3.27440292e-01,  3.83728123e+00,\n",
       "          3.16667390e+00,  7.45307088e-01, -9.21457350e-01,\n",
       "          2.69833994e+00,  2.59751987e+00,  3.57194233e+00,\n",
       "          1.53102368e-01, -1.76552022e+00, -1.32270348e+00,\n",
       "         -4.17026854e+00, -3.40654194e-01,  4.27922785e-01,\n",
       "          2.78492498e+00,  2.67387539e-01, -3.22001553e+00,\n",
       "          6.66592479e-01,  1.36779594e+00, -8.41365278e-01],\n",
       "        [ 3.68797868e-01, -3.67537141e-01, -1.06857404e-01,\n",
       "         -2.05665186e-01, -3.49044114e-01,  2.52264500e-01,\n",
       "         -1.02861851e-01, -1.48509905e-01,  2.10070103e-01,\n",
       "         -8.40236545e-02,  8.29032600e-01, -2.75075942e-01,\n",
       "          1.60795763e-01, -1.52087629e-01, -1.04324013e-01,\n",
       "          1.21966638e-01, -3.57252836e-01,  1.57534450e-01,\n",
       "          2.90335149e-01,  7.38549590e-01,  8.54309916e-01],\n",
       "        [-2.19454527e-01,  4.02008802e-01, -3.40379715e-01,\n",
       "         -3.09351623e-01,  1.91944957e-01,  1.81629390e-01,\n",
       "          4.05776143e-01, -5.48447549e-01, -2.58650444e-02,\n",
       "          2.85209745e-01, -1.08734503e-01, -1.13653332e-01,\n",
       "          2.27653816e-01,  3.47557038e-01,  4.77043748e-01,\n",
       "         -2.48792712e-02, -2.49870062e-01, -1.05418563e-02,\n",
       "         -6.39262795e-03,  1.47143155e-01, -1.80263385e-01],\n",
       "        [ 8.20635855e-02, -1.00318421e-05, -6.57534599e-02,\n",
       "          2.67259508e-01, -3.50139320e-01, -7.18465447e-03,\n",
       "          1.49978131e-01,  3.30646604e-01,  9.67067182e-02,\n",
       "         -2.44404942e-01,  3.82319897e-01, -1.15902096e-01,\n",
       "          1.42937213e-01, -4.87111211e-02,  4.17272478e-01,\n",
       "          1.69465706e-01, -8.64174664e-02, -3.91826719e-01,\n",
       "         -2.34479904e-02,  2.02383548e-01, -3.09506387e-01],\n",
       "        [-1.72677875e-01,  4.07963902e-01,  2.85686088e+00,\n",
       "          1.01411080e+00,  1.28178513e+00,  1.09641282e-02,\n",
       "         -2.39226937e+00, -8.97081792e-01, -8.93597841e-01,\n",
       "          2.58410841e-01, -4.98520374e+00, -2.71101260e+00,\n",
       "         -3.11709714e+00,  2.54290432e-01, -1.35481822e+00,\n",
       "          1.04812610e+00, -3.76038790e-01, -1.41829503e+00,\n",
       "         -1.56366014e+00, -8.98239076e-01, -1.57509267e+00],\n",
       "        [-3.19518924e-01,  3.63101989e-01, -4.53597736e+00,\n",
       "         -9.37121582e+00, -2.77242994e+00, -2.73615766e+00,\n",
       "         -4.65532827e+00, -1.33625078e+00,  2.48325515e+00,\n",
       "         -2.57851839e-01,  3.17924809e+00,  1.10602096e-01,\n",
       "          6.42651606e+00,  7.43003786e-02,  4.73837107e-01,\n",
       "          2.08963704e+00,  1.47975355e-01,  1.44898891e+00,\n",
       "          1.78112996e+00, -3.33928561e+00, -1.78759801e+00],\n",
       "        [-2.39139199e-02, -4.08620894e-01,  1.03991594e+01,\n",
       "          2.11861944e+00,  2.45575638e+01, -9.11979318e-01,\n",
       "          8.84914398e+00, -5.75728238e-01,  5.75743437e+00,\n",
       "         -2.01495811e-01,  4.87035036e+00, -2.04006290e+01,\n",
       "          4.08200026e+00, -2.16767743e-01,  1.15498734e+01,\n",
       "         -9.92878973e-01, -2.33198524e-01, -2.16220331e+00,\n",
       "         -1.29242220e+01, -3.65560389e+00, -2.29021225e+01],\n",
       "        [ 2.54171103e-01, -2.27755100e-01, -3.48183942e+00,\n",
       "          6.79082012e+00, -3.97058666e-01,  2.51811361e+00,\n",
       "         -9.45051003e+00, -4.99984407e+00, -4.93966246e+00,\n",
       "          2.69005567e-01,  3.11918020e+00,  4.31105405e-01,\n",
       "          1.14251339e+00,  2.14915246e-01, -7.65694916e-01,\n",
       "          3.50880599e+00, -1.25044882e-01, -8.25039101e+00,\n",
       "         -6.72050571e+00, -2.57451296e+00,  4.84127067e-02],\n",
       "        [-2.71894306e-01, -3.82627726e-01,  9.93328190e+00,\n",
       "          1.05291843e+01,  1.01436157e+01,  1.14433622e+01,\n",
       "          1.11984129e+01,  1.10511189e+01,  1.07760935e+01,\n",
       "         -3.57126474e-01,  9.29369736e+00,  9.61487103e+00,\n",
       "          9.75461388e+00, -2.26277128e-01,  1.07904453e+01,\n",
       "          1.10292139e+01, -3.00933242e-01,  9.44695759e+00,\n",
       "          9.48384190e+00,  9.64738750e+00,  9.31617546e+00],\n",
       "        [-6.49440885e-02, -3.07520330e-01,  3.51919115e-01,\n",
       "         -1.79789009e+01, -2.12063026e+00, -1.23189223e+00,\n",
       "         -8.77320862e+00, -7.72722840e-01, -8.91067791e+00,\n",
       "         -3.90067428e-01,  1.08962679e+00, -7.69667006e+00,\n",
       "         -2.01554108e+00, -2.23985851e-01, -1.31575871e+00,\n",
       "          2.03860545e+00,  1.28163904e-01,  5.60485601e+00,\n",
       "          1.18608490e-01,  7.92570472e-01, -5.69952154e+00],\n",
       "        [ 2.56327957e-01, -3.78075182e-01, -1.15308456e+01,\n",
       "          3.67930889e+00, -1.42685690e+01,  5.81019878e+00,\n",
       "          1.50110483e+01,  2.53081441e+00,  5.33228159e+00,\n",
       "         -3.11946869e-02,  3.46242189e+00,  7.80441570e+00,\n",
       "         -1.84908462e+00,  3.63566250e-01, -1.12874756e+01,\n",
       "         -1.15622654e+01,  4.58162129e-02,  1.78738003e+01,\n",
       "          1.31314631e+01,  9.45921803e+00,  1.21267748e+01],\n",
       "        [ 1.39155984e-02,  2.57685512e-01, -1.37703490e+00,\n",
       "         -3.25302196e+00,  3.74021918e-01, -7.45958924e-01,\n",
       "          5.66043234e+00,  1.99229848e+00,  1.86754659e-01,\n",
       "         -2.69830674e-01,  3.88827610e+00,  5.55013418e-01,\n",
       "          7.60396361e-01,  8.12200010e-02, -3.91341543e+00,\n",
       "         -5.97699690e+00,  6.08943701e-02, -1.04659414e+00,\n",
       "          5.25043845e-01, -4.23250020e-01, -3.48378968e+00],\n",
       "        [ 4.24121290e-01, -1.87640518e-01,  4.00655061e-01,\n",
       "          2.53748894e-01,  6.94619119e-02,  1.28675431e-01,\n",
       "         -2.48309076e-01, -4.23908740e-01,  2.86163151e-01,\n",
       "          2.15583891e-01, -2.50792384e-01,  4.42888141e-02,\n",
       "         -2.58251935e-01,  7.03856647e-02,  2.80940861e-01,\n",
       "         -3.07611316e-01,  1.52164549e-01, -2.62185276e-01,\n",
       "         -2.42434949e-01,  1.09496675e-01, -4.07177210e-01],\n",
       "        [-1.38874769e-01, -2.17511266e-01, -3.81118745e-01,\n",
       "         -3.71297431e+00, -2.54328418e+00, -4.70212460e+00,\n",
       "         -8.76738071e+00, -7.24021339e+00, -5.49703741e+00,\n",
       "         -3.33118111e-01,  1.36318302e+00,  4.28442955e+00,\n",
       "          7.46677399e-01, -1.14289403e-01, -7.19021988e+00,\n",
       "         -4.02740145e+00, -1.89131483e-01, -3.55582476e+00,\n",
       "         -2.38681480e-01,  1.34498096e+00, -1.44222867e+00],\n",
       "        [ 4.16927844e-01,  3.04759711e-01,  3.36989546e+00,\n",
       "          3.47474241e+00,  3.07827640e+00, -4.61222315e+00,\n",
       "         -2.67114449e+00, -2.28008449e-01, -2.44649673e+00,\n",
       "          1.33455187e-01, -1.60224006e-01,  1.15119147e+00,\n",
       "         -5.51875651e-01,  2.36922234e-01,  1.34814143e+00,\n",
       "          3.06383729e+00,  3.38315219e-01, -7.05221355e-01,\n",
       "          1.16730511e+00,  3.91413403e+00,  7.03560233e-01],\n",
       "        [-3.72971267e-01, -3.45483661e-01,  1.37656951e+00,\n",
       "         -6.44455671e-01, -1.17675114e+00, -3.24925399e+00,\n",
       "         -1.05578012e+01, -2.42270994e+00, -1.64858782e+00,\n",
       "         -7.89175928e-02, -3.74027133e-01,  4.95111132e+00,\n",
       "          5.30516863e+00,  2.35660940e-01,  2.24635452e-01,\n",
       "         -4.30840349e+00, -2.60660052e-02,  7.29588389e-01,\n",
       "         -3.43906045e+00, -2.86530066e+00, -3.60607117e-01],\n",
       "        [-5.78530729e-02,  4.62857485e-02,  4.92164612e+00,\n",
       "         -6.20750546e-01,  8.76660252e+00,  4.02659798e+00,\n",
       "          3.89345884e+00, -4.56467152e+00, -2.17792559e+00,\n",
       "          1.34916633e-01,  3.41256976e+00,  5.24872446e+00,\n",
       "          1.14644146e+01,  2.97547370e-01,  2.38996655e-01,\n",
       "          4.05785990e+00, -1.13544554e-01, -1.47677298e+01,\n",
       "         -7.04263783e+00,  6.93125427e-01, -3.55366564e+00],\n",
       "        [-4.02473688e-01, -1.95356816e-01,  3.19844460e+00,\n",
       "          2.97527575e+00,  3.74216557e+00,  2.89193082e+00,\n",
       "          6.24736118e+00,  3.54326510e+00,  3.37011528e+00,\n",
       "         -3.46695781e-02,  5.96016932e+00,  2.24260902e+00,\n",
       "          3.45571852e+00, -9.72272158e-02,  3.17665553e+00,\n",
       "          1.18780673e+00, -4.17092532e-01,  4.40998268e+00,\n",
       "          4.71096802e+00,  3.71507120e+00,  2.44206524e+00],\n",
       "        [-2.08823562e-01,  1.78019833e-02,  9.08502400e-01,\n",
       "          1.78716862e+00,  1.07644653e+00, -2.03148603e+00,\n",
       "         -1.97301936e+00,  2.05075479e+00,  2.45025039e+00,\n",
       "          6.84650242e-02, -7.79471064e+00,  3.83542538e+00,\n",
       "          1.50217259e+00, -2.99163759e-02, -3.73785043e+00,\n",
       "         -2.69601989e+00, -3.39793384e-01, -1.44002457e+01,\n",
       "          3.78894711e+00, -7.02919960e-01, -1.23055649e+01],\n",
       "        [ 1.52656287e-01, -1.61467090e-01,  9.78050053e-01,\n",
       "          5.38978502e-02,  5.33167720e-01,  4.44294691e-01,\n",
       "         -2.60345817e-01,  6.55523598e-01, -2.05517516e-01,\n",
