Cycle_GAN_real_art.ipynb

{
  "cells": [
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "view-in-github",
        "colab_type": "text"
      },
      "source": [
        "<a href=\"https://colab.research.google.com/github/rashmi-2000/AI-Generated-Art/blob/main/Cycle_GAN_real_art.ipynb\" target=\"_parent\"><img src=\"https://colab.research.google.com/assets/colab-badge.svg\" alt=\"Open In Colab\"/></a>"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "eW8Pjw1-cycJ"
      },
      "source": [
        "Cycle GAN"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "Z9J0Z1iXJ6Xr",
        "outputId": "6be1073f-7250-439f-9ff2-80071a965b59"
      },
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Looking in indexes: https://pypi.org/simple, https://us-python.pkg.dev/colab-wheels/public/simple/\n",
            "Collecting git+https://www.github.com/keras-team/keras-contrib.git\n",
            "  Cloning https://www.github.com/keras-team/keras-contrib.git to /tmp/pip-req-build-obbojrod\n",
            "  Running command git clone -q https://www.github.com/keras-team/keras-contrib.git /tmp/pip-req-build-obbojrod\n",
            "Requirement already satisfied: keras in /usr/local/lib/python3.7/dist-packages (from keras-contrib==2.0.8) (2.8.0)\n",
            "Building wheels for collected packages: keras-contrib\n",
            "  Building wheel for keras-contrib (setup.py) ... \u001b[?25l\u001b[?25hdone\n",
            "  Created wheel for keras-contrib: filename=keras_contrib-2.0.8-py3-none-any.whl size=101077 sha256=9262c6065372e2be5dfc362eb6e38b9a0338e5aa7c04bf9be4120061a3f04e6b\n",
            "  Stored in directory: /tmp/pip-ephem-wheel-cache-0xeiuikj/wheels/bb/1f/f2/b57495012683b6b20bbae94a3915ec79753111452d79886abc\n",
            "Successfully built keras-contrib\n",
            "Installing collected packages: keras-contrib\n",
            "Successfully installed keras-contrib-2.0.8\n"
          ]
        }
      ],
      "source": [
        "!pip install git+https://www.github.com/keras-team/keras-contrib.git #Keras for instance_normalization"
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "import cv2"
      ],
      "metadata": {
        "id": "axHr3ESZF5eA"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "source": [
        "Unsharp Mask Sharpening Technique\n"
      ],
      "metadata": {
        "id": "ZH-8QI5Xxss9"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "#Unsharp Mask\n",
        "def unsharp(image):\n",
        "  gaussian_3 = cv2.GaussianBlur(image, (0, 0), 2.0)\n",
        "  unsharp_image = cv2.addWeighted(image, 2.0, gaussian_3, -1.0, 0)\n",
        "  return unsharp_image"
      ],
      "metadata": {
        "id": "egDHtnyEF8Ku"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "source": [
        "Laplacian Filter Sharpening Technique"
      ],
      "metadata": {
        "id": "wkusyUJdyGMl"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "#Unsharp Mask\n",
        "import numpy as np\n",
        "def lapfil(path):\n",
        "  kernel = np.array([[0, -1, 0],\n",
        "                    [-1, 5,-1],\n",
        "                    [0, -1, 0]])\n",
        "  image_sharp = cv2.filter2D(src=path, ddepth=-1, kernel=kernel)\n",
        "  return image_sharp"
      ],
      "metadata": {
        "id": "_OVNYoYYF-ik"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "source": [
        "Highboost Filter"
      ],
      "metadata": {
        "id": "6jVCnmJDFrEh"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "#Highboost Filter\n",
        "import numpy as np\n",
        "from matplotlib import pyplot as plt\n",
        "import imageio\n",
        "def highboost(path):\n",
        "    kernel = np.array([[-1, -1, -1],\n",
        "                   [-1,  8, -1],\n",
        "                   [-1, -1, -1]])\n",
        "    image_hpf = cv2.filter2D(path, -1, kernel)\n",
        "    return image_hpf"
      ],
      "metadata": {
        "id": "LsKqFCZ4GC0g"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "7xPtRZRacVNV",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "99334b46-905a-4022-b12e-f13c026284a0"
      },
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Mounted at /content/drive\n"
          ]
        }
      ],
      "source": [
        "#accessing dataset from Google Drive\n",
        "from google.colab import drive\n",
        "drive.mount('/content/drive')"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "Ecz7HMqOJ_XU"
      },
      "outputs": [],
      "source": [
        "from random import random\n",
        "from numpy import load\n",
        "from numpy import zeros\n",
        "from numpy import ones\n",
        "from numpy import asarray\n",
        "from numpy.random import randint\n",
        "from tensorflow.keras.optimizers import Adam\n",
        "from keras.initializers import RandomNormal\n",
        "from keras.models import Model\n",
        "from keras.models import Input\n",
        "from keras.layers import Conv2D\n",
        "from keras.layers import Conv2DTranspose\n",
        "from keras.layers import LeakyReLU\n",
        "from keras.layers import Activation\n",
        "from keras.layers import Concatenate\n",
        "from keras_contrib.layers.normalization.instancenormalization import InstanceNormalization  #instance_normalization\n",
        "from matplotlib import pyplot\n",
        "\n",
        "#DISCRIMINATOR (classifying real/fake using instance normalization)\n",
        "def define_discriminator(image_shape):\n",
        "\t# weight initialization\n",
        "\tinit = RandomNormal(stddev=0.02)\n",
        " \n",
        "\t# source image input\n",
        "\tin_image = Input(shape=image_shape)\n",
        " \n",
        "\t# C64: 4x4 kernel Stride 2x2 and 64 filters\n",
        "\td = Conv2D(64, (4,4), strides=(2,2), padding='same', kernel_initializer=init)(in_image)\n",
        "\td = LeakyReLU(alpha=0.2)(d)\n",
        " \n",
        "\t# C128: 4x4 kernel Stride 2x2 and 128 filters\n",
        "\td = Conv2D(128, (4,4), strides=(2,2), padding='same', kernel_initializer=init)(d)\n",
        "\td = InstanceNormalization(axis=-1)(d)\n",
        "\td = LeakyReLU(alpha=0.2)(d)#LeakyReLU\n",
        "\n",
        "\t# C256: 4x4 kernel Stride 2x2 and 256 filters\n",
        "\td = Conv2D(256, (4,4), strides=(2,2), padding='same', kernel_initializer=init)(d)\n",
        "\td = InstanceNormalization(axis=-1)(d)\n",
        "\td = LeakyReLU(alpha=0.2)(d)\n",
        " \n",
        "\t# C512: 4x4 kernel Stride 2x2 and 512 filters\n",
        "\td = Conv2D(512, (4,4), strides=(2,2), padding='same', kernel_initializer=init)(d)\n",
        "\td = InstanceNormalization(axis=-1)(d)\n",
        "\td = LeakyReLU(alpha=0.2)(d)\n",
        " \n",
        "\t# second last output layer : 4x4 kernel but Stride 1x1\n",
        "\td = Conv2D(512, (4,4), padding='same', kernel_initializer=init)(d)\n",
        "\td = InstanceNormalization(axis=-1)(d)\n",
        "\td = LeakyReLU(alpha=0.2)(d)\n",
        " \n",
        "\t# patch output\n",
        "\tpatch_out = Conv2D(1, (4,4), padding='same', kernel_initializer=init)(d)\n",
        " \n",
        "\t# define model\n",
        "\tmodel = Model(in_image, patch_out)\n",
        "\tmodel.compile(loss='mse', optimizer=Adam(lr=0.0002, beta_1=0.5), loss_weights=[0.5])\n",
        "\treturn model\n",
        "\n",
        "#resnet block is needed as a part of generator\n",
        "def resnet_block(n_filters, input_layer):\n",
        "\t# weight initialization\n",
        "\tinit = RandomNormal(stddev=0.02)\n",
        " \n",
        "\t# first convolutional layer\n",
        "\tg = Conv2D(n_filters, (3,3), padding='same', kernel_initializer=init)(input_layer)\n",
        "\tg = InstanceNormalization(axis=-1)(g)\n",
        "\tg = Activation('relu')(g)\n",
        " \n",
        "\t# second convolutional layer\n",
        "\tg = Conv2D(n_filters, (3,3), padding='same', kernel_initializer=init)(g)\n",
        "\tg = InstanceNormalization(axis=-1)(g)\n",
        " \n",
        "\t# concatenate merge channel-wise with input layer\n",
        "\tg = Concatenate()([g, input_layer]) #concatenate output with input\n",
        "\treturn g\n",
        "\n",
        "#GENERATOR\n",
        "def define_generator(image_shape, n_resnet=9):\n",
        "\t# weight initialization\n",
        "\tinit = RandomNormal(stddev=0.02)\n",
        " \n",
        "\t# image input\n",
        "\tin_image = Input(shape=image_shape)\n",
        " \n",
        "\t# c7s1-64\n",
        "\tg = Conv2D(64, (7,7), padding='same', kernel_initializer=init)(in_image)\n",
        "\tg = InstanceNormalization(axis=-1)(g)\n",
        "\tg = Activation('relu')(g)\n",
        " \n",
        "\t# d128\n",
        "\tg = Conv2D(128, (3,3), strides=(2,2), padding='same', kernel_initializer=init)(g)\n",
        "\tg = InstanceNormalization(axis=-1)(g)\n",
        "\tg = Activation('relu')(g)\n",
        " \n",
        "\t# d256\n",
        "\tg = Conv2D(256, (3,3), strides=(2,2), padding='same', kernel_initializer=init)(g)\n",
        "\tg = InstanceNormalization(axis=-1)(g)\n",
        "\tg = Activation('relu')(g)\n",
        " \n",
        "\t# R256\n",
        "\tfor _ in range(n_resnet):\n",
        "\t\tg = resnet_block(256, g) #9 resnet blocks\n",
        "\t\n",
        "\t# u128\n",
        "\tg = Conv2DTranspose(128, (3,3), strides=(2,2), padding='same', kernel_initializer=init)(g)\n",
        "\tg = InstanceNormalization(axis=-1)(g)\n",
        "\tg = Activation('relu')(g)\n",
        " \n",
        "\t# u64\n",
        "\tg = Conv2DTranspose(64, (3,3), strides=(2,2), padding='same', kernel_initializer=init)(g)\n",
        "\tg = InstanceNormalization(axis=-1)(g)\n",
        "\tg = Activation('relu')(g)\n",
        " \n",
        "\t# c7s1-3\n",
        "\tg = Conv2D(3, (7,7), padding='same', kernel_initializer=init)(g)\n",
        "\tg = InstanceNormalization(axis=-1)(g)\n",
        "\tout_image = Activation('tanh')(g) #tanh -> -1 to 1\n",
        " \n",
        "\t# define model\n",
        "\tmodel = Model(in_image, out_image)\n",
        "\treturn model\n",
        " \n",
        "def define_composite_model(g_model_1, d_model, g_model_2, image_shape):\n",
        "\n",
        "#All other models should be constant or non-trainable when  1 model is training\n",
        "\tg_model_1.trainable = True\n",
        "\td_model.trainable = False\n",
        "\tg_model_2.trainable = False\n",
        "  \n",
        "\t#4 losses\n",
        "\n",
        "\t#adversarial loss\n",
        "\tinput_gen = Input(shape=image_shape)\n",
        "\tgen1_out = g_model_1(input_gen)\n",
        "\toutput_d = d_model(gen1_out)\n",
        " \n",
        "\t#identitiy loss\n",
        "\tinput_id = Input(shape=image_shape)\n",
        "\toutput_id = g_model_1(input_id)\n",
        " \n",
        "\t#Forward cycle consistency loss\n",
        "\toutput_f = g_model_2(gen1_out)\n",
        " \n",
        "\t#Backward cycle consistency loss\n",
        "\tgen2_out = g_model_2(input_id)\n",
        "\toutput_b = g_model_1(gen2_out)\n",
        "    \n",
        "\t#define model graph\n",
        "\tmodel = Model([input_gen, input_id], [output_d, output_id, output_f, output_b])\n",
        "\t\n",
        "#defining the optimizer-> Adam solver\n",
        "\topt = Adam(lr=0.0002, beta_1=0.5)\n",
        " \n",
        " #compile mode weighting of L1(MAE) and L2(MSE) loss [minimizing the losses]\n",
        "\tmodel.compile(loss=['mse', 'mae', 'mae', 'mae'], \n",
        "               loss_weights=[1, 5, 10, 10], optimizer=opt)\n",
        "\treturn model\n",
        "\n",
        "#Helper Functions\n",
        "#1\n",
        "#load and prepare training images\n",
        "def load_real_samples(filename):\n",
        "\n",
        "\t#load dataset\n",
        "\tdata = load(filename)\n",
        " \n",
        " #unpack arrays\n",
        "\tX1, X2 = data['arr_0'], data['arr_1']\n",
        "\n",
        "\t# scale from [0,255] to [-1,1] (tanh)\n",
        "\tX1 = (X1 - 127.5) / 127.5\n",
        "\tX2 = (X2 - 127.5) / 127.5\n",
        "\treturn [X1, X2]\n",
        "\n",
        "#2\n",
        "#select a batch of random samples, returns  images and target. (real images- (y)- 1)\n",
        "def generate_real_samples(dataset, n_samples, patch_shape):\n",
        "\n",
        "\t#choose random instance\n",
        "\tix = randint(0, dataset.shape[0], n_samples)\n",
        " \n",
        "\t#retrieve selected image\n",
        "\tX = dataset[ix]\n",
        "\n",
        "\t#generate real class labels (1)\n",
        "\ty = ones((n_samples, patch_shape, patch_shape, 1))\n",
        "\treturn X, y\n",
        "\n",
        "#3\n",
        "#select a batch of random samples, returns  images and target. (fake images- (y)- 0)\n",
        "def generate_fake_samples(g_model, dataset, patch_shape):\n",
        "\t\n",
        "\t#generate fake images\n",
        "\tX = g_model.predict(dataset)\n",
        "\t\n",
        "\t#fake class labels (0)\n",
        "\ty = zeros((len(X), patch_shape, patch_shape, 1))\n",
        "\treturn X, y\n",
        "\n",
        "#4\n",
        "def save_models(step, g_model_AtoB, g_model_BtoA):\n",
        "\n",
        "\t#saves first generator model\n",
        "\tfilename1 = '/content/drive/MyDrive/Project/CycleGAN/TrainedModels/C_Highboost_Real_Model/g_model_AtoB_%06d.h5' % (step+1)\n",
        "\tg_model_AtoB.save(filename1)\n",
        " \n",
        "\t#saves second generator model\n",
        "\tfilename2 = '/content/drive/MyDrive/Project/CycleGAN/TrainedModels/C_Highboost_Real_Model/g_model_BtoA_%06d.h5' % (step+1)\n",
        "\tg_model_BtoA.save(filename2)\n",
        "\tprint('>Saved: %s and %s' % (filename1, filename2))\n",
        "\n",
        "#5\n",
        "#periodically generate images using the save model and plot imput and output images\n",
        "def summarize_performance(step, g_model, trainX, name, n_samples=5):\n",
        "\n",
        "\t#select a sample of i/p img\n",
        "\tX_in, _ = generate_real_samples(trainX, n_samples, 0)\n",
        " \n",
        "\t#generate translated image\n",
        "\tX_out, _ = generate_fake_samples(g_model, X_in, 0)\n",
        " \n",
        "\t#scale all pixels from (-1, 1) to (0,1)\n",
        "\tX_in = (X_in + 1) / 2.0\n",
        "\tX_out = (X_out + 1) / 2.0\n",
        "\t\n",
        "\t#plot real image\n",
        "\tfor i in range(n_samples):\n",
        "\t\tpyplot.subplot(2, n_samples, 1 + i)\n",
        "\t\tpyplot.axis('off')\n",
        "\t\tpyplot.imshow(X_in[i])\n",
        "\t\n",
        "\t#plot translated image\n",
        "\tfor i in range(n_samples):\n",
        "\t\tpyplot.subplot(2, n_samples, 1 + n_samples + i)\n",
        "\t\tpyplot.axis('off')\n",
        "\t\tpyplot.imshow(X_out[i])\n",
        "\t\n",
        "\t#save plot\n",
        "\tfilename1 = '%s_generated_plot_%06d.png' % (name, (step+1))\n",
        "\tpyplot.savefig(filename1)\n",
        "\tpyplot.close()\n",
        " \n",
        " #5\n",
        "#pool of 50 fake images to reduce model oscillation\n",
        "def update_image_pool(pool, images, max_size=50):\n",
        "\tselected = list()\n",
        "\tfor image in images:\n",
        "\t\tif len(pool) < max_size:\n",
        "\t\t\t\n",
        "\t\t\t# stack the pool\n",
        "\t\t\tpool.append(image)\n",
        "\t\t\tselected.append(image)\n",
        "\t\telif random() < 0.5:\n",
        "\t\t\tselected.append(image)\n",
        "\t\telse:\n",
        "\t\t\tix = randint(0, len(pool))\n",
        "\t\t\tselected.append(pool[ix])\n",
        "\t\t\tpool[ix] = image\n",
        "\treturn asarray(selected)\n",
        "\n",
        "#TRAINING THE MODEL\n",
        "def train(d_model_A, d_model_B, g_model_AtoB, g_model_BtoA, c_model_AtoB, c_model_BtoA, dataset, epochs=5):\n",
        "\n",
        "\t# define properties of the training run\n",
        "\tn_epochs, n_batch, = epochs, 1  \n",
        "\n",
        "\t# determine the output square shape of the discriminator\n",
        "\tn_patch = d_model_A.output_shape[1]\n",
        "\n",
        "\t# unpack dataset\n",
        "\ttrainA, trainB = dataset\n",
        "\n",
        "\t# prepare image pool for fake images\n",
        "\tpoolA, poolB = list(), list()\n",
        " \n",
        "\t# calculate the number of batches per training epoch\n",
        "\tbat_per_epo = int(len(trainA) / n_batch)\n",
        " \n",
        "\t# calculate the number of training iterations\n",
        "\tn_steps = bat_per_epo * n_epochs\n",
        "    \n",
        "# manually enumerate epochs\n",
        "\tfor i in range(n_steps):\n",
        "\n",
        "\t\t# select a batch of real samples from each domain (A and B)\n",
        "\t\tX_realA, y_realA = generate_real_samples(trainA, n_batch, n_patch)\n",
        "\t\tX_realB, y_realB = generate_real_samples(trainB, n_batch, n_patch)\n",
        "\t\n",
        "\t\t# generate a batch of fake samples using both B to A and A to B generators.\n",
        "\t\tX_fakeA, y_fakeA = generate_fake_samples(g_model_BtoA, X_realB, n_patch)\n",
        "\t\tX_fakeB, y_fakeB = generate_fake_samples(g_model_AtoB, X_realA, n_patch)\n",
        "\t\n",
        "\t\t# update fake images in the pool. Remember that the paper suggstes a buffer of 50 images\n",
        "\t\tX_fakeA = update_image_pool(poolA, X_fakeA)\n",
        "\t\tX_fakeB = update_image_pool(poolB, X_fakeB)\n",
        "        \n",
        "\t\t# update generator B->A via the composite model\n",
        "\t\tg_loss2, _, _, _, _  = c_model_BtoA.train_on_batch([X_realB, X_realA], [y_realA, X_realA, X_realB, X_realA])\n",
        "\t\n",
        "\t\t# update discriminator for A -> [real/fake]\n",
        "\t\tdA_loss1 = d_model_A.train_on_batch(X_realA, y_realA)\n",
        "\t\tdA_loss2 = d_model_A.train_on_batch(X_fakeA, y_fakeA)\n",
        "\t\t\n",
        "    # update generator A->B via the composite model \n",
        "\t\tg_loss1, _, _, _, _ = c_model_AtoB.train_on_batch([X_realA, X_realB], [y_realB, X_realB, X_realA, X_realB])\n",
        "\t\n",
        "\t\t# update discriminator for B -> [real/fake]\n",
        "\t\tdB_loss1 = d_model_B.train_on_batch(X_realB, y_realB)\n",
        "\t\tdB_loss2 = d_model_B.train_on_batch(X_fakeB, y_fakeB)\n",
        "\t\t\n",
        "\t\tprint('Iteration>%d, dA[%.3f,%.3f] dB[%.3f,%.3f] g[%.3f,%.3f]' % (i+1, dA_loss1,dA_loss2, dB_loss1,dB_loss2, g_loss1,g_loss2))\n",
        "\t\n",
        "\t\t# evaluate the model performance periodically\n",
        "     #If batch size (total images)=100, performance will be summarized after every 75th iteration.\n",
        "\t\tif (i+1) % (bat_per_epo * 1) == 0:\n",
        "\n",
        "\t\t\t# plot A->B translation\n",
        "\t\t\tsummarize_performance(i, g_model_AtoB, trainA, 'AtoB')\n",
        "\t\t\t# plot B->A translation\n",
        "\n",
        "\t\t\tsummarize_performance(i, g_model_BtoA, trainB, 'BtoA')\n",
        "\t\tif (i+1) % (bat_per_epo * 5) == 0:\n",
        "\n",
        "\t\t\t# save the models\n",
        "\t\t\tsave_models(i, g_model_AtoB, g_model_BtoA)"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "Q6CoKPoPc6S4"
      },
      "source": [
        "Dataset Preprocessing"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "nfXs6hVwR-d7"
      },
      "outputs": [],
      "source": [
        "import os\n",
        "from os import listdir\n",
        "import re\n",
        "import cv2"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "z31FaknISBXL"
      },
      "outputs": [],
      "source": [
        "fd1 = '/content/drive/MyDrive/WarpedDreams/CycleGAN/Dataset/realistic_car'\n",
        "fd2='/content/drive/MyDrive/WarpedDreams/CycleGAN/Dataset/sketch_car'"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "zjJNi5L-SF0m",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "84da6bd0-8a53-444f-8ef2-7ba9641fcd73"
      },
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "{'n02958343_10092-1.jpg': ['n02958343_10092-1.png'],\n",
              " 'n02958343_10092-2.jpg': ['n02958343_10092-2.png'],\n",
              " 'n02958343_10092-3.jpg': ['n02958343_10092-3.png'],\n",
              " 'n02958343_10092-4.jpg': ['n02958343_10092-4.png'],\n",
              " 'n02958343_10092-5.jpg': ['n02958343_10092-5.png'],\n",
              " 'n02958343_10187-1.jpg': ['n02958343_10187-1.png'],\n",
              " 'n02958343_10187-2.jpg': ['n02958343_10187-2.png'],\n",
              " 'n02958343_10187-3.jpg': ['n02958343_10187-3.png'],\n",
              " 'n02958343_10187-4.jpg': ['n02958343_10187-4.png'],\n",
              " 'n02958343_10187-5.jpg': ['n02958343_10187-5.png'],\n",
              " 'n02958343_1104-1.jpg': ['n02958343_1104-1.png'],\n",
              " 'n02958343_1104-2.jpg': ['n02958343_1104-2.png'],\n",
              " 'n02958343_1104-3.jpg': ['n02958343_1104-3.png'],\n",
              " 'n02958343_1104-4.jpg': ['n02958343_1104-4.png'],\n",
              " 'n02958343_1104-5.jpg': ['n02958343_1104-5.png'],\n",
              " 'n02958343_11246-1.jpg': ['n02958343_11246-1.png'],\n",
              " 'n02958343_11246-2.jpg': ['n02958343_11246-2.png'],\n",
              " 'n02958343_11246-3.jpg': ['n02958343_11246-3.png'],\n",
              " 'n02958343_11246-4.jpg': ['n02958343_11246-4.png'],\n",
              " 'n02958343_11246-5.jpg': ['n02958343_11246-5.png'],\n",
              " 'n02958343_11365-1.jpg': ['n02958343_11365-1.png'],\n",
              " 'n02958343_11365-2.jpg': ['n02958343_11365-2.png'],\n",
              " 'n02958343_11365-3.jpg': ['n02958343_11365-3.png'],\n",
              " 'n02958343_11365-4.jpg': ['n02958343_11365-4.png'],\n",
              " 'n02958343_11365-5.jpg': ['n02958343_11365-5.png'],\n",
              " 'n02958343_12252-1.jpg': ['n02958343_12252-1.png'],\n",
              " 'n02958343_12252-2.jpg': ['n02958343_12252-2.png'],\n",
              " 'n02958343_12252-3.jpg': ['n02958343_12252-3.png'],\n",
              " 'n02958343_12252-4.jpg': ['n02958343_12252-4.png'],\n",
              " 'n02958343_12252-5.jpg': ['n02958343_12252-5.png'],\n",
              " 'n02958343_12509-1.jpg': ['n02958343_12509-1.png'],\n",
              " 'n02958343_12509-2.jpg': ['n02958343_12509-2.png'],\n",
              " 'n02958343_12509-3.jpg': ['n02958343_12509-3.png'],\n",
              " 'n02958343_12509-4.jpg': ['n02958343_12509-4.png'],\n",
              " 'n02958343_12509-5.jpg': ['n02958343_12509-5.png'],\n",
              " 'n02958343_12557-1.jpg': ['n02958343_12557-1.png'],\n",
              " 'n02958343_12557-2.jpg': ['n02958343_12557-2.png'],\n",
              " 'n02958343_12557-3.jpg': ['n02958343_12557-3.png'],\n",
              " 'n02958343_12557-4.jpg': ['n02958343_12557-4.png'],\n",
              " 'n02958343_12557-5.jpg': ['n02958343_12557-5.png'],\n",
              " 'n02958343_12557-6.jpg': ['n02958343_12557-6.png'],\n",
              " 'n02958343_12624-1.jpg': ['n02958343_12624-1.png'],\n",
              " 'n02958343_12624-2.jpg': ['n02958343_12624-2.png'],\n",
              " 'n02958343_12624-3.jpg': ['n02958343_12624-3.png'],\n",
              " 'n02958343_12624-4.jpg': ['n02958343_12624-4.png'],\n",
              " 'n02958343_12624-5.jpg': ['n02958343_12624-5.png'],\n",
              " 'n02958343_13036-1.jpg': ['n02958343_13036-1.png'],\n",
              " 'n02958343_13036-2.jpg': ['n02958343_13036-2.png'],\n",
              " 'n02958343_13036-3.jpg': ['n02958343_13036-3.png'],\n",
              " 'n02958343_13036-4.jpg': ['n02958343_13036-4.png'],\n",
              " 'n02958343_13036-5.jpg': ['n02958343_13036-5.png'],\n",
              " 'n02958343_13036-6.jpg': ['n02958343_13036-6.png'],\n",
              " 'n02958343_13615-1.jpg': ['n02958343_13615-1.png'],\n",
              " 'n02958343_13615-2.jpg': ['n02958343_13615-2.png'],\n",
              " 'n02958343_13615-3.jpg': ['n02958343_13615-3.png'],\n",
              " 'n02958343_13615-4.jpg': ['n02958343_13615-4.png'],\n",
              " 'n02958343_13615-5.jpg': ['n02958343_13615-5.png'],\n",
              " 'n02958343_14347-1.jpg': ['n02958343_14347-1.png'],\n",
              " 'n02958343_14347-2.jpg': ['n02958343_14347-2.png'],\n",
              " 'n02958343_14347-3.jpg': ['n02958343_14347-3.png'],\n",
              " 'n02958343_14347-4.jpg': ['n02958343_14347-4.png'],\n",
              " 'n02958343_14347-5.jpg': ['n02958343_14347-5.png'],\n",
              " 'n02958343_14365-1.jpg': ['n02958343_14365-1.png'],\n",
              " 'n02958343_14365-2.jpg': ['n02958343_14365-2.png'],\n",
              " 'n02958343_14365-3.jpg': ['n02958343_14365-3.png'],\n",
              " 'n02958343_14365-4.jpg': ['n02958343_14365-4.png'],\n",
              " 'n02958343_14365-5.jpg': ['n02958343_14365-5.png'],\n",
              " 'n02958343_14365-6.jpg': ['n02958343_14365-6.png'],\n",
              " 'n02958343_14732-1.jpg': ['n02958343_14732-1.png'],\n",
              " 'n02958343_14732-2.jpg': ['n02958343_14732-2.png'],\n",
              " 'n02958343_14732-3.jpg': ['n02958343_14732-3.png'],\n",
              " 'n02958343_14732-4.jpg': ['n02958343_14732-4.png'],\n",
              " 'n02958343_14732-5.jpg': ['n02958343_14732-5.png'],\n",
              " 'n02958343_1653-1.jpg': ['n02958343_1653-1.png'],\n",
              " 'n02958343_1653-2.jpg': ['n02958343_1653-2.png'],\n",
              " 'n02958343_1653-3.jpg': ['n02958343_1653-3.png'],\n",
              " 'n02958343_1653-4.jpg': ['n02958343_1653-4.png'],\n",
              " 'n02958343_1653-5.jpg': ['n02958343_1653-5.png'],\n",
              " 'n02958343_1653-6.jpg': ['n02958343_1653-6.png'],\n",
              " 'n02958343_1853-1.jpg': ['n02958343_1853-1.png'],\n",
              " 'n02958343_1853-2.jpg': ['n02958343_1853-2.png'],\n",
              " 'n02958343_1853-3.jpg': ['n02958343_1853-3.png'],\n",
              " 'n02958343_1853-4.jpg': ['n02958343_1853-4.png'],\n",
              " 'n02958343_1853-5.jpg': ['n02958343_1853-5.png'],\n",
              " 'n02958343_1853-6.jpg': ['n02958343_1853-6.png'],\n",
              " 'n02958343_20771-1.jpg': ['n02958343_20771-1.png'],\n",
              " 'n02958343_20771-2.jpg': ['n02958343_20771-2.png'],\n",
              " 'n02958343_20771-3.jpg': ['n02958343_20771-3.png'],\n",
              " 'n02958343_20771-4.jpg': ['n02958343_20771-4.png'],\n",
              " 'n02958343_20771-5.jpg': ['n02958343_20771-5.png'],\n",
              " 'n02958343_2417-1.jpg': ['n02958343_2417-1.png'],\n",
              " 'n02958343_2417-2.jpg': ['n02958343_2417-2.png'],\n",
              " 'n02958343_2417-3.jpg': ['n02958343_2417-3.png'],\n",
              " 'n02958343_2417-4.jpg': ['n02958343_2417-4.png'],\n",
              " 'n02958343_2417-5.jpg': ['n02958343_2417-5.png'],\n",
              " 'n02958343_255-1.jpg': ['n02958343_255-1.png'],\n",
              " 'n02958343_255-2.jpg': ['n02958343_255-2.png'],\n",
              " 'n02958343_255-3.jpg': ['n02958343_255-3.png'],\n",
              " 'n02958343_255-4.jpg': ['n02958343_255-4.png'],\n",
              " 'n02958343_255-5.jpg': ['n02958343_255-5.png'],\n",
              " 'n02958343_3561-1.jpg': ['n02958343_3561-1.png'],\n",
              " 'n02958343_3561-2.jpg': ['n02958343_3561-2.png'],\n",
              " 'n02958343_3561-3.jpg': ['n02958343_3561-3.png'],\n",
              " 'n02958343_3561-4.jpg': ['n02958343_3561-4.png'],\n",
              " 'n02958343_3561-5.jpg': ['n02958343_3561-5.png'],\n",