       "         -8.64220858e-02, -5.79816066e-02,  7.34026492e-01,\n",
       "          5.09706855e-01,  1.30750746e-01,  3.28538209e-01,\n",
       "          6.70802712e-01,  4.16747034e-02, -1.74098358e-01,\n",
       "         -1.10563219e-01,  1.17390227e+00, -2.30177745e-01],\n",
       "        [-3.70995283e-01,  3.72183770e-01, -5.28006029e+00,\n",
       "          7.25843477e+00, -5.17888498e+00,  1.11016369e+00,\n",
       "         -2.64245868e+00,  7.00573504e-01, -2.52143502e+00,\n",
       "         -2.17672631e-01,  2.30917577e-02,  9.58104134e+00,\n",
       "          1.39956701e+00,  1.22468472e-02, -6.16924477e+00,\n",
       "         -3.02420449e+00, -4.15682495e-01, -3.04425025e+00,\n",
       "          3.29013872e+00,  2.74292111e-01,  3.22738123e+00],\n",
       "        [-1.62041038e-01,  7.81938136e-02,  3.75392705e-01,\n",
       "         -1.70592561e-01,  8.07650626e-01,  1.96687892e-01,\n",
       "          3.87583643e-01,  6.25258088e-01,  1.97590247e-01,\n",
       "         -4.08198833e-02,  1.19178426e+00,  4.02419627e-01,\n",
       "          1.09789073e+00, -3.66667330e-01, -1.40694872e-01,\n",
       "          1.22743726e+00, -1.35762393e-02,  2.74838805e-01,\n",
       "         -1.80185884e-01,  1.06585753e+00,  1.20238900e+00]], dtype=float32),\n",
       " array([ 0.        , -0.09632435, 37.341366  , 14.458738  , 17.952812  ,\n",
       "        25.628641  , 37.547287  , 21.28078   , 33.60882   ,  0.        ,\n",
       "        35.456432  , 31.100866  , 20.993258  ,  0.        , 38.989204  ,\n",
       "        43.578514  ,  0.        , 14.221806  , 15.667272  , 23.640106  ,\n",
       "        47.05461   ], dtype=float32)]"
      ]
     },
     "execution_count": 13,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "## Verifying the 'model_weights.pkl' file generated\n",
    "\n",
    "model_weights_path = r\"model_weights.pkl\"\n",
    "model_weights= pd.read_pickle(model_weights_path)   ## List of Numpy arrays of model_weights\n",
    "model_weights"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Model Prediction (Loading Saved DQN_Model) "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<keras.engine.sequential.Sequential at 0x7ff31ad457f0>"
      ]
     },
     "execution_count": 14,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "## Loading the saved 'DQN_model.h5' (to test for prediction)\n",
    "from keras.models import load_model\n",
    "model = load_model('DQN_model.h5')  \n",
    "model"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {},
   "outputs": [],
   "source": [
    "## Creating a function to make model predictions:\n",
    "def model_predict(state, state_size= m+t+d):\n",
    "    state_encoded= np.array(env.state_encod_arch2(state)).reshape(1, state_size)\n",
    "    \n",
    "    # Predict the action for given state\n",
    "    action= action_space[np.argmax(model.predict(state_encoded))]\n",
    "    print();print(\"*\"*20+ \" Model Prediction \" +\"*\"*20)\n",
    "    print(f\"Predicted action corresponding to state {state}:\",action)\n",
    "    print(\"*\"*58)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      "******************** Model Prediction ********************\n",
      "Predicted action corresponding to state (3, 3, 4): (3, 2)\n",
      "**********************************************************\n"
     ]
    }
   ],
   "source": [
    "## Model Prediction: Example 1: state (3,3,4)\n",
    "model_predict(state=(3,3,4))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      "******************** Model Prediction ********************\n",
      "Predicted action corresponding to state (1, 10, 3): (1, 2)\n",
      "**********************************************************\n"
     ]
    }
   ],
   "source": [
    "## Model Prediction: Example 2:  state (1,10,3)\n",
    "model_predict(state=(1,10,3))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      "******************** Model Prediction ********************\n",
      "Predicted action corresponding to state (2, 2, 5): (2, 1)\n",
      "**********************************************************\n"
     ]
    }
   ],
   "source": [
    "## Model Prediction: Example 3: state (2,2,5)\n",
    "model_predict(state=(2,2,5))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      "******************** Model Prediction ********************\n",
      "Predicted action corresponding to state (1, 1, 1): (1, 2)\n",
      "**********************************************************\n"
     ]
    }
   ],
   "source": [
    "## Model Prediction: Example 4: state (1,1,1)\n",
    "model_predict(state=(1,1,1))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Tracking Convergence"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "**Q_values for Sample state-action pairs tracked**"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "[0.0,\n",
       " 0.0,\n",
       " 0.0,\n",
       " 0.0,\n",
       " 0.0,\n",
       " 0.0,\n",
       " 0.0,\n",
       " 202.46686,\n",
       " 232.74548,\n",
       " 286.4916,\n",
       " 342.78458,\n",
       " 383.08374,\n",
       " 401.2797,\n",
       " 341.01062,\n",
       " 357.11734,\n",
       " 320.26776,\n",
       " 329.2884,\n",
       " 321.45425,\n",
       " 404.1648,\n",
       " 522.9371,\n",
       " 335.5011,\n",
       " 435.6562,\n",
       " 570.26245,\n",
       " 417.67587,\n",
       " 458.152,\n",
       " 456.6614,\n",
       " 348.0986,\n",
       " 390.9257,\n",
       " 410.4838,\n",
       " 428.83002,\n",
       " 495.30554,\n",
       " 483.03937,\n",
       " 489.30417,\n",
       " 443.85794,\n",
       " 516.4765,\n",
       " 558.61945,\n",
       " 453.47025,\n",
       " 593.2853,\n",
       " 611.0213,\n",
       " 542.9027,\n",
       " 557.48175,\n",
       " 582.6229,\n",
       " 614.9456,\n",
       " 536.2225,\n",
       " 545.9276,\n",
       " 534.3977,\n",
       " 526.42896,\n",
       " 527.2378,\n",
       " 590.1236,\n",
       " 533.128,\n",
       " 522.2237,\n",
       " 591.35126,\n",
       " 571.922,\n",
       " 479.2735,\n",
       " 515.49115,\n",
       " 548.646,\n",
       " 564.8672,\n",
       " 576.6278,\n",
       " 535.3079,\n",
       " 579.3602,\n",
       " 560.25916,\n",
       " 548.2667,\n",
       " 511.54263,\n",
       " 533.09424,\n",
       " 538.5911,\n",
       " 554.512,\n",
       " 567.31604,\n",
       " 620.01697,\n",
       " 543.2202,\n",
       " 588.3278,\n",
       " 473.1441,\n",
       " 601.6133,\n",
       " 553.7058,\n",
       " 576.3743,\n",
       " 555.8202,\n",
       " 542.80457,\n",
       " 593.7231,\n",
       " 580.88885,\n",
       " 596.0504,\n",
       " 581.40564,\n",
       " 558.8374,\n",
       " 547.3252,\n",
       " 563.24835,\n",
       " 558.4266,\n",
       " 533.7048,\n",
       " 561.4748,\n",
       " 604.67163,\n",
       " 630.9177,\n",
       " 565.1645,\n",
       " 548.4188,\n",
       " 620.99225,\n",
       " 541.9731,\n",
       " 581.56366,\n",
       " 632.2404,\n",
       " 602.663,\n",
       " 619.90857,\n",
       " 567.9741,\n",
       " 561.1049,\n",
       " 571.87756,\n",
       " 579.15717,\n",
       " 576.8972,\n",
       " 587.5248,\n",
       " 615.6564,\n",
       " 608.8748,\n",
       " 578.1639,\n",
       " 577.6459,\n",
       " 642.94525,\n",
       " 586.519,\n",
       " 611.6782,\n",
       " 562.4617,\n",
       " 567.7759,\n",
       " 561.40765,\n",
       " 613.2177,\n",
       " 581.586,\n",
       " 607.86487,\n",
       " 573.4462,\n",
       " 638.627,\n",
       " 562.8234,\n",
       " 663.90424,\n",
       " 605.06573,\n",
       " 580.9453,\n",
       " 590.09955,\n",
       " 637.28406,\n",
       " 544.768,\n",
       " 559.7835,\n",
       " 560.2976,\n",
       " 613.1471,\n",
       " 600.3852,\n",
       " 585.48956,\n",
       " 622.2964,\n",
       " 566.34296,\n",
       " 595.0407,\n",
       " 576.69977,\n",
       " 613.5473,\n",
       " 598.456,\n",
       " 541.5915,\n",
       " 614.71326,\n",
       " 578.37933,\n",
       " 622.07574,\n",
       " 594.4385,\n",
       " 569.8724,\n",
       " 566.76495,\n",
       " 625.6496,\n",
       " 604.239,\n",
       " 611.2271,\n",
       " 589.66034,\n",
       " 621.97174,\n",
       " 612.7445,\n",
       " 605.9123,\n",
       " 616.1648,\n",
       " 586.18756,\n",
       " 577.9001,\n",
       " 632.1398,\n",
       " 583.3548,\n",
       " 593.36084,\n",
       " 567.71826,\n",
       " 564.69836,\n",
       " 573.4977,\n",
       " 549.2645,\n",
       " 539.1035,\n",
       " 561.17426,\n",
       " 629.3725,\n",
       " 578.3965,\n",
       " 625.61804,\n",
       " 588.24634,\n",
       " 636.87854,\n",
       " 617.84796,\n",
       " 602.1979,\n",
       " 638.1328,\n",
       " 583.50903,\n",
       " 626.1565,\n",
       " 607.8954,\n",
       " 591.74884,\n",
       " 637.4167,\n",
       " 633.01294,\n",
       " 615.65607,\n",
       " 610.74805,\n",
       " 573.03046,\n",
       " 578.71234,\n",
       " 588.2888,\n",
       " 578.79767,\n",
       " 597.7993,\n",
       " 625.487,\n",
       " 593.7841,\n",
       " 609.569,\n",
       " 594.92145,\n",
       " 621.8896,\n",
       " 598.76685,\n",
       " 594.0497,\n",
       " 557.3493,\n",
       " 573.31995,\n",
       " 562.6986,\n",
       " 585.93396,\n",
       " 556.88257,\n",
       " 545.37036,\n",
       " 591.28345,\n",
       " 577.2283,\n",
       " 576.3054,\n",
       " 573.9989,\n",
       " 649.9996,\n",
       " 597.1403,\n",
       " 579.9391,\n",
       " 579.75165,\n",
       " 584.72424,\n",
       " 618.7915,\n",
       " 586.07544,\n",
       " 608.6901,\n",
       " 569.8794,\n",
       " 588.7984,\n",
       " 572.9509,\n",
       " 577.77905,\n",
       " 599.519,\n",
       " 598.1185,\n",
       " 613.3627,\n",