              " 'n02958343_3671-1.jpg': ['n02958343_3671-1.png'],\n",
              " 'n02958343_3671-2.jpg': ['n02958343_3671-2.png'],\n",
              " 'n02958343_3671-3.jpg': ['n02958343_3671-3.png'],\n",
              " 'n02958343_3671-4.jpg': ['n02958343_3671-4.png'],\n",
              " 'n02958343_3671-5.jpg': ['n02958343_3671-5.png'],\n",
              " 'n02958343_3763-1.jpg': ['n02958343_3763-1.png'],\n",
              " 'n02958343_3763-2.jpg': ['n02958343_3763-2.png'],\n",
              " 'n02958343_3763-3.jpg': ['n02958343_3763-3.png'],\n",
              " 'n02958343_3763-4.jpg': ['n02958343_3763-4.png'],\n",
              " 'n02958343_3763-5.jpg': ['n02958343_3763-5.png'],\n",
              " 'n02958343_3787-1.jpg': ['n02958343_3787-1.png'],\n",
              " 'n02958343_3787-2.jpg': ['n02958343_3787-2.png'],\n",
              " 'n02958343_3787-3.jpg': ['n02958343_3787-3.png'],\n",
              " 'n02958343_3787-4.jpg': ['n02958343_3787-4.png'],\n",
              " 'n02958343_3787-5.jpg': ['n02958343_3787-5.png'],\n",
              " 'n02958343_3930-1.jpg': ['n02958343_3930-1.png'],\n",
              " 'n02958343_3930-2.jpg': ['n02958343_3930-2.png'],\n",
              " 'n02958343_3930-3.jpg': ['n02958343_3930-3.png'],\n",
              " 'n02958343_3930-4.jpg': ['n02958343_3930-4.png'],\n",
              " 'n02958343_3930-5.jpg': ['n02958343_3930-5.png'],\n",
              " 'n02958343_4120-1.jpg': ['n02958343_4120-1.png'],\n",
              " 'n02958343_4120-2.jpg': ['n02958343_4120-2.png'],\n",
              " 'n02958343_4120-3.jpg': ['n02958343_4120-3.png'],\n",
              " 'n02958343_4120-4.jpg': ['n02958343_4120-4.png'],\n",
              " 'n02958343_4120-5.jpg': ['n02958343_4120-5.png'],\n",
              " 'n02958343_4294-1.jpg': ['n02958343_4294-1.png'],\n",
              " 'n02958343_4294-2.jpg': ['n02958343_4294-2.png'],\n",
              " 'n02958343_4294-3.jpg': ['n02958343_4294-3.png'],\n",
              " 'n02958343_4294-4.jpg': ['n02958343_4294-4.png'],\n",
              " 'n02958343_4294-5.jpg': ['n02958343_4294-5.png'],\n",
              " 'n02958343_4414-1.jpg': ['n02958343_4414-1.png'],\n",
              " 'n02958343_4414-2.jpg': ['n02958343_4414-2.png'],\n",
              " 'n02958343_4414-3.jpg': ['n02958343_4414-3.png'],\n",
              " 'n02958343_4414-4.jpg': ['n02958343_4414-4.png'],\n",
              " 'n02958343_4414-5.jpg': ['n02958343_4414-5.png'],\n",
              " 'n02958343_4414-6.jpg': ['n02958343_4414-6.png'],\n",
              " 'n02958343_4632-1.jpg': ['n02958343_4632-1.png'],\n",
              " 'n02958343_4632-2.jpg': ['n02958343_4632-2.png'],\n",
              " 'n02958343_4632-3.jpg': ['n02958343_4632-3.png'],\n",
              " 'n02958343_4632-4.jpg': ['n02958343_4632-4.png'],\n",
              " 'n02958343_4632-5.jpg': ['n02958343_4632-5.png'],\n",
              " 'n02958343_4632-6.jpg': ['n02958343_4632-6.png'],\n",
              " 'n02958343_4750-1.jpg': ['n02958343_4750-1.png'],\n",
              " 'n02958343_4750-2.jpg': ['n02958343_4750-2.png'],\n",
              " 'n02958343_4750-3.jpg': ['n02958343_4750-3.png'],\n",
              " 'n02958343_4750-4.jpg': ['n02958343_4750-4.png'],\n",
              " 'n02958343_4750-5.jpg': ['n02958343_4750-5.png'],\n",
              " 'n02958343_494-1.jpg': ['n02958343_494-1.png'],\n",
              " 'n02958343_494-2.jpg': ['n02958343_494-2.png'],\n",
              " 'n02958343_494-3.jpg': ['n02958343_494-3.png'],\n",
              " 'n02958343_494-4.jpg': ['n02958343_494-4.png'],\n",
              " 'n02958343_494-5.jpg': ['n02958343_494-5.png'],\n",
              " 'n02958343_494-6.jpg': ['n02958343_494-6.png'],\n",
              " 'n02958343_520-1.jpg': ['n02958343_520-1.png'],\n",
              " 'n02958343_520-2.jpg': ['n02958343_520-2.png'],\n",
              " 'n02958343_520-3.jpg': ['n02958343_520-3.png'],\n",
              " 'n02958343_520-4.jpg': ['n02958343_520-4.png'],\n",
              " 'n02958343_520-5.jpg': ['n02958343_520-5.png'],\n",
              " 'n02958343_520-6.jpg': ['n02958343_520-6.png'],\n",
              " 'n02958343_5353-1.jpg': ['n02958343_5353-1.png'],\n",
              " 'n02958343_5353-2.jpg': ['n02958343_5353-2.png'],\n",
              " 'n02958343_5353-3.jpg': ['n02958343_5353-3.png'],\n",
              " 'n02958343_5353-4.jpg': ['n02958343_5353-4.png'],\n",
              " 'n02958343_5353-5.jpg': ['n02958343_5353-5.png'],\n",
              " 'n02958343_53735-1.jpg': ['n02958343_53735-1.png'],\n",
              " 'n02958343_53735-2.jpg': ['n02958343_53735-2.png'],\n",
              " 'n02958343_53735-3.jpg': ['n02958343_53735-3.png'],\n",
              " 'n02958343_53735-4.jpg': ['n02958343_53735-4.png'],\n",
              " 'n02958343_53735-5.jpg': ['n02958343_53735-5.png'],\n",
              " 'n02958343_53735-6.jpg': ['n02958343_53735-6.png'],\n",
              " 'n02958343_5640-1.jpg': ['n02958343_5640-1.png'],\n",
              " 'n02958343_5640-2.jpg': ['n02958343_5640-2.png'],\n",
              " 'n02958343_5640-3.jpg': ['n02958343_5640-3.png'],\n",
              " 'n02958343_5640-4.jpg': ['n02958343_5640-4.png'],\n",
              " 'n02958343_5640-5.jpg': ['n02958343_5640-5.png'],\n",
              " 'n02958343_5640-6.jpg': ['n02958343_5640-6.png'],\n",
              " 'n02958343_5820-1.jpg': ['n02958343_5820-1.png'],\n",
              " 'n02958343_5820-2.jpg': ['n02958343_5820-2.png'],\n",
              " 'n02958343_5820-3.jpg': ['n02958343_5820-3.png'],\n",
              " 'n02958343_5820-4.jpg': ['n02958343_5820-4.png'],\n",
              " 'n02958343_5820-5.jpg': ['n02958343_5820-5.png'],\n",
              " 'n02958343_5820-6.jpg': ['n02958343_5820-6.png'],\n",
              " 'n02958343_6233-1.jpg': ['n02958343_6233-1.png'],\n",
              " 'n02958343_6233-2.jpg': ['n02958343_6233-2.png'],\n",
              " 'n02958343_6233-3.jpg': ['n02958343_6233-3.png'],\n",
              " 'n02958343_6233-4.jpg': ['n02958343_6233-4.png'],\n",
              " 'n02958343_6233-5.jpg': ['n02958343_6233-5.png'],\n",
              " 'n02958343_6233-6.jpg': ['n02958343_6233-6.png'],\n",
              " 'n02958343_6530-1.jpg': ['n02958343_6530-1.png'],\n",
              " 'n02958343_6530-2.jpg': ['n02958343_6530-2.png'],\n",
              " 'n02958343_6530-3.jpg': ['n02958343_6530-3.png'],\n",
              " 'n02958343_6530-4.jpg': ['n02958343_6530-4.png'],\n",
              " 'n02958343_6530-5.jpg': ['n02958343_6530-5.png'],\n",
              " 'n02958343_6586-1.jpg': ['n02958343_6586-1.png'],\n",
              " 'n02958343_6586-2.jpg': ['n02958343_6586-2.png'],\n",
              " 'n02958343_6586-3.jpg': ['n02958343_6586-3.png'],\n",
              " 'n02958343_6586-4.jpg': ['n02958343_6586-4.png'],\n",
              " 'n02958343_6586-5.jpg': ['n02958343_6586-5.png'],\n",
              " 'n02958343_8028-1.jpg': ['n02958343_8028-1.png'],\n",
              " 'n02958343_8028-2.jpg': ['n02958343_8028-2.png'],\n",
              " 'n02958343_8028-3.jpg': ['n02958343_8028-3.png'],\n",
              " 'n02958343_8028-4.jpg': ['n02958343_8028-4.png'],\n",
              " 'n02958343_8028-5.jpg': ['n02958343_8028-5.png'],\n",
              " 'n02958343_8028-6.jpg': ['n02958343_8028-6.png'],\n",
              " 'n02958343_8642-1.jpg': ['n02958343_8642-1.png'],\n",
              " 'n02958343_8642-2.jpg': ['n02958343_8642-2.png'],\n",
              " 'n02958343_8642-3.jpg': ['n02958343_8642-3.png'],\n",
              " 'n02958343_8642-4.jpg': ['n02958343_8642-4.png'],\n",
              " 'n02958343_8642-5.jpg': ['n02958343_8642-5.png'],\n",
              " 'n02958343_8642-6.jpg': ['n02958343_8642-6.png'],\n",
              " 'n02958343_8789-1.jpg': ['n02958343_8789-1.png'],\n",
              " 'n02958343_8789-2.jpg': ['n02958343_8789-2.png'],\n",
              " 'n02958343_8789-3.jpg': ['n02958343_8789-3.png'],\n",
              " 'n02958343_8789-4.jpg': ['n02958343_8789-4.png'],\n",
              " 'n02958343_8789-5.jpg': ['n02958343_8789-5.png'],\n",
              " 'n02958343_8789-6.jpg': ['n02958343_8789-6.png'],\n",
              " 'n02958343_9022-1.jpg': ['n02958343_9022-1.png'],\n",
              " 'n02958343_9022-2.jpg': ['n02958343_9022-2.png'],\n",
              " 'n02958343_9022-3.jpg': ['n02958343_9022-3.png'],\n",
              " 'n02958343_9022-4.jpg': ['n02958343_9022-4.png'],\n",
              " 'n02958343_9022-5.jpg': ['n02958343_9022-5.png'],\n",
              " 'n02958343_9912-1.jpg': ['n02958343_9912-1.png'],\n",
              " 'n02958343_9912-2.jpg': ['n02958343_9912-2.png'],\n",
              " 'n02958343_9912-3.jpg': ['n02958343_9912-3.png'],\n",
              " 'n02958343_9912-4.jpg': ['n02958343_9912-4.png'],\n",
              " 'n02958343_9912-5.jpg': ['n02958343_9912-5.png'],\n",
              " 'n04166281_1027-1.jpg': ['n04166281_1027-1.png'],\n",
              " 'n04166281_1027-2.jpg': ['n04166281_1027-2.png'],\n",
              " 'n04166281_1027-3.jpg': ['n04166281_1027-3.png'],\n",
              " 'n04166281_1027-4.jpg': ['n04166281_1027-4.png'],\n",
              " 'n04166281_1027-5.jpg': ['n04166281_1027-5.png'],\n",
              " 'n04166281_1059-1.jpg': ['n04166281_1059-1.png'],\n",
              " 'n04166281_1059-2.jpg': ['n04166281_1059-2.png'],\n",
              " 'n04166281_1059-3.jpg': ['n04166281_1059-3.png'],\n",
              " 'n04166281_1059-4.jpg': ['n04166281_1059-4.png'],\n",
              " 'n04166281_1059-5.jpg': ['n04166281_1059-5.png'],\n",
              " 'n04166281_1059-6.jpg': ['n04166281_1059-6.png'],\n",
              " 'n04166281_1059-7.jpg': ['n04166281_1059-7.png'],\n",
              " 'n04166281_11736-1.jpg': ['n04166281_11736-1.png'],\n",
              " 'n04166281_11736-2.jpg': ['n04166281_11736-2.png'],\n",
              " 'n04166281_11736-3.jpg': ['n04166281_11736-3.png'],\n",
              " 'n04166281_11736-4.jpg': ['n04166281_11736-4.png'],\n",
              " 'n04166281_11736-5.jpg': ['n04166281_11736-5.png'],\n",
              " 'n04166281_11736-6.jpg': ['n04166281_11736-6.png'],\n",
              " 'n04166281_1303-1.jpg': ['n04166281_1303-1.png'],\n",
              " 'n04166281_1303-2.jpg': ['n04166281_1303-2.png'],\n",
              " 'n04166281_1303-3.jpg': ['n04166281_1303-3.png'],\n",
              " 'n04166281_1303-4.jpg': ['n04166281_1303-4.png'],\n",
              " 'n04166281_1303-5.jpg': ['n04166281_1303-5.png'],\n",
              " 'n04166281_1303-6.jpg': ['n04166281_1303-6.png'],\n",
              " 'n04166281_1303-7.jpg': ['n04166281_1303-7.png'],\n",
              " 'n04166281_13380-1.jpg': ['n04166281_13380-1.png'],\n",
              " 'n04166281_13380-2.jpg': ['n04166281_13380-2.png'],\n",
              " 'n04166281_13380-3.jpg': ['n04166281_13380-3.png'],\n",
              " 'n04166281_13380-4.jpg': ['n04166281_13380-4.png'],\n",
              " 'n04166281_13380-5.jpg': ['n04166281_13380-5.png'],\n",
              " 'n04166281_13380-6.jpg': ['n04166281_13380-6.png'],\n",
              " 'n04166281_13380-7.jpg': ['n04166281_13380-7.png'],\n",
              " 'n04166281_13380-8.jpg': ['n04166281_13380-8.png'],\n",
              " 'n04166281_13380-9.jpg': ['n04166281_13380-9.png'],\n",
              " 'n04166281_14544-1.jpg': ['n04166281_14544-1.png'],\n",
              " 'n04166281_14544-2.jpg': ['n04166281_14544-2.png'],\n",
              " 'n04166281_14544-3.jpg': ['n04166281_14544-3.png'],\n",
              " 'n04166281_14544-4.jpg': ['n04166281_14544-4.png'],\n",
              " 'n04166281_14544-5.jpg': ['n04166281_14544-5.png'],\n",
              " 'n04166281_14544-6.jpg': ['n04166281_14544-6.png'],\n",
              " 'n04166281_185-1.jpg': ['n04166281_185-1.png'],\n",
              " 'n04166281_185-2.jpg': ['n04166281_185-2.png'],\n",
              " 'n04166281_185-3.jpg': ['n04166281_185-3.png'],\n",
              " 'n04166281_185-4.jpg': ['n04166281_185-4.png'],\n",
              " 'n04166281_185-5.jpg': ['n04166281_185-5.png'],\n",
              " 'n04166281_185-6.jpg': ['n04166281_185-6.png'],\n",
              " 'n04166281_185-7.jpg': ['n04166281_185-7.png'],\n",
              " 'n04166281_185-8.jpg': ['n04166281_185-8.png'],\n",
              " 'n04166281_1917-1.jpg': ['n04166281_1917-1.png'],\n",
              " 'n04166281_1917-2.jpg': ['n04166281_1917-2.png'],\n",
              " 'n04166281_1917-3.jpg': ['n04166281_1917-3.png'],\n",
              " 'n04166281_1917-4.jpg': ['n04166281_1917-4.png'],\n",
              " 'n04166281_1917-5.jpg': ['n04166281_1917-5.png'],\n",
              " 'n04166281_1917-6.jpg': ['n04166281_1917-6.png'],\n",
              " 'n04166281_1917-7.jpg': ['n04166281_1917-7.png'],\n",
              " 'n04166281_1917-8.jpg': ['n04166281_1917-8.png'],\n",
              " 'n04166281_19230-1.jpg': ['n04166281_19230-1.png'],\n",
              " 'n04166281_19230-2.jpg': ['n04166281_19230-2.png'],\n",
              " 'n04166281_19230-3.jpg': ['n04166281_19230-3.png'],\n",
              " 'n04166281_19230-4.jpg': ['n04166281_19230-4.png'],\n",
              " 'n04166281_19230-5.jpg': ['n04166281_19230-5.png'],\n",
              " 'n04166281_19230-6.jpg': ['n04166281_19230-6.png'],\n",
              " 'n04166281_1959-1.jpg': ['n04166281_1959-1.png'],\n",
              " 'n04166281_1959-2.jpg': ['n04166281_1959-2.png'],\n",
              " 'n04166281_1959-3.jpg': ['n04166281_1959-3.png'],\n",
              " 'n04166281_1959-4.jpg': ['n04166281_1959-4.png'],\n",
              " 'n04166281_1959-5.jpg': ['n04166281_1959-5.png'],\n",
              " 'n04166281_1959-6.jpg': ['n04166281_1959-6.png'],\n",
              " 'n04166281_1959-7.jpg': ['n04166281_1959-7.png'],\n",
              " 'n04166281_21492-1.jpg': ['n04166281_21492-1.png'],\n",
              " 'n04166281_21492-2.jpg': ['n04166281_21492-2.png'],\n",
              " 'n04166281_21492-3.jpg': ['n04166281_21492-3.png'],\n",
              " 'n04166281_21492-4.jpg': ['n04166281_21492-4.png'],\n",
              " 'n04166281_21492-5.jpg': ['n04166281_21492-5.png'],\n",
              " 'n04166281_21492-6.jpg': ['n04166281_21492-6.png'],\n",
              " 'n04166281_21492-7.jpg': ['n04166281_21492-7.png'],\n",
              " 'n04166281_21492-8.jpg': ['n04166281_21492-8.png'],\n",
              " 'n04166281_21492-9.jpg': ['n04166281_21492-9.png'],\n",
              " 'n04166281_22321-1.jpg': ['n04166281_22321-10.png', 'n04166281_22321-1.png'],\n",
              " 'n04166281_22321-10.jpg': ['n04166281_22321-10.png'],\n",
              " 'n04166281_22321-2.jpg': ['n04166281_22321-2.png'],\n",
              " 'n04166281_22321-3.jpg': ['n04166281_22321-3.png'],\n",
              " 'n04166281_22321-4.jpg': ['n04166281_22321-4.png'],\n",
              " 'n04166281_22321-5.jpg': ['n04166281_22321-5.png'],\n",
              " 'n04166281_22321-6.jpg': ['n04166281_22321-6.png'],\n",
              " 'n04166281_22321-7.jpg': ['n04166281_22321-7.png'],\n",
              " 'n04166281_22321-8.jpg': ['n04166281_22321-8.png'],\n",
              " 'n04166281_22321-9.jpg': ['n04166281_22321-9.png'],\n",
              " 'n04166281_2370-1.jpg': ['n04166281_2370-1.png'],\n",
              " 'n04166281_2370-2.jpg': ['n04166281_2370-2.png'],\n",
              " 'n04166281_2370-3.jpg': ['n04166281_2370-3.png'],\n",
              " 'n04166281_2370-4.jpg': ['n04166281_2370-4.png'],\n",
              " 'n04166281_2370-5.jpg': ['n04166281_2370-5.png'],\n",
              " 'n04166281_2370-6.jpg': ['n04166281_2370-6.png'],\n",
              " 'n04166281_23980-1.jpg': ['n04166281_23980-1.png'],\n",
              " 'n04166281_23980-2.jpg': ['n04166281_23980-2.png'],\n",
              " 'n04166281_23980-3.jpg': ['n04166281_23980-3.png'],\n",
              " 'n04166281_23980-4.jpg': ['n04166281_23980-4.png'],\n",
              " 'n04166281_23980-5.jpg': ['n04166281_23980-5.png'],\n",
              " 'n04166281_23980-6.jpg': ['n04166281_23980-6.png'],\n",
              " 'n04166281_23980-7.jpg': ['n04166281_23980-7.png'],\n",
              " 'n04166281_23980-8.jpg': ['n04166281_23980-8.png'],\n",
              " 'n04166281_23980-9.jpg': ['n04166281_23980-9.png'],\n",
              " 'n04166281_241-1.jpg': ['n04166281_241-1.png'],\n",
              " 'n04166281_241-2.jpg': ['n04166281_241-2.png'],\n",
              " 'n04166281_241-3.jpg': ['n04166281_241-3.png'],\n",
              " 'n04166281_241-4.jpg': ['n04166281_241-4.png'],\n",
              " 'n04166281_241-5.jpg': ['n04166281_241-5.png'],\n",
              " 'n04166281_241-6.jpg': ['n04166281_241-6.png'],\n",
              " 'n04166281_2502-1.jpg': ['n04166281_2502-1.png'],\n",
              " 'n04166281_2502-2.jpg': ['n04166281_2502-2.png'],\n",
              " 'n04166281_2502-3.jpg': ['n04166281_2502-3.png'],\n",
              " 'n04166281_2502-4.jpg': ['n04166281_2502-4.png'],\n",
              " 'n04166281_2502-5.jpg': ['n04166281_2502-5.png'],\n",
              " 'n04166281_2502-6.jpg': ['n04166281_2502-6.png'],\n",
              " 'n04166281_2502-7.jpg': ['n04166281_2502-7.png'],\n",
              " 'n04166281_2502-8.jpg': ['n04166281_2502-8.png'],\n",
              " 'n04166281_2502-9.jpg': ['n04166281_2502-9.png'],\n",
              " 'n04166281_2596-1.jpg': ['n04166281_2596-1.png'],\n",
              " 'n04166281_2596-2.jpg': ['n04166281_2596-2.png'],\n",
              " 'n04166281_2596-3.jpg': ['n04166281_2596-3.png'],\n",
              " 'n04166281_2596-4.jpg': ['n04166281_2596-4.png'],\n",
              " 'n04166281_2596-5.jpg': ['n04166281_2596-5.png'],\n",
              " 'n04166281_2596-6.jpg': ['n04166281_2596-6.png'],\n",
              " 'n04166281_2596-7.jpg': ['n04166281_2596-7.png'],\n",
              " 'n04166281_263-1.jpg': ['n04166281_263-1.png', 'n04166281_263-10.png'],\n",
              " 'n04166281_263-10.jpg': ['n04166281_263-10.png'],\n",
              " 'n04166281_263-2.jpg': ['n04166281_263-2.png'],\n",
              " 'n04166281_263-3.jpg': ['n04166281_263-3.png'],\n",
              " 'n04166281_263-4.jpg': ['n04166281_263-4.png'],\n",
              " 'n04166281_263-5.jpg': ['n04166281_263-5.png'],\n",
              " 'n04166281_263-6.jpg': ['n04166281_263-6.png'],\n",
              " 'n04166281_263-7.jpg': ['n04166281_263-7.png'],\n",
              " 'n04166281_263-8.jpg': ['n04166281_263-8.png'],\n",
              " 'n04166281_263-9.jpg': ['n04166281_263-9.png'],\n",
              " 'n04166281_28282-1.jpg': ['n04166281_28282-1.png'],\n",
              " 'n04166281_28282-2.jpg': ['n04166281_28282-2.png'],\n",
              " 'n04166281_28282-3.jpg': ['n04166281_28282-3.png'],\n",
              " 'n04166281_28282-4.jpg': ['n04166281_28282-4.png'],\n",
              " 'n04166281_28282-5.jpg': ['n04166281_28282-5.png'],\n",
              " 'n04166281_28282-6.jpg': ['n04166281_28282-6.png'],\n",
              " 'n04166281_28282-7.jpg': ['n04166281_28282-7.png'],\n",
              " 'n04166281_28282-8.jpg': ['n04166281_28282-8.png'],\n",
              " 'n04166281_28282-9.jpg': ['n04166281_28282-9.png'],\n",
              " 'n04166281_2972-1.jpg': ['n04166281_2972-1.png'],\n",
              " 'n04166281_2972-2.jpg': ['n04166281_2972-2.png'],\n",
              " 'n04166281_2972-3.jpg': ['n04166281_2972-3.png'],\n",
              " 'n04166281_2972-4.jpg': ['n04166281_2972-4.png'],\n",
              " 'n04166281_2972-5.jpg': ['n04166281_2972-5.png'],\n",
              " 'n04166281_2972-6.jpg': ['n04166281_2972-6.png'],\n",
              " 'n04166281_3840-1.jpg': ['n04166281_3840-1.png'],\n",
              " 'n04166281_3840-2.jpg': ['n04166281_3840-2.png'],\n",
              " 'n04166281_3840-3.jpg': ['n04166281_3840-3.png'],\n",
              " 'n04166281_3840-4.jpg': ['n04166281_3840-4.png'],\n",
              " 'n04166281_3840-5.jpg': ['n04166281_3840-5.png'],\n",
              " 'n04166281_388-1.jpg': ['n04166281_388-1.png'],\n",
              " 'n04166281_388-2.jpg': ['n04166281_388-2.png'],\n",
              " 'n04166281_388-3.jpg': ['n04166281_388-3.png'],\n",
              " 'n04166281_388-4.jpg': ['n04166281_388-4.png'],\n",
              " 'n04166281_388-5.jpg': ['n04166281_388-5.png'],\n",
              " 'n04166281_4002-1.jpg': ['n04166281_4002-1.png'],\n",
              " 'n04166281_4002-2.jpg': ['n04166281_4002-2.png'],\n",
              " 'n04166281_4002-3.jpg': ['n04166281_4002-3.png'],\n",
              " 'n04166281_4002-4.jpg': ['n04166281_4002-4.png'],\n",
              " 'n04166281_4002-5.jpg': ['n04166281_4002-5.png'],\n",
              " 'n04166281_4002-6.jpg': ['n04166281_4002-6.png'],\n",
              " 'n04166281_4166-1.jpg': ['n04166281_4166-1.png'],\n",
              " 'n04166281_4166-2.jpg': ['n04166281_4166-2.png'],\n",
              " 'n04166281_4166-3.jpg': ['n04166281_4166-3.png'],\n",
              " 'n04166281_4166-4.jpg': ['n04166281_4166-4.png'],\n",
              " 'n04166281_4166-5.jpg': ['n04166281_4166-5.png'],\n",
              " 'n04166281_4376-1.jpg': ['n04166281_4376-1.png'],\n",
              " 'n04166281_4376-2.jpg': ['n04166281_4376-2.png'],\n",
              " 'n04166281_4376-3.jpg': ['n04166281_4376-3.png'],\n",
              " 'n04166281_4376-4.jpg': ['n04166281_4376-4.png'],\n",
              " 'n04166281_4376-5.jpg': ['n04166281_4376-5.png'],\n",
              " 'n04166281_4376-6.jpg': ['n04166281_4376-6.png'],\n",
              " 'n04166281_4401-1.jpg': ['n04166281_4401-1.png'],\n",
              " 'n04166281_4401-2.jpg': ['n04166281_4401-2.png'],\n",
              " 'n04166281_4401-3.jpg': ['n04166281_4401-3.png'],\n",
              " 'n04166281_4401-4.jpg': ['n04166281_4401-4.png'],\n",
              " 'n04166281_4401-5.jpg': ['n04166281_4401-5.png'],\n",
              " 'n04166281_4401-6.jpg': ['n04166281_4401-6.png'],\n",
              " 'n04166281_4416-1.jpg': ['n04166281_4416-1.png'],\n",
              " 'n04166281_4416-2.jpg': ['n04166281_4416-2.png'],\n",
              " 'n04166281_4416-3.jpg': ['n04166281_4416-3.png'],\n",
              " 'n04166281_4416-4.jpg': ['n04166281_4416-4.png'],\n",
              " 'n04166281_4416-5.jpg': ['n04166281_4416-5.png'],\n",
              " 'n04166281_4416-6.jpg': ['n04166281_4416-6.png'],\n",
              " 'n04166281_4467-1.jpg': ['n04166281_4467-1.png'],\n",
              " 'n04166281_4467-2.jpg': ['n04166281_4467-2.png'],\n",
              " 'n04166281_4467-3.jpg': ['n04166281_4467-3.png'],\n",
              " 'n04166281_4467-4.jpg': ['n04166281_4467-4.png'],\n",
              " 'n04166281_4467-5.jpg': ['n04166281_4467-5.png'],\n",
              " 'n04166281_4467-6.jpg': ['n04166281_4467-6.png'],\n",
              " 'n04166281_4467-7.jpg': ['n04166281_4467-7.png'],\n",
              " 'n04166281_4467-8.jpg': ['n04166281_4467-8.png'],\n",
              " 'n04166281_4467-9.jpg': ['n04166281_4467-9.png'],\n",
              " 'n04166281_4661-1.jpg': ['n04166281_4661-1.png'],\n",
              " 'n04166281_4661-2.jpg': ['n04166281_4661-2.png'],\n",
              " 'n04166281_4661-3.jpg': ['n04166281_4661-3.png'],\n",
              " 'n04166281_4661-4.jpg': ['n04166281_4661-4.png'],\n",
              " 'n04166281_4661-5.jpg': ['n04166281_4661-5.png'],\n",
              " 'n04166281_4869-1.jpg': ['n04166281_4869-1.png'],\n",
              " 'n04166281_4869-2.jpg': ['n04166281_4869-2.png'],\n",
              " 'n04166281_4869-3.jpg': ['n04166281_4869-3.png'],\n",
              " 'n04166281_4869-4.jpg': ['n04166281_4869-4.png'],\n",
              " 'n04166281_4869-5.jpg': ['n04166281_4869-5.png'],\n",
              " 'n04166281_4869-6.jpg': ['n04166281_4869-6.png'],\n",
              " 'n04166281_4869-7.jpg': ['n04166281_4869-7.png'],\n",
              " 'n04166281_4869-8.jpg': ['n04166281_4869-8.png'],\n",
              " 'n04166281_4915-1.jpg': ['n04166281_4915-1.png'],\n",
              " 'n04166281_4915-2.jpg': ['n04166281_4915-2.png'],\n",
              " 'n04166281_4915-3.jpg': ['n04166281_4915-3.png'],\n",
              " 'n04166281_4915-4.jpg': ['n04166281_4915-4.png'],\n",
              " 'n04166281_4915-5.jpg': ['n04166281_4915-5.png'],\n",
              " 'n04166281_4915-6.jpg': ['n04166281_4915-6.png'],\n",
              " 'n04166281_4915-7.jpg': ['n04166281_4915-7.png'],\n",
              " 'n04166281_4915-8.jpg': ['n04166281_4915-8.png'],\n",
              " 'n04166281_4975-1.jpg': ['n04166281_4975-1.png'],\n",
              " 'n04166281_4975-2.jpg': ['n04166281_4975-2.png'],\n",
              " 'n04166281_4975-3.jpg': ['n04166281_4975-3.png'],\n",
              " 'n04166281_4975-4.jpg': ['n04166281_4975-4.png'],\n",
              " 'n04166281_4975-5.jpg': ['n04166281_4975-5.png'],\n",
              " 'n04166281_4975-6.jpg': ['n04166281_4975-6.png'],\n",
              " 'n04166281_4975-7.jpg': ['n04166281_4975-7.png'],\n",
              " 'n04166281_4975-8.jpg': ['n04166281_4975-8.png'],\n",
              " 'n04166281_4975-9.jpg': ['n04166281_4975-9.png'],\n",
              " 'n04166281_5125-1.jpg': ['n04166281_5125-10.png', 'n04166281_5125-1.png'],\n",
              " 'n04166281_5125-10.jpg': ['n04166281_5125-10.png'],\n",
              " 'n04166281_5125-2.jpg': ['n04166281_5125-2.png'],\n",
              " 'n04166281_5125-3.jpg': ['n04166281_5125-3.png'],\n",
              " 'n04166281_5125-4.jpg': ['n04166281_5125-4.png'],\n",
              " 'n04166281_5125-5.jpg': ['n04166281_5125-5.png'],\n",
              " 'n04166281_5125-6.jpg': ['n04166281_5125-6.png'],\n",
              " 'n04166281_5125-7.jpg': ['n04166281_5125-7.png'],\n",
              " 'n04166281_5125-8.jpg': ['n04166281_5125-8.png'],\n",
              " 'n04166281_5125-9.jpg': ['n04166281_5125-9.png'],\n",
              " 'n04166281_5134-1.jpg': ['n04166281_5134-1.png'],\n",
              " 'n04166281_5134-2.jpg': ['n04166281_5134-2.png'],\n",
              " 'n04166281_5134-3.jpg': ['n04166281_5134-3.png'],\n",
              " 'n04166281_5134-4.jpg': ['n04166281_5134-4.png'],\n",
              " 'n04166281_5134-5.jpg': ['n04166281_5134-5.png'],\n",
              " 'n04166281_534-1.jpg': ['n04166281_534-1.png', 'n04166281_534-10.png'],\n",
              " 'n04166281_534-10.jpg': ['n04166281_534-10.png'],\n",
              " 'n04166281_534-2.jpg': ['n04166281_534-2.png'],\n",
              " 'n04166281_534-3.jpg': ['n04166281_534-3.png'],\n",
              " 'n04166281_534-4.jpg': ['n04166281_534-4.png'],\n",
              " 'n04166281_534-5.jpg': ['n04166281_534-5.png'],\n",
              " 'n04166281_534-6.jpg': ['n04166281_534-6.png'],\n",
              " 'n04166281_534-7.jpg': ['n04166281_534-7.png'],\n",
              " 'n04166281_534-8.jpg': ['n04166281_534-8.png'],\n",
              " 'n04166281_534-9.jpg': ['n04166281_534-9.png'],\n",
              " 'n04166281_5354-1.jpg': ['n04166281_5354-1.png'],\n",
              " 'n04166281_5354-2.jpg': ['n04166281_5354-2.png'],\n",
              " 'n04166281_5354-3.jpg': ['n04166281_5354-3.png'],\n",
              " 'n04166281_5354-4.jpg': ['n04166281_5354-4.png'],\n",
              " 'n04166281_5354-5.jpg': ['n04166281_5354-5.png'],\n",
              " 'n04166281_5354-6.jpg': ['n04166281_5354-6.png'],\n",
              " 'n04166281_5438-1.jpg': ['n04166281_5438-10.png', 'n04166281_5438-1.png'],\n",
              " 'n04166281_5438-10.jpg': ['n04166281_5438-10.png'],\n",
              " 'n04166281_5438-2.jpg': ['n04166281_5438-2.png'],\n",
              " 'n04166281_5438-3.jpg': ['n04166281_5438-3.png'],\n",
              " 'n04166281_5438-4.jpg': ['n04166281_5438-4.png'],\n",
              " 'n04166281_5438-5.jpg': ['n04166281_5438-5.png'],\n",