       " 556.75476,\n",
       " 576.2788,\n",
       " 559.94366,\n",
       " 583.61224,\n",
       " 595.5082,\n",
       " 607.4884,\n",
       " 606.81213,\n",
       " 577.43646,\n",
       " 559.46826,\n",
       " 583.3832,\n",
       " 567.8205,\n",
       " 600.1301,\n",
       " 571.8975,\n",
       " 585.64557,\n",
       " 589.0961,\n",
       " 606.07275,\n",
       " 564.4518,\n",
       " 566.53156,\n",
       " 599.0673,\n",
       " 609.36615,\n",
       " 642.40204,\n",
       " 629.627,\n",
       " 628.11334,\n",
       " 607.6985,\n",
       " 606.794,\n",
       " 594.56854,\n",
       " 578.19434,\n",
       " 573.26434,\n",
       " 566.3356,\n",
       " 593.43994,\n",
       " 572.324,\n",
       " 577.8001,\n",
       " 596.2307,\n",
       " 588.72375,\n",
       " 566.9082,\n",
       " 568.95026,\n",
       " 568.672,\n",
       " 570.37244,\n",
       " 603.129,\n",
       " 547.177,\n",
       " 593.9706,\n",
       " 600.59216,\n",
       " 612.02454,\n",
       " 583.20465,\n",
       " 558.50946,\n",
       " 604.08673,\n",
       " 554.0044,\n",
       " 576.0435,\n",
       " 543.19385,\n",
       " 579.71155,\n",
       " 583.7739,\n",
       " 586.6981,\n",
       " 555.5591,\n",
       " 577.7784,\n",
       " 605.08453,\n",
       " 608.5974,\n",
       " 596.7225,\n",
       " 577.7956,\n",
       " 587.03217,\n",
       " 575.52844,\n",
       " 632.9736,\n",
       " 588.2889,\n",
       " 583.08954,\n",
       " 588.74225,\n",
       " 568.3083,\n",
       " 594.3206,\n",
       " 558.3752,\n",
       " 570.29816,\n",
       " 582.18335,\n",
       " 583.80994,\n",
       " 580.87006,\n",
       " 617.6594,\n",
       " 576.1858,\n",
       " 572.67816,\n",
       " 588.7637,\n",
       " 569.55176,\n",
       " 586.0334,\n",
       " 579.4984,\n",
       " 583.5955,\n",
       " 581.41504,\n",
       " 607.5738,\n",
       " 602.67126,\n",
       " 609.7907,\n",
       " 601.2816,\n",
       " 581.7332,\n",
       " 583.3409,\n",
       " 634.0036,\n",
       " 609.3309,\n",
       " 595.38293,\n",
       " 593.84827,\n",
       " 587.88196,\n",
       " 590.76935,\n",
       " 612.63293,\n",
       " 603.5564,\n",
       " 592.3097,\n",
       " 606.9257,\n",
       " 585.5871,\n",
       " 554.57153,\n",
       " 586.67395,\n",
       " 595.0395,\n",
       " 591.23517,\n",
       " 628.5761,\n",
       " 611.2902,\n",
       " 627.8143,\n",
       " 611.3655,\n",
       " 584.7391,\n",
       " 595.0684,\n",
       " 564.1544,\n",
       " 598.80615,\n",
       " 605.25793,\n",
       " 585.77905,\n",
       " 591.29175,\n",
       " 767.0866,\n",
       " 557.6368,\n",
       " 577.17395,\n",
       " 551.7853,\n",
       " 541.73834,\n",
       " 558.3162,\n",
       " 567.59015,\n",
       " 565.7356,\n",
       " 562.2047,\n",
       " 623.8123,\n",
       " 600.974,\n",
       " 593.1644,\n",
       " 592.595,\n",
       " 607.7241,\n",
       " 630.9128,\n",
       " 588.3548,\n",
       " 608.5498,\n",
       " 583.3572,\n",
       " 584.3285,\n",
       " 610.6018,\n",
       " 600.0391,\n",
       " 574.9406,\n",
       " 600.7322,\n",
       " 598.49347,\n",
       " 593.0356,\n",
       " 591.88214,\n",
       " 606.4137,\n",
       " 599.247,\n",
       " 617.2277,\n",
       " 597.4046,\n",
       " 624.57404,\n",
       " 614.6634,\n",
       " 600.5329,\n",
       " 601.6305,\n",
       " 601.2658,\n",
       " 594.8759,\n",
       " 632.131,\n",
       " 587.91034,\n",
       " 601.60046,\n",
       " 631.6366,\n",
       " 610.55896,\n",
       " 606.8038,\n",
       " 574.7115,\n",
       " 576.8481,\n",
       " 587.67206,\n",
       " 598.30206,\n",
       " 594.3152,\n",
       " 583.6887,\n",
       " 556.2255,\n",
       " 577.1751,\n",
       " 550.00946,\n",
       " 559.4888,\n",
       " 596.3235,\n",
       " 599.8324,\n",
       " 607.1552,\n",
       " 606.1908,\n",
       " 607.1122,\n",
       " 624.89124,\n",
       " 591.2851,\n",
       " 599.67175,\n",
       " 609.4412,\n",
       " 591.4433,\n",
       " 595.00854,\n",
       " 595.0965,\n",
       " 611.0584,\n",
       " 594.6748,\n",
       " 612.6988,\n",
       " 610.3344,\n",
       " 595.98694,\n",
       " 574.87726,\n",
       " 609.6597,\n",
       " 596.57416,\n",
       " 616.1998,\n",
       " 598.94635,\n",
       " 584.0036,\n",
       " 590.2203,\n",
       " 605.894,\n",
       " 596.36523,\n",
       " 577.6794,\n",
       " 571.25073,\n",
       " 582.4644,\n",
       " 584.25745,\n",
       " 605.3829,\n",
       " 580.0642,\n",
       " 595.8745,\n",
       " 638.70215,\n",
       " 617.22144,\n",
       " 598.18646,\n",
       " 625.8206,\n",
       " 603.9856,\n",
       " 575.9135,\n",
       " 601.54724,\n",
       " 592.36334,\n",
       " 595.1613,\n",
       " 603.6184,\n",
       " 633.08167,\n",
       " 582.8215,\n",
       " 607.78424,\n",
       " 577.97516,\n",
       " 595.2504,\n",
       " 594.2716,\n",
       " 560.1764,\n",
       " 592.15533,\n",
       " 588.0622,\n",
       " 590.6476,\n",
       " 553.4494,\n",
       " 578.12317,\n",
       " 560.89435,\n",
       " 564.66943,\n",
       " 577.69,\n",
       " 606.79675,\n",
       " 576.2629,\n",
       " 602.7022,\n",
       " 571.38666,\n",
       " 576.4953,\n",
       " 584.2511,\n",
       " 550.15234,\n",
       " 569.81256,\n",
       " 586.9026,\n",
       " 608.8956,\n",
       " 595.48193,\n",
       " 599.1301,\n",
       " 572.8237,\n",
       " 594.42377,\n",
       " 620.3095,\n",
       " 606.59454,\n",
       " 592.63544,\n",
       " 583.675,\n",
       " 582.93115,\n",
       " 562.2666,\n",
       " 570.08563,\n",
       " 581.0036,\n",
       " 567.9621,\n",
       " 589.08417,\n",
       " 588.8613,\n",
       " 542.98157,\n",
       " 563.7623,\n",
       " 599.06775,\n",
       " 585.90094,\n",
       " 594.96857,\n",
       " 595.6431,\n",
       " 587.22815,\n",
       " 582.6022,\n",
       " 570.65204,\n",
       " 602.71545,\n",
       " 566.5803,\n",
       " 559.11334,\n",
       " 589.2023,\n",
       " 581.7774,\n",
       " 597.43335,\n",
       " 585.16046,\n",
       " 588.23535,\n",
       " 590.16473,\n",
       " 621.37695,\n",
       " 598.59,\n",
       " 620.94147,\n",
       " 610.56165,\n",
       " 609.33936,\n",
       " 617.1297,\n",
       " 605.1182,\n",
       " 597.76636,\n",
       " 634.3626,\n",
       " 580.59753,\n",
       " 600.2832,\n",
       " 622.73944,\n",
       " 602.51624,\n",
       " 594.0921,\n",
       " 593.0874,\n",
       " 607.4351,\n",
       " 602.3769,\n",
       " 627.7338,\n",
       " 614.03625,\n",
       " 627.00116,\n",
       " 581.5829,\n",
       " 600.7948,\n",
       " 600.8758,\n",
       " 600.9721,\n",
       " 578.4328,\n",
       " 604.9252,\n",
       " 606.2366,\n",
       " 590.9418,\n",
       " 614.1752,\n",
       " 585.7169,\n",
       " 586.1748,\n",
       " 580.27014,\n",
       " 596.2416,\n",
       " 589.4662,\n",
       " 589.2825,\n",
       " 596.68494,\n",
       " 595.29846,\n",
       " 590.2216,\n",
       " 603.95825,\n",
       " 613.94055,\n",
       " 590.298,\n",
       " 606.7235,\n",
       " 608.3599,\n",
       " 597.449,\n",
       " 593.3076,\n",
       " 591.65607,\n",
       " 599.12177,\n",
       " 606.31696,\n",
       " 621.4122,\n",
       " 637.4838,\n",
       " 603.77356,\n",
       " 631.3681,\n",
       " 605.08264,\n",
       " 602.28687,\n",
       " 610.1677,\n",
       " 587.97974,\n",
       " 605.3458,\n",
       " 599.29254,\n",
       " 591.9614,\n",
       " 595.10706,\n",
       " 617.60126,\n",
       " 593.2516,\n",
       " 591.68964,\n",
       " 556.4929,\n",
       " 599.9718,\n",
       " 624.1411,\n",
       " 583.51227,\n",
       " 590.53613,\n",
       " 575.8523,\n",
       " 571.4987,\n",
       " 549.5745,\n",
       " 579.6422,\n",
       " 577.81396,\n",
       " 598.8166,\n",
       " 587.1216,\n",
       " 563.3337,\n",
       " 574.3609,\n",
       " 586.8767,\n",
       " 618.7336,\n",
       " 594.15643,\n",
       " 659.50494,\n",
       " 604.51843,\n",
       " 625.7079,\n",
       " 633.6623,\n",
       " 616.8372,\n",
       " 624.6629,\n",
       " 599.76154,\n",
       " 595.22943,\n",
       " 625.88446,\n",
       " 589.7921,\n",
       " 604.5834,\n",
       " 602.21857,\n",
       " 588.2268,\n",
       " 609.16174,\n",
       " 603.5296,\n",
       " 608.48944,\n",
       " 595.5223,\n",
       " 578.8171,\n",
       " 609.3275,\n",
       " 620.12427,\n",
       " 586.7924,\n",
       " 572.979,\n",
       " 593.4879,\n",
       " 613.05615,\n",
       " 588.5852,\n",
       " 597.9513,\n",
       " 597.22595,\n",
       " 594.6022,\n",
       " 603.16254,\n",
       " 596.2217,\n",
       " 609.2917,\n",
       " 592.2886,\n",
       " 608.83765,\n",
       " 606.17535,\n",
       " 588.99396,\n",
       " 594.7691,\n",
       " 613.85675,\n",
       " 613.9241,\n",
       " 600.62744,\n",
       " 598.4388,\n",
       " 590.19604,\n",
       " 587.15643,\n",
       " 598.8617,\n",
       " 593.81683,\n",
       " 589.41547,\n",
       " 599.6092,\n",
       " 581.91516,\n",
       " 590.12476,\n",
       " 572.6863,\n",
       " 603.9342,\n",
       " 587.7961,\n",
       " 605.40753,\n",
       " 589.5576,\n",
       " 591.68774,\n",
       " 595.5967,\n",
       " 602.55035,\n",
       " 628.7618,\n",
       " 619.46375,\n",
       " 597.4617,\n",
       " 605.2498,\n",
       " 599.7914,\n",
       " 578.68646,\n",
       " 590.67676,\n",
       " 596.931,\n",
       " 599.7473,\n",
       " 591.4239,\n",
       " 607.063,\n",
       " 579.8323,\n",
       " 577.52997,\n",
       " 595.6113,\n",
       " 592.67505,\n",
       " 574.98376,\n",
       " 597.38043,\n",
       " 582.62286,\n",