              " 'n04166281_5438-6.jpg': ['n04166281_5438-6.png'],\n",
              " 'n04166281_5438-7.jpg': ['n04166281_5438-7.png'],\n",
              " 'n04166281_5438-8.jpg': ['n04166281_5438-8.png'],\n",
              " 'n04166281_5438-9.jpg': ['n04166281_5438-9.png'],\n",
              " 'n04166281_5506-1.jpg': ['n04166281_5506-1.png', 'n04166281_5506-10.png'],\n",
              " 'n04166281_5506-10.jpg': ['n04166281_5506-10.png'],\n",
              " 'n04166281_5506-2.jpg': ['n04166281_5506-2.png'],\n",
              " 'n04166281_5506-3.jpg': ['n04166281_5506-3.png'],\n",
              " 'n04166281_5506-4.jpg': ['n04166281_5506-4.png'],\n",
              " 'n04166281_5506-5.jpg': ['n04166281_5506-5.png'],\n",
              " 'n04166281_5506-6.jpg': ['n04166281_5506-6.png'],\n",
              " 'n04166281_5506-7.jpg': ['n04166281_5506-7.png'],\n",
              " 'n04166281_5506-8.jpg': ['n04166281_5506-8.png'],\n",
              " 'n04166281_5506-9.jpg': ['n04166281_5506-9.png'],\n",
              " 'n04166281_5545-1.jpg': ['n04166281_5545-1.png'],\n",
              " 'n04166281_5545-2.jpg': ['n04166281_5545-2.png'],\n",
              " 'n04166281_5545-3.jpg': ['n04166281_5545-3.png'],\n",
              " 'n04166281_5545-4.jpg': ['n04166281_5545-4.png'],\n",
              " 'n04166281_5545-5.jpg': ['n04166281_5545-5.png'],\n",
              " 'n04166281_5545-6.jpg': ['n04166281_5545-6.png'],\n",
              " 'n04166281_5551-1.jpg': ['n04166281_5551-1.png'],\n",
              " 'n04166281_5551-2.jpg': ['n04166281_5551-2.png'],\n",
              " 'n04166281_5551-3.jpg': ['n04166281_5551-3.png'],\n",
              " 'n04166281_5551-4.jpg': ['n04166281_5551-4.png'],\n",
              " 'n04166281_5551-5.jpg': ['n04166281_5551-5.png'],\n",
              " 'n04166281_5551-6.jpg': ['n04166281_5551-6.png'],\n",
              " 'n04166281_5551-7.jpg': ['n04166281_5551-7.png'],\n",
              " 'n04166281_5654-1.jpg': ['n04166281_5654-1.png'],\n",
              " 'n04166281_5654-2.jpg': ['n04166281_5654-2.png'],\n",
              " 'n04166281_5654-3.jpg': ['n04166281_5654-3.png'],\n",
              " 'n04166281_5654-4.jpg': ['n04166281_5654-4.png'],\n",
              " 'n04166281_5654-5.jpg': ['n04166281_5654-5.png'],\n",
              " 'n04166281_5719-1.jpg': ['n04166281_5719-1.png'],\n",
              " 'n04166281_5719-2.jpg': ['n04166281_5719-2.png'],\n",
              " 'n04166281_5719-3.jpg': ['n04166281_5719-3.png'],\n",
              " 'n04166281_5719-4.jpg': ['n04166281_5719-4.png'],\n",
              " 'n04166281_5719-5.jpg': ['n04166281_5719-5.png'],\n",
              " 'n04166281_5846-1.jpg': ['n04166281_5846-1.png'],\n",
              " 'n04166281_5846-2.jpg': ['n04166281_5846-2.png'],\n",
              " 'n04166281_5846-3.jpg': ['n04166281_5846-3.png'],\n",
              " 'n04166281_5846-4.jpg': ['n04166281_5846-4.png'],\n",
              " 'n04166281_5846-5.jpg': ['n04166281_5846-5.png'],\n",
              " 'n04166281_5846-6.jpg': ['n04166281_5846-6.png'],\n",
              " 'n04166281_5904-1.jpg': ['n04166281_5904-1.png'],\n",
              " 'n04166281_5904-2.jpg': ['n04166281_5904-2.png'],\n",
              " 'n04166281_5904-3.jpg': ['n04166281_5904-3.png'],\n",
              " 'n04166281_5904-4.jpg': ['n04166281_5904-4.png'],\n",
              " 'n04166281_5904-5.jpg': ['n04166281_5904-5.png'],\n",
              " 'n04166281_5904-6.jpg': ['n04166281_5904-6.png'],\n",
              " 'n04166281_6355-1.jpg': ['n04166281_6355-1.png'],\n",
              " 'n04166281_6355-2.jpg': ['n04166281_6355-2.png'],\n",
              " 'n04166281_6355-3.jpg': ['n04166281_6355-3.png'],\n",
              " 'n04166281_6355-4.jpg': ['n04166281_6355-4.png'],\n",
              " 'n04166281_6355-5.jpg': ['n04166281_6355-5.png'],\n",
              " 'n04166281_6355-6.jpg': ['n04166281_6355-6.png'],\n",
              " 'n04166281_6355-7.jpg': ['n04166281_6355-7.png'],\n",
              " 'n04166281_6355-8.jpg': ['n04166281_6355-8.png'],\n",
              " 'n04166281_6355-9.jpg': ['n04166281_6355-9.png'],\n",
              " 'n04166281_6358-1.jpg': ['n04166281_6358-1.png'],\n",
              " 'n04166281_6358-2.jpg': ['n04166281_6358-2.png'],\n",
              " 'n04166281_6358-3.jpg': ['n04166281_6358-3.png'],\n",
              " 'n04166281_6358-4.jpg': ['n04166281_6358-4.png'],\n",
              " 'n04166281_6358-5.jpg': ['n04166281_6358-5.png'],\n",
              " 'n04166281_6358-6.jpg': ['n04166281_6358-6.png'],\n",
              " 'n04166281_6358-7.jpg': ['n04166281_6358-7.png'],\n",
              " 'n04166281_6358-8.jpg': ['n04166281_6358-8.png'],\n",
              " 'n04166281_6358-9.jpg': ['n04166281_6358-9.png'],\n",
              " 'n04166281_6586-1.jpg': ['n04166281_6586-1.png'],\n",
              " 'n04166281_6586-2.jpg': ['n04166281_6586-2.png'],\n",
              " 'n04166281_6586-3.jpg': ['n04166281_6586-3.png'],\n",
              " 'n04166281_6586-4.jpg': ['n04166281_6586-4.png'],\n",
              " 'n04166281_6586-5.jpg': ['n04166281_6586-5.png'],\n",
              " 'n04166281_6586-6.jpg': ['n04166281_6586-6.png'],\n",
              " 'n04166281_6586-7.jpg': ['n04166281_6586-7.png'],\n",
              " 'n04166281_6586-8.jpg': ['n04166281_6586-8.png'],\n",
              " 'n04166281_6586-9.jpg': ['n04166281_6586-9.png'],\n",
              " 'n04166281_6626-1.jpg': ['n04166281_6626-1.png'],\n",
              " 'n04166281_6626-2.jpg': ['n04166281_6626-2.png'],\n",
              " 'n04166281_6626-3.jpg': ['n04166281_6626-3.png'],\n",
              " 'n04166281_6626-4.jpg': ['n04166281_6626-4.png'],\n",
              " 'n04166281_6626-5.jpg': ['n04166281_6626-5.png'],\n",
              " 'n04166281_6626-6.jpg': ['n04166281_6626-6.png'],\n",
              " 'n04166281_6665-1.jpg': ['n04166281_6665-1.png'],\n",
              " 'n04166281_6665-2.jpg': ['n04166281_6665-2.png'],\n",
              " 'n04166281_6665-3.jpg': ['n04166281_6665-3.png'],\n",
              " 'n04166281_6665-4.jpg': ['n04166281_6665-4.png'],\n",
              " 'n04166281_6665-5.jpg': ['n04166281_6665-5.png'],\n",
              " 'n04166281_6665-6.jpg': ['n04166281_6665-6.png'],\n",
              " 'n04166281_6665-7.jpg': ['n04166281_6665-7.png'],\n",
              " 'n04166281_6690-1.jpg': ['n04166281_6690-1.png'],\n",
              " 'n04166281_6690-2.jpg': ['n04166281_6690-2.png'],\n",
              " 'n04166281_6690-3.jpg': ['n04166281_6690-3.png'],\n",
              " 'n04166281_6690-4.jpg': ['n04166281_6690-4.png'],\n",
              " 'n04166281_6690-5.jpg': ['n04166281_6690-5.png'],\n",
              " 'n04166281_6690-6.jpg': ['n04166281_6690-6.png'],\n",
              " 'n04166281_6808-1.jpg': ['n04166281_6808-10.png', 'n04166281_6808-1.png'],\n",
              " 'n04166281_6808-10.jpg': ['n04166281_6808-10.png'],\n",
              " 'n04166281_6808-2.jpg': ['n04166281_6808-2.png'],\n",
              " 'n04166281_6808-3.jpg': ['n04166281_6808-3.png'],\n",
              " 'n04166281_6808-4.jpg': ['n04166281_6808-4.png'],\n",
              " 'n04166281_6808-5.jpg': ['n04166281_6808-5.png'],\n",
              " 'n04166281_6808-6.jpg': ['n04166281_6808-6.png'],\n",
              " 'n04166281_6808-7.jpg': ['n04166281_6808-7.png'],\n",
              " 'n04166281_6808-8.jpg': ['n04166281_6808-8.png'],\n",
              " 'n04166281_6808-9.jpg': ['n04166281_6808-9.png'],\n",
              " 'n04166281_69-1.jpg': ['n04166281_69-1.png'],\n",
              " 'n04166281_69-2.jpg': ['n04166281_69-2.png'],\n",
              " 'n04166281_69-3.jpg': ['n04166281_69-3.png'],\n",
              " 'n04166281_69-4.jpg': ['n04166281_69-4.png'],\n",
              " 'n04166281_69-5.jpg': ['n04166281_69-5.png'],\n",
              " 'n04166281_69-6.jpg': ['n04166281_69-6.png'],\n",
              " 'n04166281_69-7.jpg': ['n04166281_69-7.png'],\n",
              " 'n04166281_69-8.jpg': ['n04166281_69-8.png'],\n",
              " 'n04166281_7105-1.jpg': ['n04166281_7105-1.png'],\n",
              " 'n04166281_7105-2.jpg': ['n04166281_7105-2.png'],\n",
              " 'n04166281_7105-3.jpg': ['n04166281_7105-3.png'],\n",
              " 'n04166281_7105-4.jpg': ['n04166281_7105-4.png'],\n",
              " 'n04166281_7105-5.jpg': ['n04166281_7105-5.png'],\n",
              " 'n04166281_7105-6.jpg': ['n04166281_7105-6.png'],\n",
              " 'n04166281_7105-7.jpg': ['n04166281_7105-7.png'],\n",
              " 'n04166281_7105-8.jpg': ['n04166281_7105-8.png'],\n",
              " 'n04166281_7105-9.jpg': ['n04166281_7105-9.png'],\n",
              " 'n04166281_73-1.jpg': ['n04166281_73-1.png'],\n",
              " 'n04166281_73-2.jpg': ['n04166281_73-2.png'],\n",
              " 'n04166281_73-3.jpg': ['n04166281_73-3.png'],\n",
              " 'n04166281_73-4.jpg': ['n04166281_73-4.png'],\n",
              " 'n04166281_73-5.jpg': ['n04166281_73-5.png'],\n",
              " 'n04166281_73-6.jpg': ['n04166281_73-6.png'],\n",
              " 'n04166281_8236-1.jpg': ['n04166281_8236-1.png'],\n",
              " 'n04166281_8236-2.jpg': ['n04166281_8236-2.png'],\n",
              " 'n04166281_8236-3.jpg': ['n04166281_8236-3.png'],\n",
              " 'n04166281_8236-4.jpg': ['n04166281_8236-4.png'],\n",
              " 'n04166281_8236-5.jpg': ['n04166281_8236-5.png'],\n",
              " 'n04166281_8236-6.jpg': ['n04166281_8236-6.png'],\n",
              " 'n04166281_8236-7.jpg': ['n04166281_8236-7.png'],\n",
              " 'n04166281_927-1.jpg': ['n04166281_927-1.png'],\n",
              " 'n04166281_927-2.jpg': ['n04166281_927-2.png'],\n",
              " 'n04166281_927-3.jpg': ['n04166281_927-3.png'],\n",
              " 'n04166281_927-4.jpg': ['n04166281_927-4.png'],\n",
              " 'n04166281_927-5.jpg': ['n04166281_927-5.png'],\n",
              " 'n04166281_990-1.jpg': ['n04166281_990-1.png'],\n",
              " 'n04166281_990-2.jpg': ['n04166281_990-2.png'],\n",
              " 'n04166281_990-3.jpg': ['n04166281_990-3.png'],\n",
              " 'n04166281_990-4.jpg': ['n04166281_990-4.png'],\n",
              " 'n04166281_990-5.jpg': ['n04166281_990-5.png'],\n",
              " 'n04166281_990-6.jpg': ['n04166281_990-6.png']}"
            ]
          },
          "metadata": {},
          "execution_count": 10
        }
      ],
      "source": [
        "\n",
        "data=dict()\n",
        "for images in os.listdir(fd1):\n",
        "  iml=list()\n",
        "  if images.endswith(\".jpg\"):\n",
        "    img=images[:-4]\n",
        "    for ima in os.listdir(fd2):\n",
        "      if ima.endswith(\".png\") and ima.startswith(img):\n",
        "        iml.append(ima)\n",
        "    data[images]=iml\n",
        "data \n"
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "UNSHARP "
      ],
      "metadata": {
        "id": "ew82wRMNIdup"
      }
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "Q-_nxsvmSICg"
      },
      "outputs": [],
      "source": [
        "\n",
        "# count=0\n",
        "# trainlen=len(data)*0.8\n",
        "# for imagesA, imagesB in data.items():\n",
        "#   count+=1\n",
        "#   imgA = cv2.imread('/content/drive/MyDrive/Project/CycleGAN/Dataset/realistic_car/'+imagesA, 1)\n",
        "#   image=imagesB[0]\n",
        "#   imgB = cv2.imread('/content/drive/MyDrive/Project/CycleGAN/Dataset/sketch_car/'+image, 1)\n",
        "#   imgB= unsharp(imgB)\n",
        "#   if count>trainlen:\n",
        "    \n",
        "    \n",
        "#     os.chdir('/content/drive/MyDrive/Project/CycleGAN/Dataset/Test_Train_Split/C_Unsharp_Real_Dataset/testA/')\n",
        "#     filename=image\n",
        "#     cv2.imwrite(filename, imgA)\n",
        "    \n",
        "    \n",
        "#     os.chdir('/content/drive/MyDrive/Project/CycleGAN/Dataset/Test_Train_Split/C_Unsharp_Real_Dataset/testB/')\n",
        "#     filename=image\n",
        "#     cv2.imwrite(filename, imgB)\n",
        "  \n",
        "#   else:\n",
        "#     os.chdir('/content/drive/MyDrive/Project/CycleGAN/Dataset/Test_Train_Split/C_Unsharp_Real_Dataset/trainA/')\n",
        "#     filename=image\n",
        "#     cv2.imwrite(filename, imgA)\n",
        "    \n",
        "    \n",
        "#     os.chdir('/content/drive/MyDrive/Project/CycleGAN/Dataset/Test_Train_Split/C_Unsharp_Real_Dataset/trainB/')\n",
        "#     filename=image\n",
        "#     cv2.imwrite(filename, imgB)\n",
        "\n",
        "\n"
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "LAPLACIAN SHARPENING"
      ],
      "metadata": {
        "id": "xbJ4QK0tIYuW"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "# count=0\n",
        "# trainlen=len(data)*0.8\n",
        "# for imagesA, imagesB in data.items():\n",
        "#   count+=1\n",
        "#   imgA = cv2.imread('/content/drive/MyDrive/realistic_car/'+imagesA, 1)\n",
        "#   image=imagesB[0]\n",
        "#   imgB = cv2.imread('/content/drive/MyDrive/sketch_car/'+image, 1)\n",
        "#   imgB= lapfil(imgB)\n",
        "#   if count>trainlen:\n",
        "    \n",
        "    \n",
        "#     os.chdir('/content/drive/MyDrive/Cycle-Laplacian-Data/testA/')\n",
        "#     filename=image\n",
        "#     cv2.imwrite(filename, imgA)\n",
        "    \n",
        "    \n",
        "#     os.chdir('/content/drive/MyDrive/Cycle-Laplacian-Data/testB/')\n",
        "#     filename=image\n",
        "#     cv2.imwrite(filename, imgB)\n",
        "  \n",
        "#   else:\n",
        "#     os.chdir('/content/drive/MyDrive/Cycle-Laplacian-Data/trainA/')\n",
        "#     filename=image\n",
        "#     cv2.imwrite(filename, imgA)\n",
        "    \n",
        "    \n",
        "#     os.chdir('/content/drive/MyDrive/Cycle-Laplacian-Data/trainB/')\n",
        "#     filename=image\n",
        "#     cv2.imwrite(filename, imgB)\n",
        "\n",
        "\n"
      ],
      "metadata": {
        "id": "LiwK7Kn7IV1R"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "source": [
        "HIGHBOOST"
      ],
      "metadata": {
        "id": "lwpaQJgYIh3g"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "# count=0\n",
        "# trainlen=len(data)*0.8\n",
        "# for imagesA, imagesB in data.items():\n",
        "#   count+=1\n",
        "#   imgA = cv2.imread('/content/drive/MyDrive/Project/CycleGAN/Dataset/realistic_car/'+imagesA, 1)\n",
        "#   image=imagesB[0]\n",
        "#   imgB = cv2.imread('/content/drive/MyDrive/Project/CycleGAN/Dataset/sketch_car/'+image, 1)\n",
        "#   imgB= highboost(imgB)\n",
        "#   if count>trainlen:\n",
        "    \n",
        "    \n",
        "#     os.chdir('/content/drive/MyDrive/Project/CycleGAN/Dataset/Test_Train_Split/C_Highboost_Real_Dataset/testA/')\n",
        "#     filename=image\n",
        "#     cv2.imwrite(filename, imgA)\n",
        "    \n",
        "    \n",
        "#     os.chdir('/content/drive/MyDrive/Project/CycleGAN/Dataset/Test_Train_Split/C_Highboost_Real_Dataset/testB/')\n",
        "#     filename=image\n",
        "#     cv2.imwrite(filename, imgB)\n",
        "  \n",
        "#   else:\n",
        "#     os.chdir('/content/drive/MyDrive/Project/CycleGAN/Dataset/Test_Train_Split/C_Highboost_Real_Dataset/trainA/')\n",
        "#     filename=image\n",
        "#     cv2.imwrite(filename, imgA)\n",
        "    \n",
        "    \n",
        "#     os.chdir('/content/drive/MyDrive/Project/CycleGAN/Dataset/Test_Train_Split/C_Highboost_Real_Dataset/trainB/')\n",
        "#     filename=image\n",
        "#     cv2.imwrite(filename, imgB)\n",
        "\n",
        "\n"
      ],
      "metadata": {
        "id": "00h-HUxFIhf0"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "UIAM7356HR0E"
      },
      "outputs": [],
      "source": [
        "from os import listdir\n",
        "from numpy import asarray\n",
        "from numpy import vstack\n",
        "from keras.preprocessing.image import img_to_array\n",
        "from keras.preprocessing.image import load_img\n",
        "from matplotlib import pyplot as plt\n",
        "import numpy as np\n",
        "from sklearn.utils import resample"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "-VaGDkozHa8b"
      },
      "outputs": [],
      "source": [
        "def load_images(path, size=(256,256)):\n",
        "\tdata_list = list()\n",
        "\t# enumerate filenames in directory, assume all are images\n",
        "\tfor filename in listdir(path):\n",
        "\n",
        "\t\t# load and resize the image\n",
        "\t\tpixels = load_img(path + filename, target_size=size)\n",
        "\t\n",
        "\t\t# convert to numpy array\n",
        "\t\tpixels = img_to_array(pixels)\n",
        "\t\n",
        "\t\t# store\n",
        "\t\tdata_list.append(pixels)\n",
        "\treturn asarray(data_list)"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "XRV6A_deHgoG"
      },
      "outputs": [],
      "source": [
        "path = '/content/drive/MyDrive/WarpedDreams/CycleGAN/Dataset/Test_Train_Split/C_Unsharp_Real_Dataset/'\n",
        "# path = '/content/drive/MyDrive/Project/CycleGAN/Dataset/Test_Train_Split/C_Laplacian_Real_Dataset/'\n",
        "# path= '/content/drive/MyDrive/Project/CycleGAN/Dataset/Test_Train_Split/C_Highboost_Real_Dataset/'"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "BMXCGQXHHrx6",
        "outputId": "77876087-4db7-4774-8167-93147ae57860"
      },
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Loaded dataA:  (509, 256, 256, 3)\n"
          ]
        }
      ],
      "source": [
        "dataA_all = load_images(path + 'trainA/')\n",
        "print('Loaded dataA: ', dataA_all.shape)"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "aqN4S2UDHyNO"
      },
      "outputs": [],
      "source": [
        "#subsampling\n",
        "dataA = resample(dataA_all, \n",
        "                 replace=False,     \n",
        "                 n_samples=80,    \n",
        "                 random_state=42) "
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "K3CAI9R_IA0T",
        "outputId": "685d7e0b-be93-4fc9-8712-12eef944032c"
      },
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Loaded dataB:  (509, 256, 256, 3)\n"
          ]
        }
      ],
      "source": [
        "dataB_all = load_images(path + 'trainB/')\n",
        "print('Loaded dataB: ', dataB_all.shape)"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "vUPL6wQsIGCQ"
      },
      "outputs": [],
      "source": [
        "dataB = resample(dataB_all, \n",
        "                 replace=False,     \n",
        "                 n_samples=80,    \n",
        "                 random_state=42) "
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 248
        },
        "id": "BEk_wt7NILuM",
        "outputId": "6e43b07a-634f-44e5-801d-ce89eec1e08f"
      },
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 432x288 with 6 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {
            "needs_background": "light"
          }
        }
      ],
      "source": [
        "n_samples = 3\n",
        "for i in range(n_samples):\n",
        "\tplt.subplot(2, n_samples, 1 + i)\n",
        "\tplt.axis('off')\n",
        "\tplt.imshow(dataA[i].astype('uint8'))\n",
        " \n",
        "# plot target image\n",
        "for i in range(n_samples):\n",
        "\tplt.subplot(2, n_samples, 1 + n_samples + i)\n",
        "\tplt.axis('off')\n",
        "\tplt.imshow(dataB[i].astype('uint8'))\n",
        "plt.show()"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "-UnIzBlSIQ2z",
        "outputId": "0d445336-2631-4208-88b7-c3d0091a7254"
      },
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Loaded (80, 256, 256, 3) (80, 256, 256, 3)\n"
          ]
        }
      ],
      "source": [
        "data = [dataA, dataB]\n",
        "print('Loaded', data[0].shape, data[1].shape)"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "prV89GMPIVXB"
      },
      "outputs": [],
      "source": [
        "def preprocess_data(data):\n",
        "\t# load compressed arrays\n",
        "\t# unpack arrays\n",
        "\tX1, X2 = data[0], data[1]\n",
        "\t\n",
        "\t# scale from [0,255] to [-1,1]\n",
        "\tX1 = (X1 - 127.5) / 127.5\n",
        "\tX2 = (X2 - 127.5) / 127.5\n",
        "\treturn [X1, X2]"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "4W7rzeUuIZLe"
      },
      "outputs": [],
      "source": [
        "dataset = preprocess_data(data)"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "J4J30uU1Id75",
        "outputId": "03ec3e2d-92ac-479e-8a81-da7316e895fd"
      },
      "outputs": [
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "/usr/local/lib/python3.7/dist-packages/keras/optimizer_v2/adam.py:105: UserWarning: The `lr` argument is deprecated, use `learning_rate` instead.\n",
            "  super(Adam, self).__init__(name, **kwargs)\n"
          ]
        }
      ],
      "source": [
        "image_shape = dataset[0].shape[1:]\n",
        "# generator: A -> B\n",
        "g_model_AtoB = define_generator(image_shape)\n",
        "\n",
        "# generator: B -> A\n",
        "g_model_BtoA = define_generator(image_shape)\n",
        "\n",
        "# discriminator: A -> [real/fake]\n",
        "d_model_A = define_discriminator(image_shape)\n",
        "\n",
        "# discriminator: B -> [real/fake]\n",
        "d_model_B = define_discriminator(image_shape)\n",
        "\n",
        "# composite: A -> B -> [real/fake, A]\n",
        "c_model_AtoB = define_composite_model(g_model_AtoB, d_model_B, g_model_BtoA, image_shape)\n",
        "\n",
        "# composite: B -> A -> [real/fake, B]\n",
        "c_model_BtoA = define_composite_model(g_model_BtoA, d_model_A, g_model_AtoB, image_shape)\n"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "QNIP_IH3IqDW",
        "outputId": "15a4b7c6-00a1-4f9d-fea6-40f91181758b"
      },
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "WARNING:tensorflow:5 out of the last 24001 calls to <function Model.make_train_function.<locals>.train_function at 0x7f97e0604e60> triggered tf.function retracing. Tracing is expensive and the excessive number of tracings could be due to (1) creating @tf.function repeatedly in a loop, (2) passing tensors with different shapes, (3) passing Python objects instead of tensors. For (1), please define your @tf.function outside of the loop. For (2), @tf.function has experimental_relax_shapes=True option that relaxes argument shapes that can avoid unnecessary retracing. For (3), please refer to https://www.tensorflow.org/guide/function#controlling_retracing and https://www.tensorflow.org/api_docs/python/tf/function for  more details.\n",
            "Iteration>1, dA[1.822,2.258] dB[1.262,1.875] g[27.202,29.487]\n",
            "Iteration>2, dA[5.020,6.564] dB[1.059,6.491] g[28.168,32.842]\n",
            "Iteration>3, dA[9.754,6.408] dB[1.251,8.396] g[39.909,26.177]\n",
            "Iteration>4, dA[1.267,2.825] dB[1.906,8.586] g[36.851,24.078]\n",
            "Iteration>5, dA[1.273,1.439] dB[1.944,4.362] g[26.881,22.033]\n",
            "Iteration>6, dA[0.685,0.682] dB[0.847,5.020] g[24.198,20.513]\n",
            "Iteration>7, dA[0.326,0.437] dB[0.773,1.183] g[22.088,20.914]\n",
            "Iteration>8, dA[0.431,0.433] dB[0.701,0.898] g[20.413,18.898]\n",
            "Iteration>9, dA[0.312,0.339] dB[0.484,0.551] g[19.466,18.538]\n",
            "Iteration>10, dA[0.325,0.385] dB[0.616,0.590] g[21.621,21.070]\n",
            "Iteration>11, dA[0.437,0.348] dB[0.738,0.550] g[19.207,17.684]\n",
            "Iteration>12, dA[0.337,0.259] dB[0.535,0.497] g[18.260,16.641]\n",
            "Iteration>13, dA[0.248,0.254] dB[0.450,0.616] g[18.739,16.677]\n",
            "Iteration>14, dA[0.346,0.512] dB[0.375,0.650] g[18.555,16.822]\n",
            "Iteration>15, dA[0.364,0.295] dB[0.478,0.433] g[20.204,19.664]\n",
            "Iteration>16, dA[0.422,0.299] dB[0.424,0.542] g[18.232,16.687]\n",
            "Iteration>17, dA[0.333,0.262] dB[0.504,0.332] g[17.416,15.589]\n",
            "Iteration>18, dA[0.226,0.261] dB[0.438,0.341] g[17.694,15.960]\n",
            "Iteration>19, dA[0.340,0.283] dB[0.409,0.381] g[16.387,14.741]\n",
            "Iteration>20, dA[0.260,0.161] dB[0.308,0.297] g[19.582,19.052]\n",
            "Iteration>21, dA[0.206,0.193] dB[0.462,0.341] g[18.367,16.713]\n",
            "Iteration>22, dA[0.171,0.340] dB[0.332,0.342] g[19.724,19.150]\n",
            "Iteration>23, dA[0.179,0.203] dB[0.259,0.321] g[18.141,17.004]\n",
            "Iteration>24, dA[0.206,0.190] dB[0.293,0.290] g[17.936,16.511]\n",
            "Iteration>25, dA[0.422,0.220] dB[0.440,0.300] g[16.951,15.299]\n",
            "Iteration>26, dA[0.440,0.411] dB[0.354,0.250] g[16.046,13.859]\n",
            "Iteration>27, dA[0.357,0.413] dB[0.219,0.270] g[16.731,14.957]\n",
            "Iteration>28, dA[0.260,0.189] dB[0.340,0.216] g[16.407,14.295]\n",
            "Iteration>29, dA[0.221,0.222] dB[0.400,0.195] g[16.194,14.225]\n",
            "Iteration>30, dA[0.223,0.158] dB[0.361,0.300] g[16.189,14.087]\n",
            "Iteration>31, dA[0.178,0.201] dB[0.320,0.403] g[17.513,16.120]\n",
            "Iteration>32, dA[0.174,0.177] dB[0.352,0.343] g[19.219,18.564]\n",
            "Iteration>33, dA[0.275,0.152] dB[0.165,0.248] g[16.015,14.800]\n",
            "Iteration>34, dA[0.127,0.120] dB[0.170,0.130] g[17.706,16.516]\n",
            "Iteration>35, dA[0.132,0.203] dB[0.237,0.147] g[15.528,13.044]\n",
            "Iteration>36, dA[0.164,0.260] dB[0.287,0.330] g[19.429,18.801]\n",
            "Iteration>37, dA[0.246,0.146] dB[0.174,0.258] g[16.227,14.544]\n",
            "Iteration>38, dA[0.196,0.151] dB[0.201,0.295] g[16.114,13.959]\n",
            "Iteration>39, dA[0.205,0.196] dB[0.246,0.176] g[17.974,16.763]\n",
            "Iteration>40, dA[0.171,0.184] dB[0.191,0.087] g[17.690,16.357]\n",
            "Iteration>41, dA[0.160,0.199] dB[0.150,0.118] g[16.900,15.561]\n",
            "Iteration>42, dA[0.143,0.117] dB[0.213,0.121] g[17.223,16.145]\n",
            "Iteration>43, dA[0.249,0.162] dB[0.066,0.132] g[15.893,14.027]\n",
            "Iteration>44, dA[0.148,0.095] dB[0.107,0.091] g[17.267,16.030]\n",
            "Iteration>45, dA[0.146,0.109] dB[0.174,0.073] g[17.401,16.139]\n",
            "Iteration>46, dA[0.135,0.152] dB[0.118,0.216] g[16.195,14.062]\n",
            "Iteration>47, dA[0.164,0.144] dB[0.068,0.291] g[18.483,17.102]\n",
            "Iteration>48, dA[0.196,0.185] dB[0.092,0.511] g[17.454,15.933]\n",
            "Iteration>49, dA[0.129,0.093] dB[0.085,0.312] g[18.535,17.939]\n",
            "Iteration>50, dA[0.219,0.130] dB[0.196,0.101] g[16.446,14.413]\n",
            "Iteration>51, dA[0.101,0.070] dB[0.151,0.247] g[17.214,15.630]\n",
            "Iteration>52, dA[0.073,0.118] dB[0.369,0.087] g[18.960,17.792]\n",
            "Iteration>53, dA[0.115,0.063] dB[0.198,0.069] g[17.242,15.686]\n",
            "Iteration>54, dA[0.127,0.127] dB[0.098,0.128] g[15.496,13.393]\n",
            "Iteration>55, dA[0.059,0.154] dB[0.159,0.080] g[17.801,16.401]\n",
            "Iteration>56, dA[0.198,0.154] dB[0.160,0.052] g[17.109,15.769]\n",
            "Iteration>57, dA[0.073,0.096] dB[0.165,0.075] g[18.874,18.338]\n",
            "Iteration>58, dA[0.178,0.133] dB[0.149,0.172] g[14.630,12.656]\n",
            "Iteration>59, dA[0.127,0.149] dB[0.402,0.061] g[18.962,18.578]\n",
            "Iteration>60, dA[0.111,0.104] dB[0.120,0.090] g[17.373,16.291]\n",
            "Iteration>61, dA[0.096,0.063] dB[0.126,0.068] g[18.410,18.045]\n",
            "Iteration>62, dA[0.059,0.096] dB[0.121,0.118] g[18.669,18.390]\n",