       " 586.5602,\n",
       " 584.69055,\n",
       " 597.38086,\n",
       " 602.186,\n",
       " 609.71967,\n",
       " 600.45905,\n",
       " 602.44617,\n",
       " 595.775,\n",
       " 612.5316,\n",
       " 607.1913,\n",
       " 604.7421,\n",
       " 609.8055,\n",
       " 607.0114,\n",
       " 599.0925,\n",
       " 590.87335,\n",
       " 611.8175,\n",
       " 606.3611,\n",
       " 609.7774,\n",
       " 599.9677,\n",
       " 598.4219,\n",
       " 609.90955,\n",
       " 617.88464,\n",
       " 613.4009,\n",
       " 594.8819,\n",
       " 602.75244,\n",
       " 611.1773,\n",
       " 579.3006,\n",
       " 584.42017,\n",
       " 597.3133,\n",
       " 602.38513,\n",
       " 584.36774,\n",
       " 567.59875,\n",
       " 590.5491,\n",
       " 579.0808,\n",
       " 585.5297,\n",
       " 584.1633,\n",
       " 596.8593,\n",
       " 602.37225,\n",
       " 605.15906,\n",
       " 604.8665,\n",
       " 593.8446,\n",
       " 614.2987,\n",
       " 605.0187,\n",
       " 597.5653,\n",
       " 604.6741,\n",
       " 606.8673,\n",
       " 622.333,\n",
       " 610.62573,\n",
       " 606.7,\n",
       " 606.0821,\n",
       " 607.04297,\n",
       " 603.9857,\n",
       " 589.1999,\n",
       " 585.5633,\n",
       " 610.8363,\n",
       " 630.9389,\n",
       " 604.04016,\n",
       " 576.87274,\n",
       " 587.07635,\n",
       " 579.7195,\n",
       " 575.39764,\n",
       " 616.1171,\n",
       " 604.5982,\n",
       " 611.30115,\n",
       " 610.3358,\n",
       " 603.31506,\n",
       " 605.3184,\n",
       " 604.731,\n",
       " 610.27844,\n",
       " 606.66113,\n",
       " 607.0991,\n",
       " 610.61,\n",
       " 609.18384,\n",
       " 616.26794,\n",
       " 613.99835,\n",
       " 593.8772,\n",
       " 600.7387,\n",
       " 608.3222,\n",
       " 597.18646,\n",
       " 596.23535,\n",
       " 602.2974,\n",
       " 605.6691,\n",
       " 604.0037,\n",
       " 608.0404,\n",
       " 582.13245,\n",
       " 598.5891,\n",
       " 593.8605,\n",
       " 613.354,\n",
       " 605.14386,\n",
       " 626.0077,\n",
       " 612.9217,\n",
       " 598.5608,\n",
       " 604.1714,\n",
       " 607.8055,\n",
       " 588.9269,\n",
       " 590.022,\n",
       " 589.82135,\n",
       " 579.68567,\n",
       " 602.34796,\n",
       " 599.20276,\n",
       " 606.54816,\n",
       " 608.4639,\n",
       " 601.3269,\n",
       " 608.343,\n",
       " 606.9259,\n",
       " 600.4072,\n",
       " 612.4666,\n",
       " 610.4849,\n",
       " 611.9089,\n",
       " 597.06287,\n",
       " 609.8536,\n",
       " 607.4464,\n",
       " 601.81494,\n",
       " 610.6309,\n",
       " 617.434,\n",
       " 601.79364,\n",
       " 604.426,\n",
       " 605.1932,\n",
       " 600.7889,\n",
       " 614.23724,\n",
       " 602.3726,\n",
       " 613.25616,\n",
       " 631.8823,\n",
       " 608.0648,\n",
       " 608.60706,\n",
       " 613.9691,\n",
       " 622.6462,\n",
       " 622.4017,\n",
       " 610.602,\n",
       " 611.44763,\n",
       " 592.39984,\n",
       " 607.17224,\n",
       " 611.7038,\n",
       " 595.02844,\n",
       " 613.07764,\n",
       " 615.85333,\n",
       " 626.3268,\n",
       " 630.25653,\n",
       " 637.5989,\n",
       " 626.2422,\n",
       " 614.3787,\n",
       " 617.97675,\n",
       " 614.3458,\n",
       " 614.22424,\n",
       " 627.55865,\n",
       " 605.20074,\n",
       " 616.1984,\n",
       " 620.656,\n",
       " 609.83563,\n",
       " 586.4494,\n",
       " 585.36755,\n",
       " 606.4822,\n",
       " 601.1391,\n",
       " 610.217,\n",
       " 616.2967,\n",
       " 609.54755,\n",
       " 616.78296,\n",
       " 609.7026,\n",
       " 604.47064,\n",
       " 625.2051,\n",
       " 620.3998,\n",
       " 621.2052,\n",
       " 624.23175,\n",
       " 617.4486,\n",
       " 617.75146,\n",
       " 612.4227,\n",
       " 607.17035,\n",
       " 586.3845,\n",
       " 589.9935,\n",
       " 604.9006,\n",
       " 600.66064,\n",
       " 606.63666,\n",
       " 603.7494,\n",
       " 611.8197,\n",
       " 613.46075,\n",
       " 628.147,\n",
       " 608.5467,\n",
       " 607.18604,\n",
       " 601.2524,\n",
       " 610.59424,\n",
       " 615.7897,\n",
       " 618.5977,\n",
       " 617.75946,\n",
       " 619.53864,\n",
       " 625.8255,\n",
       " 594.8929,\n",
       " 606.98944,\n",
       " 610.1679,\n",
       " 590.0358,\n",
       " 601.2285,\n",
       " 594.664,\n",
       " 612.30457,\n",
       " 600.1455,\n",
       " 621.2033,\n",
       " 600.41077,\n",
       " 590.9619,\n",
       " 611.9873,\n",
       " 599.59204,\n",
       " 604.07056,\n",
       " 608.86664,\n",
       " 598.4178,\n",
       " 594.0226,\n",
       " 602.6933,\n",
       " 612.5341,\n",
       " 621.84656,\n",
       " 612.6576,\n",
       " 619.91644,\n",
       " 629.31976,\n",
       " 613.56335,\n",
       " 603.437,\n",
       " 609.545,\n",
       " 603.33606,\n",
       " 590.0431,\n",
       " 599.3795,\n",
       " 591.2315,\n",
       " 587.47284,\n",
       " 600.07214,\n",
       " 601.3886,\n",
       " 603.1443,\n",
       " 614.66144,\n",
       " 607.02075,\n",
       " 608.8737,\n",
       " 603.65125,\n",
       " 599.9822,\n",
       " 603.16125,\n",
       " 599.711,\n",
       " 596.54645,\n",
       " 590.65265,\n",
       " 599.1151,\n",
       " 605.0192,\n",
       " 617.7971,\n",
       " 603.2543,\n",
       " 591.4603,\n",
       " 603.0306,\n",
       " 589.8033,\n",
       " 610.25775,\n",
       " 594.56757,\n",
       " 613.197,\n",
       " 605.2776,\n",
       " 590.6026,\n",
       " 599.5292,\n",
       " 590.9795,\n",
       " 585.518,\n",
       " 596.93134,\n",
       " 587.0669,\n",
       " 591.83716,\n",
       " 617.1286,\n",
       " 598.64984,\n",
       " 617.9906,\n",
       " 601.44855,\n",
       " 601.5345,\n",
       " 596.72797,\n",
       " 591.95953,\n",
       " 595.16583,\n",
       " 589.81104,\n",
       " 607.25104,\n",
       " 595.4563,\n",
       " 600.66943,\n",
       " 612.5229,\n",
       " 606.5016,\n",
       " 603.486,\n",
       " 615.2553,\n",
       " 612.90594,\n",
       " 620.8859,\n",
       " 621.27264,\n",
       " 620.10376,\n",
       " 614.82434,\n",
       " 604.1102,\n",
       " 597.9068,\n",
       " 585.84204,\n",
       " 586.0825,\n",
       " 596.23346,\n",
       " 594.9891,\n",
       " 584.9614,\n",
       " 614.4181,\n",
       " 602.3153,\n",
       " 617.8298,\n",
       " 603.0285,\n",
       " 599.27246,\n",
       " 600.5537,\n",
       " 594.61316,\n",
       " 596.19916,\n",
       " 602.1044,\n",
       " 599.85956,\n",
       " 590.9826,\n",
       " 587.5151,\n",
       " 600.8741,\n",
       " 609.17914,\n",
       " 604.69116,\n",
       " 592.38214,\n",
       " 591.9195,\n",
       " 601.51166,\n",
       " 597.96063,\n",
       " 596.5335,\n",
       " 600.69446,\n",
       " 605.36816,\n",
       " 599.7285,\n",
       " 613.3443,\n",
       " 620.0521,\n",
       " 609.867,\n",
       " 627.3955,\n",
       " 609.77454,\n",
       " 607.6447,\n",
       " 599.18634,\n",
       " 612.023,\n",
       " 613.8544,\n",
       " 599.9279,\n",
       " 595.38873,\n",
       " 598.69086,\n",
       " 592.5133,\n",
       " 591.35803,\n",
       " 602.9775,\n",
       " 589.3773,\n",
       " 604.8331,\n",
       " 605.6196,\n",
       " 606.3737,\n",
       " 614.32776,\n",
       " 597.0913,\n",
       " 588.5864,\n",
       " 595.35095,\n",
       " 589.634,\n",
       " 582.61835,\n",
       " 606.72095,\n",
       " 594.8306,\n",
       " 605.44775,\n",
       " 600.5194,\n",
       " 609.66315,\n",
       " 598.9296,\n",
       " 597.4522,\n",
       " 590.5011,\n",
       " 595.8609,\n",
       " 603.0062,\n",
       " 606.4938,\n",
       " 609.17957,\n",
       " 610.2626,\n",
       " 603.38983,\n",
       " 612.00977,\n",
       " 594.18164,\n",
       " 591.5206,\n",
       " 606.8228,\n",
       " 605.8482,\n",
       " 583.0888,\n",
       " 590.8722,\n",
       " 596.9343,\n",
       " 636.1234,\n",
       " 629.8531,\n",
       " 606.9331,\n",
       " 606.1643,\n",
       " 599.10175,\n",
       " 608.90186,\n",
       " 600.04675,\n",
       " 593.32074,\n",
       " 586.71466,\n",
       " 591.59827,\n",
       " 581.9241,\n",
       " 577.38043,\n",
       " 586.78235,\n",
       " 588.0761,\n",
       " 596.2516,\n",
       " 598.93713,\n",
       " 613.44104,\n",
       " 594.0091,\n",
       " 594.7266,\n",
       " 600.0808,\n",
       " 604.262,\n",
       " 600.8419,\n",
       " 598.1429,\n",
       " ...]"
      ]
     },
     "execution_count": 20,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# For Sample state_action pair 1: State (2,4,6), Action (2,3)\n",
    "agent.states_tracked_1"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "[0.0,\n",
       " 0.0,\n",
       " 0.0,\n",
       " 0.0,\n",
       " 0.0,\n",
       " 0.0,\n",
       " 0.0,\n",
       " 0.0,\n",
       " 0.0,\n",
       " 0.0,\n",
       " 0.0,\n",
       " 0.0,\n",
       " 0.0,\n",
       " 0.0,\n",
       " 0.0,\n",
       " 0.0,\n",
       " 0.0,\n",
       " 0.0,\n",
       " 0.0,\n",
       " 0.0,\n",
       " 0.0,\n",
       " 0.0,\n",
       " 0.0,\n",
       " 0.0,\n",
       " 0.0,\n",
       " 480.48544,\n",
       " 313.78128,\n",
       " 365.62976,\n",
       " 396.77023,\n",
       " 419.49426,\n",
       " 505.84085,\n",
       " 492.5905,\n",
       " 492.58725,\n",
       " 462.485,\n",
       " 522.4017,\n",
       " 591.9612,\n",
       " 462.64716,\n",
       " 624.704,\n",
       " 650.055,\n",
       " 571.94226,\n",
       " 566.78156,\n",
       " 589.99677,\n",
       " 647.0921,\n",
       " 539.3229,\n",
       " 551.6705,\n",
       " 554.74536,\n",
       " 555.41327,\n",
       " 568.5403,\n",
       " 612.3995,\n",
       " 580.40894,\n",
       " 548.5284,\n",
       " 621.78516,\n",
       " 600.0667,\n",
       " 510.15616,\n",
       " 553.15076,\n",
       " 600.5629,\n",
       " 573.95026,\n",
       " 569.43823,\n",
       " 551.8759,\n",
       " 583.38367,\n",
       " 581.451,\n",
       " 590.56665,\n",
       " 548.2514,\n",
       " 562.9753,\n",
       " 557.1728,\n",
       " 584.77985,\n",
       " 597.7382,\n",
       " 629.89825,\n",
       " 555.15106,\n",
       " 590.4082,\n",
       " 499.02914,\n",
       " 621.7016,\n",
       " 600.00555,\n",
       " 608.7267,\n",
       " 589.4367,\n",
       " 599.58453,\n",
       " 621.53955,\n",
       " 612.6231,\n",
       " 610.64703,\n",
       " 559.38055,\n",
       " 564.5657,\n",
       " 553.4553,\n",
       " 573.7744,\n",
       " 573.2059,\n",
       " 542.1269,\n",
       " 567.1982,\n",
       " 611.1668,\n",
       " 655.7537,\n",
       " 600.5758,\n",
       " 566.9176,\n",
       " 618.2054,\n",
       " 530.1247,\n",
       " 598.5604,\n",
       " 633.34326,\n",
       " 589.8481,\n",
       " 632.2058,\n",
       " 573.14514,\n",
       " 573.4387,\n",
       " 575.1856,\n",
       " 607.4992,\n",
       " 579.5916,\n",
       " 600.59766,\n",
       " 614.64606,\n",
       " 621.7975,\n",
       " 606.46027,\n",
       " 581.336,\n",
       " 662.34344,\n",
       " 602.065,\n",
       " 623.1874,\n",
       " 575.2514,\n",
       " 569.1628,\n",
       " 576.4675,\n",
       " 621.425,\n",
       " 598.43604,\n",
       " 633.51825,\n",
       " 604.1867,\n",
       " 653.951,\n",
       " 555.87775,\n",
       " 662.55475,\n",
       " 599.7032,\n",
       " 597.5866,\n",
       " 604.4465,\n",
       " 646.83374,\n",
       " 560.0979,\n",
       " 566.50806,\n",
       " 554.3069,\n",
       " 621.2626,\n",
       " 592.89557,\n",
       " 604.36035,\n",
       " 627.23804,\n",
       " 597.86383,\n",
       " 577.3367,\n",
       " 593.54114,\n",
       " 627.99426,\n",
       " 605.39545,\n",
       " 551.2226,\n",
       " 608.35315,\n",
       " 571.2896,\n",
       " 608.05804,\n",
       " 581.10175,\n",
       " 555.68396,\n",
       " 571.0023,\n",
       " 624.3528,\n",
       " 602.55396,\n",
       " 609.6292,\n",
       " 599.0879,\n",
       " 628.5892,\n",
       " 609.70325,\n",
       " 625.5856,\n",
       " 622.6517,\n",
       " 608.1703,\n",
       " 590.2977,\n",
       " 620.2877,\n",
       " 596.5211,\n",
       " 608.61096,\n",
       " 571.68726,\n",
       " 573.0829,\n",
       " 598.36584,\n",
       " 595.49713,\n",
       " 556.5772,\n",
       " 583.35706,\n",
       " 627.9019,\n",
       " 596.6294,\n",
       " 616.4064,\n",
       " 604.2416,\n",
       " 642.20624,\n",
       " 634.1786,\n",
       " 614.6822,\n",
       " 620.82196,\n",
       " 581.18823,\n",
       " 606.13617,\n",
       " 621.8485,\n",
       " 601.44995,\n",
       " 634.0965,\n",
       " 626.1929,\n",
       " 617.4986,\n",
       " 614.4285,\n",
       " 583.5618,\n",
       " 592.0913,\n",
       " 608.7799,\n",
       " 596.0856,\n",
       " 593.6793,\n",
       " 632.3872,\n",
       " 579.65594,\n",
       " 604.096,\n",
       " 601.0938,\n",
       " 641.266,\n",
       " 638.4989,\n",
       " 610.4974,\n",
       " 578.4901,\n",
       " 602.58826,\n",
       " 582.9563,\n",
       " 610.22235,\n",
       " 575.8774,\n",
       " 562.6879,\n",
       " 599.6736,\n",
       " 584.53467,\n",
       " 585.7872,\n",
       " 567.23016,\n",
       " 646.55164,\n",
       " 600.04333,\n",
       " 592.44244,\n",
       " 592.2057,\n",
       " 612.8509,\n",
       " 651.84424,\n",
       " 606.27386,\n",
       " 621.32544,\n",
       " 582.71814,\n",
       " 595.6933,\n",
       " 583.6277,\n",
       " 589.1598,\n",
       " 592.3662,\n",
       " 604.2049,\n",
       " 622.2233,\n",
       " 558.50354,\n",
       " 577.5024,\n",
       " 568.70746,\n",
       " 584.8787,\n",
       " 606.65063,\n",
       " 609.0546,\n",
       " 605.2865,\n",
       " 593.6273,\n",
       " 563.3702,\n",
       " 579.38904,\n",
       " 587.112,\n",
       " 606.3108,\n",
       " 576.07654,\n",
       " 575.8717,\n",
       " 594.21967,\n",
       " 593.2572,\n",
       " 565.6159,\n",
       " 550.03143,\n",
       " 601.3721,\n",
       " 588.9773,\n",
       " 626.18536,\n",
       " 614.9428,\n",
       " 620.28577,\n",
       " 588.7226,\n",
       " 596.4232,\n",
       " 597.65564,\n",
       " 593.31195,\n",
       " 590.59033,\n",
       " 573.0985,\n",
       " 606.20355,\n",
       " 587.951,\n",
       " 592.0919,\n",
       " 613.8117,\n",
       " 595.147,\n",
       " 574.4957,\n",
       " 572.3481,\n",
       " 564.975,\n",
       " 571.64453,\n",
       " 590.69977,\n",
       " 531.66364,\n",
       " 583.25653,\n",
       " 601.60425,\n",
       " 594.78455,\n",
       " 585.6093,\n",
       " 560.37555,\n",
       " 602.1355,\n",
       " 559.1571,\n",
       " 584.12665,\n",
       " 571.47034,\n",
       " 585.7782,\n",
       " 600.8886,\n",
       " 596.8388,\n",
       " 565.71356,\n",
       " 599.1428,\n",
       " 602.75244,\n",
       " 612.2392,\n",
       " 606.892,\n",
       " 600.21545,\n",
       " 611.9579,\n",
       " 613.8601,\n",
       " 658.2645,\n",
       " 621.5839,\n",
       " 610.9867,\n",
       " 621.9147,\n",
       " 584.576,\n",
       " 601.9009,\n",
       " 562.25665,\n",
       " 586.1727,\n",
       " 602.6263,\n",
       " 595.71136,\n",
       " 604.9486,\n",
       " 619.0232,\n",
       " 588.31714,\n",
       " 570.38684,\n",
       " 598.79175,\n",
       " 576.87646,\n",
       " 584.8727,\n",
       " 597.11707,\n",
       " 594.12683,\n",
       " 585.8612,\n",
       " 609.3297,\n",
       " 612.76794,\n",
       " 608.4358,\n",
       " 627.40265,\n",
       " 593.4875,\n",
       " 622.2298,\n",
       " 645.7899,\n",
       " 629.74194,\n",
       " 599.02045,\n",
       " 598.14484,\n",
       " 604.0825,\n",
       " 596.646,\n",
       " 629.22675,\n",
       " 620.375,\n",
       " 603.555,\n",
       " 626.2817,\n",
       " 584.8819,\n",
       " 558.0202,\n",
       " 596.6611,\n",
       " 620.12317,\n",
       " 585.43787,\n",
       " 615.8865,\n",
       " 607.073,\n",
       " 622.40216,\n",
       " 614.8524,\n",
       " 587.07196,\n",
       " 589.5751,\n",
       " 557.99,\n",
       " 616.66766,\n",
       " 615.1822,\n",
       " 603.5987,\n",
       " 585.54285,\n",
       " 788.3868,\n",
       " 590.8267,\n",
       " 599.8523,\n",
       " 570.0204,\n",
       " 549.90265,\n",
       " 571.6908,\n",
       " 586.0981,\n",
       " 578.4237,\n",
       " 580.54535,\n",
       " 630.5663,\n",
       " 611.582,\n",
       " 588.88745,\n",
       " 591.7197,\n",
       " 598.6003,\n",
       " 614.7659,\n",
       " 591.45044,\n",
       " 594.4158,\n",
       " 600.9173,\n",
       " 597.7828,\n",
       " 612.41626,\n",
       " 596.9888,\n",
       " 575.3855,\n",
       " 589.51,\n",
       " 587.1986,\n",
       " 583.97864,\n",
       " 582.56433,\n",
       " 606.67194,\n",
       " 611.76465,\n",
       " 596.60065,\n",
       " 632.9276,\n",
       " 632.07513,\n",
       " 610.612,\n",
       " 594.7133,\n",
       " 606.7423,\n",
       " 600.9904,\n",
       " 588.94727,\n",
       " 631.4646,\n",
       " 596.4399,\n",
       " 614.9853,\n",
       " 646.4436,\n",
       " 612.3059,\n",
       " 608.30255,\n",
       " 572.171,\n",
       " 592.5107,\n",
       " 610.3755,\n",
       " 602.23486,\n",
       " 605.1975,\n",
       " 601.72833,\n",
       " 574.45605,\n",
       " 589.0228,\n",
       " 564.67065,\n",
       " 571.45087,\n",
       " 603.33325,\n",
       " 585.4221,\n",
       " 588.8625,\n",
       " 587.7011,\n",
       " 599.5462,\n",
       " 607.7983,\n",
       " 588.56287,\n",
       " 589.8523,\n",
       " 613.45074,\n",
       " 589.8783,\n",
       " 595.446,\n",
       " 585.4034,\n",
       " 609.50275,\n",
       " 588.2479,\n",
       " 612.3843,\n",
       " 615.6559,\n",
       " 582.29614,\n",
       " 582.3169,\n",
       " 605.6754,\n",
       " 596.28577,\n",
       " 611.934,\n",
       " 595.61786,\n",
       " 583.7258,\n",
       " 602.56946,\n",
       " 609.4661,\n",
       " 592.71796,\n",
       " 581.01575,\n",
       " 581.98315,\n",
       " 586.3154,\n",
       " 581.84973,\n",
       " 599.38995,\n",
       " 569.3347,\n",
       " 584.792,\n",
       " 637.7367,\n",
       " 611.3412,\n",
       " 604.93445,\n",
       " 618.713,\n",
       " 615.21533,\n",
       " 578.5331,\n",
       " 603.35175,\n",
       " 591.62115,\n",
       " 596.1533,\n",
       " 603.6497,\n",
       " 625.4529,\n",
       " 586.7969,\n",
       " 609.4388,\n",
       " 579.74866,\n",
       " 596.1961,\n",
       " 595.2798,\n",
       " 569.33997,\n",
       " 605.22656,\n",
       " 613.1846,\n",
       " 618.3086,\n",
       " 581.7059,\n",
       " 586.4,\n",
       " 567.9693,\n",
       " 578.06946,\n",
       " 597.0409,\n",
       " 612.2919,\n",
       " 581.5331,\n",
       " 610.59735,\n",
       " 587.4134,\n",
       " 604.6472,\n",
       " 588.55975,\n",
       " 552.5046,\n",
       " 573.30554,\n",
       " 600.662,\n",
       " 618.4148,\n",
       " 612.66327,\n",
       " 615.77405,\n",
       " 593.5266,\n",
       " 617.16565,\n",
       " 630.0402,\n",
       " 611.6336,\n",
       " 612.4522,\n",
       " 595.8028,\n",
       " 578.83905,\n",
       " 568.2126,\n",
       " 575.7591,\n",
       " 587.80927,\n",
       " 572.61633,\n",
       " 594.2894,\n",
       " 599.7111,\n",
       " 552.2056,\n",
       " 581.4476,\n",
       " 604.9945,\n",
       " 598.19995,\n",
       " 594.629,\n",
       " 607.5452,\n",
       " 596.90295,\n",
       " 596.14453,\n",