            "Iteration>63, dA[0.156,0.443] dB[0.041,0.120] g[16.588,15.217]\n",
            "Iteration>64, dA[0.280,0.148] dB[0.091,0.056] g[17.119,16.341]\n",
            "Iteration>65, dA[0.191,0.338] dB[0.064,0.066] g[15.538,13.269]\n",
            "Iteration>66, dA[0.085,0.347] dB[0.066,0.028] g[17.808,16.319]\n",
            "Iteration>67, dA[0.312,0.141] dB[0.251,0.078] g[16.703,15.518]\n",
            "Iteration>68, dA[0.089,0.135] dB[0.176,0.074] g[16.789,15.526]\n",
            "Iteration>69, dA[0.096,0.146] dB[0.064,0.103] g[17.394,16.053]\n",
            "Iteration>70, dA[0.120,0.097] dB[0.079,0.064] g[14.871,13.166]\n",
            "Iteration>71, dA[0.083,0.044] dB[0.101,0.085] g[15.788,13.368]\n",
            "Iteration>72, dA[0.041,0.088] dB[0.167,0.047] g[17.623,16.135]\n",
            "Iteration>73, dA[0.094,0.059] dB[0.152,0.101] g[17.028,15.547]\n",
            "Iteration>74, dA[0.148,0.045] dB[0.103,0.039] g[16.014,14.072]\n",
            "Iteration>75, dA[0.099,0.191] dB[0.080,0.101] g[16.917,15.260]\n",
            "Iteration>76, dA[0.233,0.187] dB[0.058,0.058] g[17.668,16.991]\n",
            "Iteration>77, dA[0.143,0.082] dB[0.034,0.052] g[14.821,12.296]\n",
            "Iteration>78, dA[0.104,0.243] dB[0.035,0.037] g[17.626,16.249]\n",
            "Iteration>79, dA[0.109,0.212] dB[0.140,0.010] g[17.170,15.984]\n",
            "Iteration>80, dA[0.198,0.077] dB[0.026,0.145] g[14.370,12.141]\n",
            "Iteration>81, dA[0.068,0.210] dB[0.196,0.101] g[17.004,15.853]\n",
            "Iteration>82, dA[0.147,0.074] dB[0.169,0.077] g[15.722,13.668]\n",
            "Iteration>83, dA[0.088,0.071] dB[0.106,0.020] g[17.373,16.279]\n",
            "Iteration>84, dA[0.076,0.035] dB[0.040,0.078] g[17.320,16.121]\n",
            "Iteration>85, dA[0.167,0.088] dB[0.052,0.042] g[15.956,14.485]\n",
            "Iteration>86, dA[0.057,0.167] dB[0.097,0.049] g[18.241,16.822]\n",
            "Iteration>87, dA[0.093,0.055] dB[0.036,0.043] g[15.635,13.836]\n",
            "Iteration>88, dA[0.056,0.166] dB[0.054,0.042] g[17.649,16.237]\n",
            "Iteration>89, dA[0.159,0.047] dB[0.043,0.048] g[16.729,15.126]\n",
            "Iteration>90, dA[0.048,0.076] dB[0.033,0.033] g[17.371,15.697]\n",
            "Iteration>91, dA[0.043,0.101] dB[0.083,0.033] g[16.938,15.124]\n",
            "Iteration>92, dA[0.114,0.149] dB[0.027,0.038] g[15.948,14.066]\n",
            "Iteration>93, dA[0.114,0.030] dB[0.039,0.104] g[14.491,11.703]\n",
            "Iteration>94, dA[0.060,0.330] dB[0.046,0.034] g[15.523,13.510]\n",
            "Iteration>95, dA[0.321,0.043] dB[0.101,0.071] g[16.825,15.378]\n",
            "Iteration>96, dA[0.039,0.262] dB[0.018,0.024] g[16.488,14.306]\n",
            "Iteration>97, dA[0.118,0.202] dB[0.073,0.086] g[14.982,12.641]\n",
            "Iteration>98, dA[0.136,0.024] dB[0.021,0.021] g[18.303,18.043]\n",
            "Iteration>99, dA[0.164,0.191] dB[0.065,0.045] g[14.826,12.285]\n",
            "Iteration>100, dA[0.076,0.097] dB[0.100,0.066] g[16.380,15.056]\n",
            "Iteration>101, dA[0.118,0.031] dB[0.052,0.025] g[18.786,18.629]\n",
            "Iteration>102, dA[0.233,0.122] dB[0.025,0.038] g[16.019,14.211]\n",
            "Iteration>103, dA[0.052,0.047] dB[0.058,0.047] g[15.291,13.019]\n",
            "Iteration>104, dA[0.052,0.231] dB[0.061,0.019] g[16.915,15.518]\n",
            "Iteration>105, dA[0.078,0.047] dB[0.039,0.064] g[18.769,18.410]\n",
            "Iteration>106, dA[0.074,0.132] dB[0.037,0.038] g[18.341,17.162]\n",
            "Iteration>107, dA[0.087,0.042] dB[0.159,0.058] g[15.494,13.280]\n",
            "Iteration>108, dA[0.034,0.061] dB[0.110,0.035] g[16.398,15.066]\n",
            "Iteration>109, dA[0.064,0.064] dB[0.040,0.028] g[16.638,15.475]\n",
            "Iteration>110, dA[0.190,0.099] dB[0.030,0.092] g[14.700,12.499]\n",
            "Iteration>111, dA[0.150,0.421] dB[0.154,0.096] g[14.125,11.669]\n",
            "Iteration>112, dA[0.075,0.353] dB[0.039,0.065] g[17.338,16.234]\n",
            "Iteration>113, dA[0.289,0.119] dB[0.094,0.034] g[16.846,15.923]\n",
            "Iteration>114, dA[0.223,0.139] dB[0.065,0.063] g[15.195,13.698]\n",
            "Iteration>115, dA[0.097,0.083] dB[0.059,0.027] g[16.957,15.864]\n",
            "Iteration>116, dA[0.146,0.235] dB[0.028,0.028] g[15.168,12.824]\n",
            "Iteration>117, dA[0.112,0.027] dB[0.028,0.013] g[18.779,18.459]\n",
            "Iteration>118, dA[0.127,0.129] dB[0.060,0.026] g[16.895,15.789]\n",
            "Iteration>119, dA[0.071,0.057] dB[0.100,0.023] g[17.421,16.005]\n",
            "Iteration>120, dA[0.040,0.181] dB[0.086,0.033] g[18.581,17.638]\n",
            "Iteration>121, dA[0.125,0.037] dB[0.054,0.035] g[16.701,15.877]\n",
            "Iteration>122, dA[0.208,0.091] dB[0.041,0.103] g[15.443,13.787]\n",
            "Iteration>123, dA[0.030,0.173] dB[0.048,0.055] g[17.331,15.503]\n",
            "Iteration>124, dA[0.132,0.086] dB[0.082,0.127] g[15.731,14.310]\n",
            "Iteration>125, dA[0.080,0.024] dB[0.066,0.024] g[17.118,15.856]\n",
            "Iteration>126, dA[0.036,0.092] dB[0.031,0.036] g[19.728,19.377]\n",
            "Iteration>127, dA[0.260,0.192] dB[0.069,0.023] g[14.901,12.756]\n",
            "Iteration>128, dA[0.065,0.093] dB[0.081,0.121] g[16.217,14.730]\n",
            "Iteration>129, dA[0.079,0.099] dB[0.089,0.028] g[16.168,14.930]\n",
            "Iteration>130, dA[0.103,0.103] dB[0.047,0.127] g[15.956,14.703]\n",
            "Iteration>131, dA[0.031,0.092] dB[0.025,0.041] g[16.493,15.042]\n",
            "Iteration>132, dA[0.021,0.035] dB[0.103,0.032] g[18.326,18.047]\n",
            "Iteration>133, dA[0.034,0.032] dB[0.013,0.013] g[17.230,16.491]\n",
            "Iteration>134, dA[0.095,0.196] dB[0.031,0.007] g[13.874,11.668]\n",
            "Iteration>135, dA[0.058,0.051] dB[0.027,0.015] g[16.462,15.121]\n",
            "Iteration>136, dA[0.051,0.174] dB[0.073,0.041] g[17.872,17.094]\n",
            "Iteration>137, dA[0.280,0.130] dB[0.121,0.026] g[15.628,14.832]\n",
            "Iteration>138, dA[0.091,0.231] dB[0.077,0.035] g[16.184,15.352]\n",
            "Iteration>139, dA[0.134,0.046] dB[0.050,0.239] g[16.484,16.007]\n",
            "Iteration>140, dA[0.148,0.289] dB[0.142,0.069] g[14.512,11.804]\n",
            "Iteration>141, dA[0.092,0.175] dB[0.143,0.048] g[15.034,12.901]\n",
            "Iteration>142, dA[0.204,0.041] dB[0.051,0.028] g[15.658,15.013]\n",
            "Iteration>143, dA[0.128,0.109] dB[0.022,0.848] g[16.752,15.904]\n",
            "Iteration>144, dA[0.034,0.021] dB[0.149,0.119] g[15.611,13.525]\n",
            "Iteration>145, dA[0.064,0.195] dB[0.092,0.048] g[14.164,11.958]\n",
            "Iteration>146, dA[0.066,0.093] dB[0.033,0.462] g[18.827,17.945]\n",
            "Iteration>147, dA[0.067,0.167] dB[0.141,0.065] g[19.428,19.884]\n",
            "Iteration>148, dA[0.029,0.091] dB[0.122,0.159] g[18.329,17.744]\n",
            "Iteration>149, dA[0.037,0.058] dB[0.103,0.046] g[16.870,15.646]\n",
            "Iteration>150, dA[0.093,0.067] dB[0.095,0.022] g[15.895,13.987]\n",
            "Iteration>151, dA[0.023,0.114] dB[0.078,0.013] g[18.057,16.984]\n",
            "Iteration>152, dA[0.290,0.099] dB[0.046,0.037] g[14.928,13.618]\n",
            "Iteration>153, dA[0.020,0.172] dB[0.064,0.019] g[17.472,15.758]\n",
            "Iteration>154, dA[0.101,0.042] dB[0.040,0.016] g[17.370,16.585]\n",
            "Iteration>155, dA[0.124,0.120] dB[0.018,0.041] g[14.043,11.967]\n",
            "Iteration>156, dA[0.024,0.054] dB[0.034,0.023] g[15.363,12.990]\n",
            "Iteration>157, dA[0.028,0.022] dB[0.039,0.132] g[17.062,16.003]\n",
            "Iteration>158, dA[0.066,0.060] dB[0.055,0.040] g[15.519,13.335]\n",
            "Iteration>159, dA[0.039,0.076] dB[0.106,0.059] g[16.669,14.541]\n",
            "Iteration>160, dA[0.050,0.045] dB[0.032,0.057] g[17.049,16.057]\n",
            "Iteration>161, dA[0.021,0.116] dB[0.026,0.014] g[18.297,17.468]\n",
            "Iteration>162, dA[0.152,0.022] dB[0.031,0.015] g[16.099,14.219]\n",
            "Iteration>163, dA[0.053,0.342] dB[0.051,0.013] g[16.603,14.938]\n",
            "Iteration>164, dA[0.137,0.405] dB[0.015,0.103] g[14.880,13.914]\n",
            "Iteration>165, dA[0.268,0.075] dB[0.058,0.033] g[18.712,18.262]\n",
            "Iteration>166, dA[0.189,0.072] dB[0.033,0.013] g[16.424,14.662]\n",
            "Iteration>167, dA[0.081,0.065] dB[0.026,0.026] g[17.733,17.252]\n",
            "Iteration>168, dA[0.201,0.212] dB[0.036,0.022] g[14.870,12.961]\n",
            "Iteration>169, dA[0.038,0.024] dB[0.036,0.025] g[15.595,14.126]\n",
            "Iteration>170, dA[0.115,0.091] dB[0.043,0.090] g[15.186,13.004]\n",
            "Iteration>171, dA[0.027,0.167] dB[0.056,0.011] g[17.659,16.273]\n",
            "Iteration>172, dA[0.036,0.039] dB[0.022,0.007] g[16.840,15.262]\n",
            "Iteration>173, dA[0.117,0.026] dB[0.069,0.024] g[15.481,14.315]\n",
            "Iteration>174, dA[0.036,0.227] dB[0.048,0.035] g[18.513,17.674]\n",
            "Iteration>175, dA[0.082,0.133] dB[0.019,0.013] g[17.559,16.529]\n",
            "Iteration>176, dA[0.296,0.295] dB[0.019,0.010] g[15.455,13.886]\n",
            "Iteration>177, dA[0.162,0.254] dB[0.023,0.008] g[17.091,15.810]\n",
            "Iteration>178, dA[0.125,0.086] dB[0.012,0.010] g[18.169,17.922]\n",
            "Iteration>179, dA[0.020,0.023] dB[0.022,0.006] g[18.594,17.941]\n",
            "Iteration>180, dA[0.032,0.044] dB[0.023,0.020] g[17.710,17.276]\n",
            "Iteration>181, dA[0.187,0.099] dB[0.034,0.014] g[15.174,13.448]\n",
            "Iteration>182, dA[0.051,0.151] dB[0.023,0.015] g[17.613,16.136]\n",
            "Iteration>183, dA[0.056,0.056] dB[0.048,0.024] g[18.365,17.953]\n",
            "Iteration>184, dA[0.107,0.056] dB[0.026,0.013] g[17.042,15.933]\n",
            "Iteration>185, dA[0.027,0.019] dB[0.017,0.023] g[17.264,16.259]\n",
            "Iteration>186, dA[0.020,0.015] dB[0.033,0.019] g[16.462,15.299]\n",
            "Iteration>187, dA[0.218,0.158] dB[0.039,0.027] g[14.418,12.469]\n",
            "Iteration>188, dA[0.042,0.063] dB[0.024,0.027] g[14.773,12.077]\n",
            "Iteration>189, dA[0.034,0.021] dB[0.022,0.009] g[16.344,14.895]\n",
            "Iteration>190, dA[0.036,0.069] dB[0.012,0.014] g[17.451,16.782]\n",
            "Iteration>191, dA[0.049,0.034] dB[0.008,0.020] g[16.535,15.409]\n",
            "Iteration>192, dA[0.031,0.016] dB[0.015,0.009] g[17.138,15.578]\n",
            "Iteration>193, dA[0.040,0.021] dB[0.019,0.006] g[16.710,14.708]\n",
            "Iteration>194, dA[0.023,0.023] dB[0.007,0.017] g[16.692,15.139]\n",
            "Iteration>195, dA[0.012,0.075] dB[0.056,0.029] g[18.000,17.429]\n",
            "Iteration>196, dA[0.022,0.043] dB[0.043,0.011] g[18.571,18.066]\n",
            "Iteration>197, dA[0.049,0.028] dB[0.051,0.022] g[17.830,17.238]\n",
            "Iteration>198, dA[0.021,0.094] dB[0.043,0.049] g[16.235,14.998]\n",
            "Iteration>199, dA[0.221,0.193] dB[0.017,0.041] g[15.252,13.246]\n",
            "Iteration>200, dA[0.030,0.067] dB[0.036,0.005] g[17.976,17.334]\n",
            "Iteration>201, dA[0.279,0.131] dB[0.033,0.035] g[14.462,13.031]\n",
            "Iteration>202, dA[0.048,0.064] dB[0.016,0.017] g[13.984,11.185]\n",
            "Iteration>203, dA[0.037,0.025] dB[0.030,0.009] g[15.754,13.702]\n",
            "Iteration>204, dA[0.038,0.109] dB[0.012,0.012] g[16.832,15.012]\n",
            "Iteration>205, dA[0.096,0.157] dB[0.029,0.027] g[16.044,15.000]\n",
            "Iteration>206, dA[0.066,0.231] dB[0.016,0.022] g[14.868,13.159]\n",
            "Iteration>207, dA[0.076,0.026] dB[0.020,0.005] g[14.960,12.872]\n",
            "Iteration>208, dA[0.049,0.213] dB[0.027,0.024] g[16.632,15.521]\n",
            "Iteration>209, dA[0.179,0.294] dB[0.025,0.016] g[14.943,13.907]\n",
            "Iteration>210, dA[0.120,0.232] dB[0.062,0.041] g[14.874,12.837]\n",
            "Iteration>211, dA[0.141,0.170] dB[0.026,0.041] g[15.260,13.559]\n",
            "Iteration>212, dA[0.165,0.033] dB[0.031,0.029] g[13.579,11.866]\n",
            "Iteration>213, dA[0.031,0.123] dB[0.033,0.018] g[16.379,14.580]\n",
            "Iteration>214, dA[0.311,0.128] dB[0.044,0.043] g[13.508,10.907]\n",
            "Iteration>215, dA[0.022,0.045] dB[0.010,0.029] g[16.588,14.781]\n",
            "Iteration>216, dA[0.039,0.028] dB[0.025,0.012] g[18.316,17.344]\n",
            "Iteration>217, dA[0.030,0.021] dB[0.017,0.011] g[18.360,17.339]\n",
            "Iteration>218, dA[0.038,0.300] dB[0.016,0.019] g[15.589,13.432]\n",
            "Iteration>219, dA[0.234,0.012] dB[0.011,0.019] g[14.544,12.901]\n",
            "Iteration>220, dA[0.018,0.261] dB[0.025,0.010] g[17.629,16.443]\n",
            "Iteration>221, dA[0.093,0.080] dB[0.014,0.016] g[16.576,15.663]\n",
            "Iteration>222, dA[0.182,0.194] dB[0.010,0.022] g[14.970,13.365]\n",
            "Iteration>223, dA[0.025,0.069] dB[0.030,0.016] g[16.129,14.235]\n",
            "Iteration>224, dA[0.093,0.082] dB[0.020,0.010] g[16.481,14.883]\n",
            "Iteration>225, dA[0.041,0.037] dB[0.035,0.038] g[16.403,15.035]\n",
            "Iteration>226, dA[0.052,0.149] dB[0.027,0.048] g[14.738,12.716]\n",
            "Iteration>227, dA[0.017,0.051] dB[0.029,0.044] g[16.056,15.146]\n",
            "Iteration>228, dA[0.115,0.107] dB[0.042,0.041] g[14.610,12.729]\n",
            "Iteration>229, dA[0.092,0.329] dB[0.025,0.024] g[15.862,13.205]\n",
            "Iteration>230, dA[0.226,0.104] dB[0.041,0.022] g[17.302,16.678]\n",
            "Iteration>231, dA[0.083,0.018] dB[0.016,0.029] g[17.577,17.475]\n",
            "Iteration>232, dA[0.015,0.068] dB[0.008,0.022] g[17.945,17.161]\n",