       " 571.7639,\n",
       " 607.46326,\n",
       " 565.902,\n",
       " 571.7874,\n",
       " 592.7585,\n",
       " 586.2814,\n",
       " 590.10114,\n",
       " 592.5291,\n",
       " 592.33777,\n",
       " 584.66235,\n",
       " 612.31006,\n",
       " 603.27405,\n",
       " 606.63257,\n",
       " 598.9145,\n",
       " 605.468,\n",
       " 607.9367,\n",
       " 604.82623,\n",
       " 606.2883,\n",
       " 632.22424,\n",
       " 576.61505,\n",
       " 592.88837,\n",
       " 604.8095,\n",
       " 590.2972,\n",
       " 582.31714,\n",
       " 592.7724,\n",
       " 607.26465,\n",
       " 604.5691,\n",
       " 621.9355,\n",
       " 617.0414,\n",
       " 617.15607,\n",
       " 587.7715,\n",
       " 605.7374,\n",
       " 608.455,\n",
       " 607.40454,\n",
       " 589.33563,\n",
       " 595.7763,\n",
       " 598.8681,\n",
       " 584.1555,\n",
       " 618.2355,\n",
       " 602.8427,\n",
       " 596.0733,\n",
       " 598.8966,\n",
       " 608.4186,\n",
       " 602.1404,\n",
       " 597.0781,\n",
       " 617.1906,\n",
       " 608.15967,\n",
       " 599.1137,\n",
       " 613.93286,\n",
       " 624.9994,\n",
       " 606.3579,\n",
       " 619.21533,\n",
       " 615.03955,\n",
       " 581.03143,\n",
       " 589.6564,\n",
       " 584.3219,\n",
       " 614.6561,\n",
       " 607.566,\n",
       " 615.1095,\n",
       " 634.4308,\n",
       " 615.2925,\n",
       " 636.7475,\n",
       " 609.3458,\n",
       " 605.45624,\n",
       " 617.2383,\n",
       " 588.693,\n",
       " 606.3242,\n",
       " 591.19824,\n",
       " 601.6129,\n",
       " 603.1627,\n",
       " 622.1903,\n",
       " 600.0574,\n",
       " 617.41486,\n",
       " 561.26355,\n",
       " 604.2602,\n",
       " 624.9065,\n",
       " 584.87427,\n",
       " 599.01056,\n",
       " 585.41003,\n",
       " 579.3137,\n",
       " 573.33105,\n",
       " 595.8331,\n",
       " 597.0331,\n",
       " 582.6785,\n",
       " 568.1472,\n",
       " 554.9687,\n",
       " 570.05865,\n",
       " 578.8536,\n",
       " 608.9865,\n",
       " 589.0476,\n",
       " 648.6541,\n",
       " 608.4593,\n",
       " 635.3643,\n",
       " 624.2938,\n",
       " 609.3177,\n",
       " 611.96564,\n",
       " 607.20636,\n",
       " 594.2493,\n",
       " 616.8333,\n",
       " 595.312,\n",
       " 603.9167,\n",
       " 598.3002,\n",
       " 593.0782,\n",
       " 612.18945,\n",
       " 599.0633,\n",
       " 603.02246,\n",
       " 602.3457,\n",
       " 585.55524,\n",
       " 612.46124,\n",
       " 621.9107,\n",
       " 591.0042,\n",
       " 583.3476,\n",
       " 597.41016,\n",
       " 608.4604,\n",
       " 601.5387,\n",
       " 602.1965,\n",
       " 600.0488,\n",
       " 608.3473,\n",
       " 616.77606,\n",
       " 605.2808,\n",
       " 605.1685,\n",
       " 593.4952,\n",
       " 614.1979,\n",
       " 626.0867,\n",
       " 602.8359,\n",
       " 599.38403,\n",
       " 619.0704,\n",
       " 616.3692,\n",
       " 608.0345,\n",
       " 603.9314,\n",
       " 599.11163,\n",
       " 597.23145,\n",
       " 611.2736,\n",
       " 611.89215,\n",
       " 609.402,\n",
       " 612.36066,\n",
       " 601.01904,\n",
       " 602.4186,\n",
       " 583.5923,\n",
       " 610.8362,\n",
       " 594.74567,\n",
       " 605.0278,\n",
       " 595.92236,\n",
       " 606.5891,\n",
       " 603.29834,\n",
       " 602.7669,\n",
       " 636.1262,\n",
       " 633.0431,\n",
       " 607.2635,\n",
       " 615.73346,\n",
       " 619.53094,\n",
       " 594.3066,\n",
       " 607.7126,\n",
       " 608.2985,\n",
       " 616.1529,\n",
       " 605.37,\n",
       " 622.9934,\n",
       " 595.64276,\n",
       " 593.95593,\n",
       " 606.0812,\n",
       " 596.4553,\n",
       " 597.66626,\n",
       " 617.40674,\n",
       " 603.0598,\n",
       " 602.4002,\n",
       " 598.4441,\n",
       " 606.4438,\n",
       " 612.3135,\n",
       " 616.5485,\n",
       " 609.5416,\n",
       " 611.7906,\n",
       " 602.20465,\n",
       " 610.71124,\n",
       " 598.7269,\n",
       " 608.56714,\n",
       " 601.7046,\n",
       " 609.30896,\n",
       " 601.17865,\n",
       " 585.77,\n",
       " 607.3195,\n",
       " 605.17773,\n",
       " 610.4085,\n",
       " 596.3435,\n",
       " 600.0318,\n",
       " 615.049,\n",
       " 617.78546,\n",
       " 621.099,\n",
       " 593.39716,\n",
       " 589.63324,\n",
       " 596.67773,\n",
       " 583.1802,\n",
       " 597.8362,\n",
       " 602.4768,\n",
       " 605.36786,\n",
       " 589.20917,\n",
       " 588.85876,\n",
       " 603.8064,\n",
       " 583.0137,\n",
       " 598.57697,\n",
       " 593.8562,\n",
       " 599.07153,\n",
       " 608.6884,\n",
       " 619.2928,\n",
       " 600.8282,\n",
       " 599.3087,\n",
       " 610.84546,\n",
       " 602.827,\n",
       " 592.7797,\n",
       " 603.0293,\n",
       " 608.4908,\n",
       " 623.31287,\n",
       " 608.6734,\n",
       " 605.30255,\n",
       " 606.8749,\n",
       " 615.8628,\n",
       " 610.49927,\n",
       " 604.4068,\n",
       " 595.51666,\n",
       " 610.7684,\n",
       " 624.79865,\n",
       " 600.1934,\n",
       " 580.6813,\n",
       " 593.5001,\n",
       " 593.65106,\n",
       " 595.43256,\n",
       " 628.8565,\n",
       " 612.5586,\n",
       " 634.39966,\n",
       " 631.88104,\n",
       " 618.2486,\n",
       " 608.7499,\n",
       " 599.8654,\n",
       " 605.4462,\n",
       " 610.8752,\n",
       " 615.32635,\n",
       " 618.40497,\n",
       " 612.6771,\n",
       " 611.1386,\n",
       " 622.2771,\n",
       " 601.7651,\n",
       " 600.534,\n",
       " 607.8746,\n",
       " 605.4847,\n",
       " 598.1991,\n",
       " 597.54724,\n",
       " 597.7262,\n",
       " 596.2072,\n",
       " 605.94543,\n",
       " 607.8456,\n",
       " 622.07,\n",
       " 601.78046,\n",
       " 623.76154,\n",
       " 611.2885,\n",
       " 626.4016,\n",
       " 612.30414,\n",
       " 600.2133,\n",
       " 606.0605,\n",
       " 597.351,\n",
       " 594.79944,\n",
       " 598.2874,\n",
       " 590.2251,\n",
       " 584.9799,\n",
       " 616.6739,\n",
       " 597.74036,\n",
       " 603.5033,\n",
       " 606.9869,\n",
       " 586.2348,\n",
       " 603.18896,\n",
       " 611.29254,\n",
       " 602.63806,\n",
       " 603.71814,\n",
       " 618.71277,\n",
       " 614.1244,\n",
       " 599.0538,\n",
       " 613.58386,\n",
       " 609.61505,\n",
       " 618.1121,\n",
       " 614.8664,\n",
       " 620.4,\n",
       " 599.19324,\n",
       " 605.653,\n",
       " 613.6422,\n",
       " 617.88116,\n",
       " 611.34467,\n",
       " 613.716,\n",
       " 617.51636,\n",
       " 641.28314,\n",
       " 618.09064,\n",
       " 610.4927,\n",
       " 613.3095,\n",
       " 620.4396,\n",
       " 617.37463,\n",
       " 608.8918,\n",
       " 611.8764,\n",
       " 605.7483,\n",
       " 622.3448,\n",
       " 622.8025,\n",
       " 603.37305,\n",
       " 622.456,\n",
       " 618.23505,\n",
       " 622.9064,\n",
       " 632.9008,\n",
       " 646.77216,\n",
       " 635.37683,\n",
       " 623.3428,\n",
       " 614.6294,\n",
       " 618.898,\n",
       " 616.7627,\n",
       " 623.8013,\n",
       " 605.8037,\n",
       " 617.6028,\n",
       " 619.13416,\n",
       " 605.6877,\n",
       " 597.34247,\n",
       " 604.9611,\n",
       " 620.06586,\n",
       " 603.5922,\n",
       " 606.18243,\n",
       " 621.96606,\n",
       " 619.6167,\n",
       " 617.1491,\n",
       " 617.4554,\n",
       " 623.2956,\n",
       " 629.25867,\n",
       " 623.32275,\n",
       " 627.2702,\n",
       " 625.6294,\n",
       " 617.9857,\n",
       " 618.5404,\n",
       " 607.9927,\n",
       " 603.8693,\n",
       " 588.7701,\n",
       " 594.50586,\n",
       " 609.6623,\n",
       " 605.56,\n",
       " 614.292,\n",
       " 607.1566,\n",
       " 623.3683,\n",
       " 628.88635,\n",
       " 627.2572,\n",
       " 619.94666,\n",
       " 617.5859,\n",
       " 606.78577,\n",
       " 600.98444,\n",
       " 604.3111,\n",
       " 617.2135,\n",
       " 618.97473,\n",
       " 621.2806,\n",
       " 624.377,\n",
       " 605.197,\n",
       " 622.62976,\n",
       " 618.3208,\n",
       " 608.9222,\n",
       " 622.72974,\n",
       " 616.2766,\n",
       " 625.8389,\n",
       " 608.30804,\n",
       " 622.7285,\n",
       " 616.5624,\n",
       " 607.73895,\n",
       " 614.5405,\n",
       " 612.7123,\n",
       " 605.0383,\n",
       " 610.69415,\n",
       " 602.67236,\n",
       " 607.8373,\n",
       " 606.5003,\n",
       " 620.6135,\n",
       " 619.5386,\n",
       " 615.6631,\n",
       " 612.60974,\n",
       " 618.7111,\n",
       " 606.75214,\n",
       " 602.4024,\n",
       " 610.3337,\n",
       " 602.0341,\n",
       " 594.44556,\n",
       " 604.28516,\n",
       " 600.5268,\n",
       " 582.7548,\n",
       " 605.92993,\n",
       " 600.63635,\n",
       " 609.77814,\n",
       " 611.974,\n",
       " 614.904,\n",
       " 619.8056,\n",
       " 612.7465,\n",
       " 604.26465,\n",
       " 612.2819,\n",
       " 605.3794,\n",
       " 612.48145,\n",
       " 594.50867,\n",
       " 616.66406,\n",
       " 607.0476,\n",
       " 627.69824,\n",
       " 610.18463,\n",
       " 592.33844,\n",
       " 607.14044,\n",
       " 602.94446,\n",
       " 615.12177,\n",
       " 602.5146,\n",
       " 613.9813,\n",
       " 616.0861,\n",
       " 597.7431,\n",
       " 602.6604,\n",
       " 602.9379,\n",
       " 602.53815,\n",
       " 609.605,\n",
       " 597.57477,\n",
       " 598.3426,\n",
       " 611.0603,\n",
       " 607.4069,\n",
       " 615.2786,\n",
       " 600.8916,\n",
       " 601.53845,\n",
       " 607.2962,\n",
       " 599.514,\n",
       " 606.5647,\n",
       " 602.2899,\n",
       " 603.38995,\n",
       " 596.6907,\n",
       " 604.86237,\n",
       " 615.3353,\n",
       " 615.6251,\n",
       " 607.744,\n",
       " 602.42346,\n",
       " 602.9128,\n",
       " 613.93335,\n",
       " 630.08264,\n",
       " 626.89014,\n",
       " 630.7625,\n",
       " 614.058,\n",
       " 615.62067,\n",
       " 605.06555,\n",
       " 608.07025,\n",
       " 612.92755,\n",
       " 602.2742,\n",
       " 601.50397,\n",
       " 621.23126,\n",
       " 611.59735,\n",
       " 619.50336,\n",
       " 612.0562,\n",
       " 604.3452,\n",
       " 614.25433,\n",
       " 602.46063,\n",
       " 598.5121,\n",
       " 611.2387,\n",
       " 605.1156,\n",
       " 595.0964,\n",
       " 600.54285,\n",
       " 612.09424,\n",
       " 625.11615,\n",
       " 611.77856,\n",
       " 608.92706,\n",
       " 606.1645,\n",
       " 615.6234,\n",
       " 598.90015,\n",
       " 598.74945,\n",
       " 609.78577,\n",
       " 612.819,\n",
       " 607.48706,\n",
       " 618.5518,\n",
       " 628.0312,\n",
       " 612.16345,\n",
       " 627.1307,\n",
       " 612.80035,\n",
       " 617.09076,\n",
       " 601.9101,\n",
       " 606.5597,\n",
       " 617.109,\n",
       " 605.9999,\n",
       " 607.226,\n",
       " 603.03955,\n",
       " 600.8683,\n",
       " 601.69135,\n",
       " 607.21606,\n",
       " 603.9986,\n",
       " 601.59827,\n",
       " 610.39557,\n",
       " 619.5,\n",
       " 626.40546,\n",
       " 599.7075,\n",
       " 586.3997,\n",
       " 610.4306,\n",
       " 605.77423,\n",
       " 587.44586,\n",
       " 610.2985,\n",
       " 599.4567,\n",
       " 610.85925,\n",
       " 612.17303,\n",
       " 620.3202,\n",
       " 602.4346,\n",
       " 597.83246,\n",
       " 593.4053,\n",
       " 601.90894,\n",
       " 606.2764,\n",
       " 612.0944,\n",
       " 623.055,\n",
       " 619.07104,\n",
       " 615.4873,\n",
       " 618.7329,\n",
       " 603.07446,\n",
       " 601.8417,\n",
       " 616.4246,\n",
       " 604.0266,\n",
       " 592.9908,\n",
       " 596.8841,\n",
       " 598.5163,\n",
       " 621.7478,\n",
       " 613.3183,\n",
       " 607.57086,\n",
       " 618.54675,\n",
       " 606.6945,\n",
       " 606.1156,\n",
       " 605.08875,\n",
       " 597.41833,\n",
       " 590.38055,\n",
       " 595.9502,\n",
       " 586.2013,\n",
       " 588.31866,\n",
       " 582.5831,\n",
       " 584.9117,\n",
       " 591.8782,\n",
       " 594.89105,\n",
       " 615.2722,\n",
       " 603.6983,\n",
       " 606.54315,\n",
       " 610.2988,\n",
       " 608.9189,\n",
       " 616.0764,\n",
       " 615.3021,\n",
       " ...]"
      ]
     },
     "execution_count": 21,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# For Sample state_action pair 2: State (1,2,3), Action (1,2)\n",
    "agent.states_tracked_2"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Tracking the state-action pairs for checking convergence\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### (i) Sample a few state-action pairs and plot their Q-values along episodes (Convergence Graphs)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "metadata": {},
   "outputs": [],
   "source": [
    "def convergence_graph_q_val(fig_num, state, action, states_tracked):\n",
    "    plt.figure(fig_num, figsize=(10,4))\n",
    "    plt.title(f\"Convergence of Q_values for state {state} and action {action}\", fontsize=14, fontweight='bold')\n",
    "    xaxis = np.asarray(range(0, len(states_tracked)))\n",
    "    plt.plot(xaxis,np.asarray(states_tracked))\n",
    "    plt.ylabel(\"Q_values\", fontsize=13, fontstyle='italic')\n",
    "    plt.xlabel(\"No. of Episodes  (step increment of 5)\", fontsize=13, fontstyle='italic')\n",
    "    plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "metadata": {},
   "outputs": [],
   "source": [
    "def convergence_graph_q_val_logscale(fig_num, state, action, states_tracked):\n",
    "    plt.figure(fig_num, figsize=(10,4))\n",
    "    plt.title(f\"Convergence of Q_values (log scale) for state {state} and action {action}\", fontsize=14, fontweight='bold')\n",
    "    xaxis = np.asarray(range(0, len(states_tracked)))\n",
    "    plt.yscale('log')\n",
    "    plt.plot(xaxis,np.asarray(states_tracked))\n",
    "    plt.ylabel(\"Q_values (log_scale)\", fontsize=13, fontstyle='italic')\n",
    "    plt.xlabel(\"No. of Episodes  (step increment of 5)\", fontsize=13, fontstyle='italic')\n",
    "    plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "metadata": {
    "scrolled": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      "\u001b[1mTracking Convergence for state-action pair 1: State (2,4,6), Action (2,3)\n",
      "\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 720x288 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 720x288 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      "\n",
      "\n",
      "\u001b[1mTracking Convergence for state-action pair 2: State (1,2,3), Action (1,2)\n",
      "\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 720x288 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 720x288 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "## Convergence Graph 1: For Sample state-action pair 1: State (2,4,6), Action (2,3)\n",
    "print('\\n\\033[1m'+\"Tracking Convergence for state-action pair 1: State (2,4,6), Action (2,3)\\n\")\n",
    "convergence_graph_q_val(fig_num=1, state=(2,4,6), action=(2,3), states_tracked= agent.states_tracked_1)\n",
    "convergence_graph_q_val_logscale(fig_num=2, state=(2,4,6), action=(2,3), states_tracked= agent.states_tracked_1)\n",
    "\n",
    "## Convergence Graph 2: For Sample state-action pair 2: State (1,2,3), Action (1,2)\n",
    "print('\\n\\n\\n\\033[1m'+\"Tracking Convergence for state-action pair 2: State (1,2,3), Action (1,2)\\n\")\n",
    "convergence_graph_q_val(fig_num=3, state=(1,2,3), action=(1,2), states_tracked= agent.states_tracked_2)\n",
    "convergence_graph_q_val_logscale(fig_num=4, state=(1,2,3), action=(1,2), states_tracked= agent.states_tracked_2)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Thus, from the graphs we can deduce that the sample state-action pairs `'State (2,4,6), Action (2,3)'` and `'State (1,2,3), Action (1,2)'` show convergence over episodes."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### (ii) Check whether the total rewards earned per episode are showing stability"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 864x288 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "## Rewards (y-axis) vs episodes (x-axis)\n",
    "plt.figure(5, figsize=(12,4))\n",
    "plt.title('Rewards per Episode', fontsize=15, fontweight='bold')\n",
    "xaxis = np.asarray(range(0, len(score_tracked)))\n",
    "plt.plot(xaxis,np.asarray(score_tracked))\n",
    "plt.ylabel(\"Rewards\", fontsize=13, fontstyle='italic')\n",
    "plt.xlabel(\"Number of Episodes\", fontsize=13, fontstyle='italic')\n",
    "plt.grid(True)\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 864x288 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "## Rewards (y-axis log scale) vs episodes\n",
    "plt.figure(6, figsize=(12,4))\n",
    "plt.title('Rewards per Episode (Log Scale)', fontsize=15, fontweight='bold')\n",
    "xaxis = np.asarray(range(0, len(score_tracked)))\n",
    "plt.plot(xaxis,np.asarray(score_tracked))\n",
    "plt.yscale('log')                                                 ## Log scale: On y-axis\n",
    "plt.ylabel(\"Rewards (log scale)\", fontsize=13, fontstyle='italic')\n",
    "plt.xlabel(\"Number of Episodes\", fontsize=13, fontstyle='italic')\n",
    "plt.grid(True)\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "From Rewards per Episode (Log scale) graph, we clearly see that the `rewards are quite consistent after 7500 episodes. Thus, showing the model stability.` Hence we can conclude that our model shows decent convergence post 7500 episodes."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Epsilon-Decay Check "
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "<div class=\"alert alert-block alert-info\">\n",
    "Try building a similar epsilon-decay function for your model.\n",
    "</div>"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAXcAAAD8CAYAAACMwORRAAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAALEgAACxIB0t1+/AAAADl0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uIDMuMC4yLCBodHRwOi8vbWF0cGxvdGxpYi5vcmcvOIA7rQAAH6VJREFUeJzt3Xt8XHWd//HXZya3Nte2SZsm6b2F0lQKbeSOCnIprLSuiMLidVX0p6yy6j4e+PP3YBH399uf4C6ugiCy/FxUQMBbdavcLSAUSEtb6D0ttE2vaZveaJrr5/fHnJRpmjSTdJKTmXk/H495zJnvfGfOJ2cm7znzPWfOMXdHRETSSyTsAkREJPkU7iIiaUjhLiKShhTuIiJpSOEuIpKGFO4iImlI4S4ikoYU7iIiaUjhLiKShrLCmnFpaalPnDgxrNmLiKSkJUuW7Hb3st76hRbuEydOpLa2NqzZi4ikJDPblEg/DcuIiKQhhbuISBpSuIuIpCGFu4hIGlK4i4ikoV7D3cweMLNdZvZmD/ebmf3QzOrMbIWZzU5+mSIi0heJrLn/DJh7gvuvAKYFlxuAe06+LBERORm9hru7Pw/sPUGX+cCDHrMYKDGzsckqsKvat/fyvT+vQacHFBHpWTLG3CuBLXG364O245jZDWZWa2a1DQ0N/ZrZm1v3c89fNtBwsLlfjxcRyQTJCHfrpq3b1Wp3v8/da9y9pqys11/PduvU8iIAVu842K/Hi4hkgmSEez0wLu52FbAtCc/brenlhQCs3XFgoGYhIpLykhHuC4BPBXvNnAPsd/ftSXjebo3Iz6G8KI8127XmLiLSk14PHGZmDwMfAErNrB74ZyAbwN3vBRYCVwJ1wGHgswNVbKdTyws1LCMicgK9hru7X9fL/Q58JWkVJWD62EJe3rCH1vYOsqP6HZaISFcpmYynlRfR0t7BW7vfCbsUEZEhKSXD/dRgo+rq7dqoKiLSnZQM9yllBWRFjLUadxcR6VZKhntOVoSpowtYo3AXEelWSoY7xIZm1mhYRkSkWykb7tPLi9i2/wj7m1rDLkVEZMhJ4XDv/KWqhmZERLpK3XAfGwv3NToMgYjIcVI23MuL8igels1qHYZAROQ4KRvuZhbbqKo1dxGR46RsuAPMGFvEmu0Hae/QiTtEROKldLhXVxTR1NquwxCIiHSR0uE+s7IYgJXb9odciYjI0JLS4T51dAE5WRFWbtO4u4hIvJQO9+xohOnlhby5VWvuIiLxUjrcAaorilm57QCxw8qLiAikRbgXsb+plfrGprBLEREZMlI+3N/dqKpxdxGRTikf7tPLC4lGTHvMiIjESflwz8uOMrWsQBtVRUTipHy4A1RXFmlYRkQkTnqEe0Uxuw42s+vgkbBLEREZEtIi3GdWFAHaqCoi0iktwn1GZ7hr3F1EBEiTcC/My2ZSaT5vKNxFRIA0CXeA06uKWVGvcBcRgTQK91lVJWzff4SdB7RRVUQkfcJ9XOyXqsu37Au5EhGR8KVNuFdXFBONGMvrFe4iImkT7nnZUaaXF2rcXUSEBMPdzOaa2VozqzOzm7u5f7yZPWdmr5vZCjO7Mvml9m7WuBKWb9lHh86pKiIZrtdwN7MocDdwBTADuM7MZnTp9r+AR939TOBa4MfJLjQRZ1SVcOBIG2/v0TlVRSSzJbLmfhZQ5+4b3b0FeASY36WPA0XBdDGwLXklJm7WuBIAjbuLSMZLJNwrgS1xt+uDtni3Ap8ws3pgIfAPSamuj6aOLmB4TpTlWzTuLiKZLZFwt27aug5qXwf8zN2rgCuBn5vZcc9tZjeYWa2Z1TY0NPS92l5EI8bMymKWaXdIEclwiYR7PTAu7nYVxw+7fA54FMDdXwbygNKuT+Tu97l7jbvXlJWV9a/iXpwxroRV2w/Q0tYxIM8vIpIKEgn314BpZjbJzHKIbTBd0KXPZuCDAGZ2GrFwT/6qeQJmVZXQ0tbB2h0Hw5i9iMiQ0Gu4u3sbcCPwBLCa2F4xK83sNjObF3T7BvAFM1sOPAx8xt1D2R+x85eqy7Y0hjF7EZEhISuRTu6+kNiG0vi2W+KmVwHnJ7e0/qksGUZZYS5LN+/jk+eGXY2ISDjS5heqncyMmgkjqN20N+xSRERCk3bhDjBnwgi27G1il44QKSIZKm3DHWDJJo27i0hmSstwr64oJjcronAXkYyVluGekxVhVlUJtQp3EclQaRnuALMnjGDltv0caW0PuxQRkUGXtuFeM2EEre2u47uLSEZK23CfHWxU1S6RIpKJ0jbcR+bnMLksn6UadxeRDJS24Q6xoZklmxoJ6UgIIiKhSetwnzNhBI2HW9nQoDMziUhmSetwr5k4EoDX3ta4u4hklrQO98ml+ZQV5vLKxj1hlyIiMqjSOtzNjLMnjWTxxr0adxeRjJLW4Q5wzuRR7DhwhE17DoddiojIoMmAcI+Nu7/yloZmRCRzpH24TykroLQgh8UbtVFVRDJH2od7bNx9FK9s3KNxdxHJGGkf7hAbmtm2/whb9jaFXYqIyKDIiHA/e/IoABZrl0gRyRAZEe7TRhcwMj+HxdqoKiIZIiPCvXN/91e0UVVEMkRGhDvE9nffuq+JzdrfXUQyQMaE+/lTSwF4oa4h5EpERAZexoT7lLJ8KorzeGHd7rBLEREZcBkT7mbGBdNKeWnDbto7tL+7iKS3jAl3gAumlXHgSBsr6veFXYqIyIDKqHA/f0psf/cX12toRkTSW0aF+6iCXKorinihTuEuIukto8Id4IJppby+uZFDzW1hlyIiMmAyLtzfN62M1nbX2ZlEJK0lFO5mNtfM1ppZnZnd3EOfj5nZKjNbaWYPJbfM5JkzYQS5WRFe0Li7iKSxrN46mFkUuBu4FKgHXjOzBe6+Kq7PNOBbwPnu3mhmoweq4JOVlx3lrEkjeWG9fswkIukrkTX3s4A6d9/o7i3AI8D8Ln2+ANzt7o0A7r4ruWUm1/tPKWNDwzts2atDEYhIekok3CuBLXG364O2eKcAp5jZX81ssZnN7e6JzOwGM6s1s9qGhvDWnC+eHvti8dzaIf0ZJCLSb4mEu3XT1vUnnlnANOADwHXA/WZWctyD3O9z9xp3rykrK+trrUkzuayAiaOG8+wahbuIpKdEwr0eGBd3uwrY1k2f37t7q7u/BawlFvZD1kXTR/Pyhj00tbSHXYqISNIlEu6vAdPMbJKZ5QDXAgu69PkdcBGAmZUSG6bZmMxCk+3i6aNpbuvgpQ3aa0ZE0k+v4e7ubcCNwBPAauBRd19pZreZ2byg2xPAHjNbBTwH/JO7D+kdyc+aNJL8nCjPaGhGRNJQr7tCArj7QmBhl7Zb4qYd+HpwSQm5WVEumFbKc2t24e6YdbdpQUQkNWXcL1TjXTx9NNv3H2HNjoNhlyIiklQZHe4XnRrbJVJ7zYhIusnocB9dlMd7Kot5ZvXOsEsREUmqjA53gMtmjGHp5n3sOnAk7FJERJIm48N97sxyAJ5YpbV3EUkfGR/uU0cXMLk0nydX7gi7FBGRpMn4cDczLp9Zzssb9rDvcEvY5YiIJEXGhzvA3Opy2jqcZ1ZrrxkRSQ8Kd+D0qmLGFufxZw3NiEiaULgTDM1Ul/P8ugYOt+jcqiKS+hTugcury2lu62DRWp2hSURSn8I98N6JIxiZn8N/v7E97FJERE6awj2QFY1wxcxynlm9S0MzIpLyFO5x5s2qoKm1naf0gyYRSXEK9zjvnTiSscV5/GF51xNNiYikFoV7nEjE+NDpY1m0rkE/aBKRlKZw72LerEpa250/v6l93kUkdSncu5hZWcSk0nwWaGhGRFKYwr0LM+OqWRW8vHGPDgMsIilL4d6NebMqcEdr7yKSshTu3Zg6uoDTq4p5fEk9sXN/i4ikFoV7D66ZU8WaHQdZue1A2KWIiPSZwr0H82ZVkhON8PiS+rBLERHpM4V7D4qHZ3Np9Rh+t2wrzW3tYZcjItInCvcTuGZOFfsOt+okHiKSchTuJ3DhtDLKi/J4rHZL2KWIiPSJwv0EohHjI7MrWbSugZ3a511EUojCvRcfnVNFh6O1dxFJKQr3XkwuK+D8qaN46JXNtHdon3cRSQ0K9wR84uwJbNt/hGfXaMOqiKSGhMLdzOaa2VozqzOzm0/Q76Nm5mZWk7wSw3fJjDGMKcrlF4s3hV2KiEhCeg13M4sCdwNXADOA68xsRjf9CoGvAq8ku8iwZUcjXPve8Sxa18CmPe+EXY6ISK8SWXM/C6hz943u3gI8Aszvpt93gduBtNyt5LqzxhONGA+9sjnsUkREepVIuFcC8buK1AdtR5nZmcA4d/9jEmsbUsqL87j0tDE8WruFI636xaqIDG2JhLt103Z0txEziwB3At/o9YnMbjCzWjOrbWhoSLzKIeJT506g8XArC5bpUMAiMrQlEu71wLi421VAfLoVAjOBv5jZ28A5wILuNqq6+33uXuPuNWVlZf2vOiTnThnF9PJC7n9xow4FLCJDWiLh/howzcwmmVkOcC2woPNOd9/v7qXuPtHdJwKLgXnuXjsgFYfIzPjChZNZt/MQz6/fHXY5IiI96jXc3b0NuBF4AlgNPOruK83sNjObN9AFDjVXzapgTFEu97+wMexSRER6lJVIJ3dfCCzs0nZLD30/cPJlDV05WRE+fd5Ebv/zWlZvP8BpY4vCLklE5Dj6hWo//N1Z4xmWHeU/X3wr7FJERLqlcO+HkuE5fKymit8v28qO/Wm5W7+IpDiFez99/sLJuMO9izaEXYqIyHEU7v00buRw/vbMSh5+dTO7DmrtXUSGFoX7SfjKRVNpbe/g/hc09i4iQ4vC/SRMLM1n3qwKfv7yJvYcag67HBGRoxTuJ+nGi6dypK1de86IyJCicD9JU0cXcuV7xvLgy5vY+05L2OWIiAAK96S46YPTONzSxo+fqwu7FBERQOGeFNPGFHL17CoeXLyJrfuawi5HREThniw3XXoKAD94al3IlYiIKNyTprJkGJ86ZwK/XlrP+p0Hwy5HRDKcwj2JvnzRVPJzsrjjibVhlyIiGU7hnkQj83P44vsn8+Sqnby8YU/Y5YhIBlO4J9nnL5xMZckwvvOHlbS1d4RdjohkKIV7kuVlR/n235zGmh0Hefi1Lb0/QERkACjcB8AVM8s5Z/JI/u3Jtew7rB82icjgU7gPADPjn6+q5kBTK3dq10gRCYHCfYCcNraI68+ewM8Xb+KN+v1hlyMiGUbhPoC+efmplBbkcvNvVmjjqogMKoX7ACoels135lWzctsBHTVSRAaVwn2AzZ1ZzqUzxnDn0+vYvOdw2OWISIZQuA8wM+O782eSFYnwP3/7Bu4edkkikgEU7oOgvDiPm6+Yzot1u/nFK5vDLkdEMoDCfZBcf/Z43ndKGf/7v1exoeFQ2OWISJpTuA8SM+OOj55OXnaUr/9qGa3ae0ZEBpDCfRCNKcrj//zte1hev5+7ntVZm0Rk4CjcB9mV7xnLR86s5EfPrmfxRh05UkQGhsI9BN+ZX82EUfl89eHXaTjYHHY5IpKGFO4hKMzL5sfXz2Z/Uytfe+R12ju0e6SIJJfCPSSnjS3iux+eyUsb9vCDp3VwMRFJroTC3czmmtlaM6szs5u7uf/rZrbKzFaY2TNmNiH5paafj9WM45o5Vfzo2Tr+/OaOsMsRkTTSa7ibWRS4G7gCmAFcZ2YzunR7Hahx99OBx4Hbk11ouvruh2cya1wJ//irZby5VUePFJHkSGTN/Sygzt03unsL8AgwP76Duz/n7p0HTlkMVCW3zPSVlx3lp5+cQ8nwbL7wYC27Dh4JuyQRSQOJhHslEH++uPqgrSefA/7U3R1mdoOZ1ZpZbUNDQ+JVprnRRXn89FM17Dvcyg0PLuFIa3vYJYlIiksk3K2btm537zCzTwA1wB3d3e/u97l7jbvXlJWVJV5lBphZWcydHz+D5fX7uPGh13X8dxE5KYmEez0wLu52FbCtayczuwT4NjDP3bXzdj/MnVnOd+ZV8/TqnXzrNzqCpIj0X1YCfV4DppnZJGArcC3wd/EdzOxM4CfAXHfflfQqM8inzp3InkMt/Mcz6xlZkMO3rjgt7JJEJAX1Gu7u3mZmNwJPAFHgAXdfaWa3AbXuvoDYMEwB8JiZAWx293kDWHdau+mSaex9p4WfLNpIYW4WN148LeySRCTFJLLmjrsvBBZ2abslbvqSJNeV0cyMW+dVc6i5je8/uY4Oh69+UAEvIolLKNxl8EUjxvevmYUZ/PtT6+hw56ZLTgm7LBFJEQr3ISwaMe746CwiZvzg6fW0tTvfuOwUgqEvEZEeKdyHuGjEuP3q08mKGHc9V8fuQ838y4dnkhXVYYFEpGcK9xQQiRj/+pH3UFqQezTgf3TdbIblRMMuTUSGKK3+pQgz45uXn8p351fzzJpdXH//Yh0LXkR6pHBPMZ88dyL3XD+bVdsPMP+uF3WwMRHplsI9Bc2dOZbHv3QeAFff8xK/X7Y15IpEZKhRuKeomZXFLPiHC5hVVcLXHlnGbX9YRXObDjgmIjEK9xRWWpDLLz5/Np85byIP/PUtrr7nJd7a/U7YZYnIEKBwT3E5WRFunVfNfZ+cQ31jEx/64Qv8Zmm9DjomkuEU7mnisupyFn71Qqorivn6o8v50i+W6MQfIhlM4Z5GKkqG8dAXzuZbV0znubUNXPrvz/PrJVqLF8lECvc0kxWN8MX3T+FPX7uQaaML+MZjy/n0/3uNjQ2Hwi5NRAaRwj1NTSkr4NEvnsutV83g9U2NXP6D5/nXP63mUHNb2KWJyCBQuKexSMT4zPmTePabH+DDZ1Tyk0Ubufj7f+Gx2i20d2ioRiSdKdwzQFlhLndcM4vffeV8xpYM458eX8Fldy5i4Rvb6VDIi6QlhXsGOWNcCb/78nnc+4k5RMz48i+XctVdL/L0qp0KeZE0Y2HtSVFTU+O1tbWhzFugvcP5/bKt3Pn0OrbsbeKUMQXc8L4pzJtVQU6WPvNFhiozW+LuNb32U7hnttb2Dv64Yhs/WbSRNTsOMrY4j8+eP5Fr5oxjRH5O2OWJSBcKd+kTd+cvaxu4Z9EGXn1rLzlZET50+liuP3sCs8eX6OxPIkNEouGuk3UIEDte/EXTR3PR9NGs2naAh17dxG+XbuU3S7dy2tgirp5dybxZFYwuygu7VBFJgNbcpUeHmttYsGwbD7+6mTe27idicP7UUuafUcnl1WMozMsOu0SRjKNhGUmqul0H+f2ybfxu2Va27G0iJyvC+VNGcemMci6ZMZrRhVqjFxkMCncZEO7O0s2N/HHFdp5atZP6xiYgtpvlpTPGcOG0UqoriolGNEYvMhAU7jLg3J01Ow7y1KqdPLVqJ28Ep/wrHpbNeVNGcd7UUi6YWsrEUcO1QVYkSRTuMuh2HTzCyxv28OL63fy1bjfb9scOOVxakMucCSXMHj+CORNGMLOymLzsaMjViqQmhbuEyt15e89h/lq3m6WbGlmyuZFNew4DkB01ZlQUM7OiiBkVRcwYW8T08iKG5SjwRXqjcJchZ/ehZpZuamTp5n28vrmR1dsPcOBI7CiVEYNJpfmcNraIU8YUMqWsgMll+Uwqzddavkgc7ecuQ05pQS6XVZdzWXU5EFu7r29sYtX2A6zadoBV2w/w+uZ9/HHF9qOPMYOqEcOYXFrAlLICxo8cRtWI4VQF1wW5eguLdEf/GRIaM2PcyOGMGzmcy4PABzjc0sZbu99hQ8M7bGw4dPT61bf20tTafsxzlAzPpmrEMKpKhlNRMowxRbmMLspldGEeowtj10XDsrRBVzJOQuFuZnOB/wCiwP3u/n+73J8LPAjMAfYAH3f3t5NbqmSK4TlZVFcUU11RfEy7u7PnnRbqG5uobzx89HprYxMbGg7x/PoGDre0H/d8uVkRygpzGV2Yy6iCXEYMz2ZEfg4jhucwYng2JcNj0yPzY9Mlw7LJiurgaZLaeg13M4sCdwOXAvXAa2a2wN1XxXX7HNDo7lPN7Frge8DHB6JgyVxmRmlBLqUFuZwxrqTbPoea29h14Ai7Djaz88ARGg42s+tg89G2LXsPs3xLC/sOt9LS3tHjvPJzohTkZVGQm0VBXjaFuZ3TsevCvHdv5+dkkZcdZVhOlGHZUfKyI8F19Gh7XlZEHxgyqBJZcz8LqHP3jQBm9ggwH4gP9/nArcH048BdZmauMzPLICvIzaKgrIDJZQUn7OfuHG5pp/FwLOgbD7fQeLiVfYdb2PtOC4eOtHGouY2DzW1HpxsONsfajrRyqLmNvh4CPztqsbAPQj83K0J2NEJ2VoScqJEdjZATtOVEI2R3bcs6ti0nGiEasXcvZsfe7qktaM+KGhEzsiIRIhGIRoysSKwtGlxD7IxeBkTMiBhgsel32yxoC9qDa+IeY3HXMjgSCfdKYEvc7Xrg7J76uHubme0HRgG7k1GkSLKZGfm5WeTnZlE1ou+Pd3eaWtuPBv+R1g6aWts5ElyaWttpaum8Hbuvs625rZ3DLe20tnfQ0ua0tHfQ2tZBS1sH7zS30dLuwX0dtLZ3xE3H+qb6KRKty4fDMR8McW12tL8dfRy8237MfXHPHd+r62PiP1usxz7HfgAd85ijfU/82C5lHNfnax+cxlWzKhhIiYR7dx+1Xd9difTBzG4AbgAYP358ArMWGZrMjOE5WQzPyWL0IM+7vSMI//YOOjqctg4/et3eefG46RO1BZe2Dqejy/0OdLiDx647b3c4EFx7cN0RfEnvONoe19YRtHHsY+Ofs3MeXR/bqXMQwI9pC66D1ndvH3s/Xe7vy2M9fo7H9fEeHnPs/cfUHUwUDxv4g+4lEu71wLi421XAth761JtZFlAM7O36RO5+H3AfxPZz70/BIpkuNrQS1f7/ckKJbOF5DZhmZpPMLAe4FljQpc8C4NPB9EeBZzXeLiISnl7X3IMx9BuBJ4jtCvmAu680s9uAWndfAPwn8HMzqyO2xn7tQBYtIiInltB+7u6+EFjYpe2WuOkjwDXJLU1ERPpLO96KiKQhhbuISBpSuIuIpCGFu4hIGlK4i4ikodBO1mFmDcCmfj68lKF5aAPV1Teqq++Gam2qq29Opq4J7l7WW6fQwv1kmFltImciGWyqq29UV98N1dpUV98MRl0alhERSUMKdxGRNJSq4X5f2AX0QHX1jerqu6Fam+rqmwGvKyXH3EVE5MRSdc1dREROIOXC3czmmtlaM6szs5sHeF7jzOw5M1ttZivN7GtB+61mttXMlgWXK+Me862gtrVmdvlA1m1mb5vZG0ENtUHbSDN7yszWB9cjgnYzsx8G819hZrPjnufTQf/1ZvbpnuaXYE2nxi2XZWZ2wMxuCmOZmdkDZrbLzN6Ma0va8jGzOcHyrwsem9A55Hqo6w4zWxPM+7dmVhK0TzSzprjldm9v8+/pb+xnXUl73Sx22PBXgrp+ZbFDiPe3rl/F1fS2mS0LYXn1lA+hv8eA2NlCUuVC7JDDG4DJQA6wHJgxgPMbC8wOpguBdcAMYueL/WY3/WcENeUCk4JaowNVN/A2UNql7Xbg5mD6ZuB7wfSVwJ+InTXrHOCVoH0ksDG4HhFMj0ji67UDmBDGMgPeB8wG3hyI5QO8CpwbPOZPwBUnUddlQFYw/b24uibG9+vyPN3Ov6e/sZ91Je11Ax4Frg2m7wX+R3/r6nL/vwG3hLC8esqH0N9j7p5ya+5HT9bt7i1A58m6B4S7b3f3pcH0QWA1sfPF9mQ+8Ii7N7v7W0BdUPNg1j0f+K9g+r+AD8e1P+gxi4ESMxsLXA485e573b0ReAqYm6RaPghscPcT/VhtwJaZuz/P8WcES8ryCe4rcveXPfZf+GDcc/W5Lnd/0t3bgpuLiZ3xrEe9zL+nv7HPdZ1An163YI3zYuDxZNYVPO/HgIdP9BwDtLx6yofQ32OQesMy3Z2s+0RhmzRmNhE4E3glaLox+Gr1QNzXuJ7qG6i6HXjSzJZY7Py0AGPcfTvE3nxw9BSfg10bxE7aEv9PNxSWWbKWT2Uwnez6AP6e2Fpap0lm9rqZLTKzC+Pq7Wn+Pf2N/ZWM120UsC/uAyxZy+tCYKe7r49rG/Tl1SUfhsR7LNXCPaETcSd9pmYFwK+Bm9z9AHAPMAU4A9hO7GvhieobqLrPd/fZwBXAV8zsfSfoO6i1BeOp84DHgqahssx60tc6Bmq5fRtoA34ZNG0Hxrv7mcDXgYfMrGig5t+NZL1uA1XvdRy7AjHoy6ubfOixaw81DMgyS7VwT+Rk3UllZtnEXrhfuvtvANx9p7u3u3sH8FNiX0VPVN+A1O3u24LrXcBvgzp2Bl/nOr+K7gqjNmIfOEvdfWdQ45BYZiRv+dRz7NDJSdcXbEj7EHB98DWcYNhjTzC9hNh49im9zL+nv7HPkvi67SY2DJHVpb3fguf6CPCruHoHdXl1lw8neL7BfY8lOjg/FC7ETgu4kdgGnM6NNdUDOD8jNs71gy7tY+Om/5HY2CNANcduZNpIbANT0usG8oHCuOmXiI2V38GxG3NuD6b/hmM35rzq727MeYvYhpwRwfTIJCy7R4DPhr3M6LKBLZnLh9jJ48/h3Y1dV55EXXOBVUBZl35lQDSYngxs7W3+Pf2N/awraa8bsW9x8RtUv9zfuuKW2aKwlhc958PQeI+d7D/xYF+IbXFeR+wT+dsDPK8LiH0NWgEsCy5XAj8H3gjaF3T5B/h2UNta4rZsJ7vu4I27PLis7HxOYmObzwDrg+vON4kBdwfzfwOoiXuuvye2QayOuEA+idqGA3uA4ri2QV9mxL6ubwdaia0FfS6ZyweoAd4MHnMXwY8C+1lXHbFx18732b1B36uD13c5sBS4qrf59/Q39rOupL1uwXv21eBvfQzI7W9dQfvPgC916TuYy6unfAj9Pebu+oWqiEg6SrUxdxERSYDCXUQkDSncRUTSkMJdRCQNKdxFRNKQwl1EJA0p3EVE0pDCXUQkDf1/fVqc+K0jucAAAAAASUVORK5CYII=\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "time = np.arange(0,20000)              \n",
    "epsilon = []\n",
    "for i in range(0,20000):\n",
    "    epsilon.append(0.0000001 + (1 - 0.0000001) * np.exp(-0.00045*i))\n",
    "\n",
    "plt.plot(time, epsilon)\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.7.9"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}