            "Iteration>233, dA[0.144,0.039] dB[0.016,0.012] g[15.591,14.013]\n",
            "Iteration>234, dA[0.060,0.077] dB[0.016,0.023] g[15.934,14.154]\n",
            "Iteration>235, dA[0.035,0.012] dB[0.024,0.032] g[15.803,14.197]\n",
            "Iteration>236, dA[0.050,0.155] dB[0.024,0.034] g[13.821,11.778]\n",
            "Iteration>237, dA[0.016,0.031] dB[0.021,0.029] g[17.605,16.200]\n",
            "Iteration>238, dA[0.074,0.087] dB[0.026,0.093] g[14.709,13.234]\n",
            "Iteration>239, dA[0.016,0.012] dB[0.026,0.026] g[16.255,14.470]\n",
            "Iteration>240, dA[0.040,0.040] dB[0.091,0.113] g[15.786,13.859]\n",
            "Iteration>241, dA[0.026,0.055] dB[0.028,0.045] g[16.069,13.983]\n",
            "Iteration>242, dA[0.121,0.012] dB[0.081,0.094] g[17.132,15.894]\n",
            "Iteration>243, dA[0.031,0.018] dB[0.047,0.016] g[15.439,12.945]\n",
            "Iteration>244, dA[0.013,0.128] dB[0.053,0.061] g[16.475,14.378]\n",
            "Iteration>245, dA[0.087,0.091] dB[0.018,0.015] g[16.449,15.259]\n",
            "Iteration>246, dA[0.073,0.052] dB[0.043,0.032] g[16.112,14.627]\n",
            "Iteration>247, dA[0.044,0.035] dB[0.033,0.026] g[16.228,14.699]\n",
            "Iteration>248, dA[0.393,0.069] dB[0.028,0.010] g[13.977,12.911]\n",
            "Iteration>249, dA[0.134,0.187] dB[0.028,0.036] g[16.250,14.792]\n",
            "Iteration>250, dA[0.075,0.080] dB[0.025,0.021] g[18.307,18.080]\n",
            "Iteration>251, dA[0.122,0.264] dB[0.040,0.057] g[15.880,14.400]\n",
            "Iteration>252, dA[0.057,0.030] dB[0.008,0.060] g[16.315,15.247]\n",
            "Iteration>253, dA[0.097,0.037] dB[0.033,0.022] g[16.090,14.316]\n",
            "Iteration>254, dA[0.020,0.146] dB[0.071,0.061] g[16.201,14.295]\n",
            "Iteration>255, dA[0.049,0.039] dB[0.021,0.041] g[17.618,17.482]\n",
            "Iteration>256, dA[0.019,0.015] dB[0.057,0.062] g[18.247,18.092]\n",
            "Iteration>257, dA[0.027,0.055] dB[0.038,0.715] g[15.771,14.857]\n",
            "Iteration>258, dA[0.085,0.173] dB[0.281,0.093] g[15.935,12.685]\n",
            "Iteration>259, dA[0.035,0.029] dB[0.036,0.353] g[18.584,17.540]\n",
            "Iteration>260, dA[0.125,0.050] dB[0.093,0.040] g[16.483,16.058]\n",
            "Iteration>261, dA[0.051,0.012] dB[0.112,0.043] g[17.744,16.610]\n",
            "Iteration>262, dA[0.018,0.023] dB[0.122,0.096] g[17.084,15.496]\n",
            "Iteration>263, dA[0.076,0.118] dB[0.031,0.026] g[15.061,13.422]\n",
            "Iteration>264, dA[0.053,0.106] dB[0.025,0.020] g[16.760,15.618]\n",
            "Iteration>265, dA[0.401,0.200] dB[0.020,0.017] g[14.062,11.690]\n",
            "Iteration>266, dA[0.037,0.191] dB[0.017,0.024] g[15.596,12.751]\n",
            "Iteration>267, dA[0.327,0.039] dB[0.031,0.005] g[14.065,12.011]\n",
            "Iteration>268, dA[0.050,0.087] dB[0.044,0.029] g[17.057,15.843]\n",
            "Iteration>269, dA[0.073,0.046] dB[0.012,0.009] g[15.000,12.954]\n",
            "Iteration>270, dA[0.036,0.023] dB[0.024,0.011] g[14.830,12.635]\n",
            "Iteration>271, dA[0.023,0.020] dB[0.014,0.024] g[15.735,13.989]\n",
            "Iteration>272, dA[0.067,0.070] dB[0.026,0.011] g[14.367,12.490]\n",
            "Iteration>273, dA[0.013,0.017] dB[0.012,0.006] g[16.228,14.599]\n",
            "Iteration>274, dA[0.018,0.020] dB[0.047,0.009] g[16.910,15.948]\n",
            "Iteration>275, dA[0.033,0.013] dB[0.020,0.016] g[17.527,17.338]\n",
            "Iteration>276, dA[0.007,0.027] dB[0.022,0.007] g[16.067,14.996]\n",
            "Iteration>277, dA[0.021,0.079] dB[0.024,0.012] g[15.390,13.519]\n",
            "Iteration>278, dA[0.027,0.019] dB[0.020,0.020] g[16.691,15.695]\n",
            "Iteration>279, dA[0.337,0.406] dB[0.010,0.005] g[14.458,12.347]\n",
            "Iteration>280, dA[0.047,0.073] dB[0.022,0.011] g[16.707,15.618]\n",
            "Iteration>281, dA[0.155,0.045] dB[0.028,0.011] g[16.022,15.249]\n",
            "Iteration>282, dA[0.053,0.083] dB[0.022,0.022] g[15.992,14.746]\n",
            "Iteration>283, dA[0.028,0.108] dB[0.023,0.031] g[15.107,13.435]\n",
            "Iteration>284, dA[0.361,0.146] dB[0.032,0.014] g[14.694,13.087]\n",
            "Iteration>285, dA[0.077,0.340] dB[0.018,0.006] g[16.268,14.331]\n",
            "Iteration>286, dA[0.447,0.221] dB[0.017,0.003] g[14.142,12.679]\n",
            "Iteration>287, dA[0.033,0.055] dB[0.020,0.004] g[17.984,17.278]\n",
            "Iteration>288, dA[0.105,0.059] dB[0.015,0.006] g[16.196,14.926]\n",
            "Iteration>289, dA[0.084,0.058] dB[0.025,0.013] g[14.871,13.221]\n",
            "Iteration>290, dA[0.015,0.072] dB[0.020,0.009] g[16.054,14.556]\n",
            "Iteration>291, dA[0.143,0.153] dB[0.019,0.023] g[14.992,13.447]\n",
            "Iteration>292, dA[0.063,0.032] dB[0.018,0.015] g[15.731,14.317]\n",
            "Iteration>293, dA[0.026,0.031] dB[0.030,0.034] g[16.803,16.134]\n",
            "Iteration>294, dA[0.051,0.031] dB[0.009,0.022] g[15.458,13.459]\n",
            "Iteration>295, dA[0.046,0.091] dB[0.020,0.011] g[17.677,16.705]\n",
            "Iteration>296, dA[0.585,0.271] dB[0.020,0.017] g[13.557,12.193]\n",
            "Iteration>297, dA[0.055,0.046] dB[0.015,0.006] g[16.208,14.971]\n",
            "Iteration>298, dA[0.106,0.059] dB[0.029,0.020] g[15.432,14.460]\n",
            "Iteration>299, dA[0.038,0.171] dB[0.031,0.022] g[14.887,12.351]\n",
            "Iteration>300, dA[0.135,0.036] dB[0.024,0.006] g[14.883,12.846]\n",
            "Iteration>301, dA[0.029,0.013] dB[0.029,0.018] g[15.811,14.546]\n",
            "Iteration>302, dA[0.018,0.034] dB[0.015,0.027] g[15.616,13.991]\n",
            "Iteration>303, dA[0.127,0.365] dB[0.017,0.006] g[13.478,11.394]\n",
            "Iteration>304, dA[0.194,0.019] dB[0.014,0.007] g[15.910,14.971]\n",
            "Iteration>305, dA[0.148,0.370] dB[0.010,0.008] g[13.308,10.825]\n",
            "Iteration>306, dA[0.032,0.014] dB[0.010,0.006] g[16.026,14.835]\n",
            "Iteration>307, dA[0.060,0.020] dB[0.024,0.009] g[16.550,15.854]\n",
            "Iteration>308, dA[0.098,0.133] dB[0.009,0.012] g[14.261,12.096]\n",
            "Iteration>309, dA[0.033,0.011] dB[0.014,0.008] g[15.409,13.405]\n",
            "Iteration>310, dA[0.032,0.069] dB[0.005,0.004] g[15.645,14.032]\n",
            "Iteration>311, dA[0.086,0.035] dB[0.011,0.007] g[15.029,13.230]\n",
            "Iteration>312, dA[0.025,0.078] dB[0.061,0.027] g[14.241,11.885]\n",
            "Iteration>313, dA[0.120,0.222] dB[0.032,0.029] g[13.696,11.656]\n",
            "Iteration>314, dA[0.018,0.024] dB[0.018,0.009] g[17.241,16.381]\n",
            "Iteration>315, dA[0.021,0.083] dB[0.033,0.017] g[17.558,17.211]\n",
            "Iteration>316, dA[0.052,0.176] dB[0.017,0.010] g[17.570,16.970]\n",
            "Iteration>317, dA[0.182,0.071] dB[0.015,0.004] g[14.244,12.298]\n",
            "Iteration>318, dA[0.028,0.183] dB[0.010,0.002] g[16.136,14.508]\n",
            "Iteration>319, dA[0.075,0.035] dB[0.014,0.008] g[16.054,15.283]\n",
            "Iteration>320, dA[0.047,0.037] dB[0.016,0.004] g[16.123,15.240]\n",
            "Iteration>321, dA[0.016,0.037] dB[0.014,0.011] g[17.402,16.980]\n",
            "Iteration>322, dA[0.050,0.032] dB[0.043,0.040] g[15.986,14.910]\n",
            "Iteration>323, dA[0.026,0.099] dB[0.059,0.020] g[17.426,16.383]\n",
            "Iteration>324, dA[0.028,0.060] dB[0.074,0.031] g[17.681,17.600]\n",
            "Iteration>325, dA[0.056,0.028] dB[0.020,0.040] g[15.645,14.735]\n",
            "Iteration>326, dA[0.042,0.006] dB[0.026,0.025] g[15.973,14.476]\n",
            "Iteration>327, dA[0.077,0.024] dB[0.027,0.019] g[14.743,12.288]\n",
            "Iteration>328, dA[0.054,0.075] dB[0.010,0.015] g[16.032,14.014]\n",
            "Iteration>329, dA[0.048,0.032] dB[0.036,0.013] g[17.290,16.625]\n",
            "Iteration>330, dA[0.393,0.121] dB[0.010,0.015] g[13.659,11.294]\n",
            "Iteration>331, dA[0.074,0.285] dB[0.010,0.003] g[17.424,15.933]\n",
            "Iteration>332, dA[0.144,0.041] dB[0.014,0.010] g[15.695,14.913]\n",
            "Iteration>333, dA[0.134,0.073] dB[0.008,0.006] g[15.103,13.804]\n",
            "Iteration>334, dA[0.031,0.131] dB[0.019,0.006] g[15.960,14.483]\n",
            "Iteration>335, dA[0.080,0.028] dB[0.006,0.006] g[14.440,12.569]\n",
            "Iteration>336, dA[0.023,0.124] dB[0.006,0.003] g[16.051,14.376]\n",
            "Iteration>337, dA[0.048,0.016] dB[0.011,0.005] g[16.712,15.659]\n",
            "Iteration>338, dA[0.024,0.035] dB[0.012,0.005] g[15.754,14.751]\n",
            "Iteration>339, dA[0.068,0.160] dB[0.004,0.007] g[14.539,12.283]\n",
            "Iteration>340, dA[0.023,0.039] dB[0.010,0.011] g[16.086,14.886]\n",
            "Iteration>341, dA[0.005,0.008] dB[0.022,0.007] g[17.365,17.010]\n",
            "Iteration>342, dA[0.156,0.085] dB[0.028,0.009] g[14.100,12.062]\n",
            "Iteration>343, dA[0.073,0.017] dB[0.011,0.013] g[16.063,14.167]\n",
            "Iteration>344, dA[0.066,0.033] dB[0.013,0.011] g[13.682,11.446]\n",
            "Iteration>345, dA[0.040,0.215] dB[0.024,0.009] g[13.641,10.423]\n",
            "Iteration>346, dA[0.049,0.042] dB[0.043,0.010] g[15.868,14.793]\n",
            "Iteration>347, dA[0.028,0.059] dB[0.020,0.016] g[15.577,14.410]\n",
            "Iteration>348, dA[0.005,0.020] dB[0.018,0.007] g[16.113,14.734]\n",
            "Iteration>349, dA[0.007,0.027] dB[0.018,0.007] g[16.896,16.306]\n",
            "Iteration>350, dA[0.040,0.084] dB[0.025,0.031] g[15.596,14.464]\n",
            "Iteration>351, dA[0.010,0.022] dB[0.013,0.032] g[16.795,16.158]\n",
            "Iteration>352, dA[0.022,0.017] dB[0.016,0.017] g[16.451,15.520]\n",
            "Iteration>353, dA[0.006,0.012] dB[0.017,0.007] g[16.115,14.859]\n",
            "Iteration>354, dA[0.034,0.014] dB[0.014,0.008] g[14.195,12.099]\n",
            "Iteration>355, dA[0.038,0.030] dB[0.025,0.013] g[15.259,13.485]\n",
            "Iteration>356, dA[0.029,0.150] dB[0.013,0.010] g[13.770,11.297]\n",
            "Iteration>357, dA[0.043,0.054] dB[0.033,0.012] g[16.045,15.153]\n",
            "Iteration>358, dA[0.008,0.026] dB[0.025,0.016] g[16.900,16.334]\n",
            "Iteration>359, dA[0.021,0.027] dB[0.014,0.006] g[15.828,14.898]\n",
            "Iteration>360, dA[0.218,0.271] dB[0.018,0.015] g[14.814,13.009]\n",
            "Iteration>361, dA[0.084,0.035] dB[0.030,0.030] g[16.870,16.052]\n",
            "Iteration>362, dA[0.082,0.118] dB[0.022,0.014] g[18.189,18.639]\n",
            "Iteration>363, dA[0.020,0.034] dB[0.024,0.005] g[16.951,16.569]\n",
            "Iteration>364, dA[0.021,0.030] dB[0.016,0.009] g[16.351,14.971]\n",
            "Iteration>365, dA[0.069,0.072] dB[0.012,0.014] g[14.631,12.698]\n",
            "Iteration>366, dA[0.088,0.029] dB[0.008,0.020] g[17.075,16.336]\n",
            "Iteration>367, dA[0.043,0.029] dB[0.023,0.003] g[14.761,13.063]\n",
            "Iteration>368, dA[0.026,0.032] dB[0.018,0.036] g[14.383,12.140]\n",
            "Iteration>369, dA[0.012,0.017] dB[0.029,0.007] g[14.640,12.033]\n",
            "Iteration>370, dA[0.011,0.059] dB[0.021,0.019] g[15.139,13.799]\n",
            "Iteration>371, dA[0.078,0.023] dB[0.008,0.012] g[14.286,13.002]\n",
            "Iteration>372, dA[0.034,0.084] dB[0.022,0.005] g[15.204,13.562]\n",
            "Iteration>373, dA[0.072,0.006] dB[0.016,0.037] g[15.153,13.855]\n",
            "Iteration>374, dA[0.034,0.173] dB[0.033,0.014] g[12.982,11.466]\n",
            "Iteration>375, dA[0.035,0.038] dB[0.009,0.011] g[15.219,13.439]\n",
            "Iteration>376, dA[0.081,0.015] dB[0.030,0.040] g[13.825,12.242]\n",
            "Iteration>377, dA[0.027,0.079] dB[0.019,0.011] g[14.451,12.408]\n",
            "Iteration>378, dA[0.063,0.021] dB[0.010,0.009] g[15.966,14.921]\n",
            "Iteration>379, dA[0.018,0.018] dB[0.007,0.016] g[17.119,16.835]\n",
            "Iteration>380, dA[0.012,0.045] dB[0.034,0.014] g[16.326,14.946]\n",
            "Iteration>381, dA[0.023,0.023] dB[0.009,0.023] g[15.455,14.221]\n",
            "Iteration>382, dA[0.015,0.054] dB[0.068,0.057] g[16.673,15.794]\n",
            "Iteration>383, dA[0.022,0.046] dB[0.038,0.074] g[16.857,16.692]\n",
            "Iteration>384, dA[0.085,0.127] dB[0.066,0.009] g[14.902,13.123]\n",
            "Iteration>385, dA[0.012,0.027] dB[0.025,0.031] g[15.877,14.724]\n",
            "Iteration>386, dA[0.010,0.010] dB[0.012,0.005] g[15.919,14.370]\n",
            "Iteration>387, dA[0.010,0.013] dB[0.016,0.010] g[15.585,14.161]\n",
            "Iteration>388, dA[0.021,0.022] dB[0.008,0.006] g[16.044,15.265]\n",
            "Iteration>389, dA[0.018,0.009] dB[0.006,0.012] g[15.480,13.549]\n",
            "Iteration>390, dA[0.016,0.044] dB[0.017,0.005] g[17.235,16.372]\n",
            "Iteration>391, dA[0.295,0.187] dB[0.019,0.004] g[13.849,12.557]\n",
            "Iteration>392, dA[0.039,0.024] dB[0.006,0.003] g[16.835,16.041]\n",
            "Iteration>393, dA[0.023,0.179] dB[0.004,0.006] g[16.104,14.340]\n",
            "Iteration>394, dA[0.169,0.038] dB[0.006,0.004] g[14.055,12.219]\n",
            "Iteration>395, dA[0.026,0.051] dB[0.006,0.002] g[15.444,13.956]\n",
            "Iteration>396, dA[0.009,0.059] dB[0.016,0.014] g[17.094,16.087]\n",
            "Iteration>397, dA[0.058,0.030] dB[0.004,0.019] g[14.559,12.449]\n",
            "Iteration>398, dA[0.043,0.145] dB[0.054,0.058] g[14.953,13.129]\n",
            "Iteration>399, dA[0.183,0.015] dB[0.053,0.035] g[12.957,11.114]\n",
            "Iteration>400, dA[0.019,0.138] dB[0.040,0.008] g[15.050,12.924]\n",
            "WARNING:tensorflow:Compiled the loaded model, but the compiled metrics have yet to be built. `model.compile_metrics` will be empty until you train or evaluate the model.\n",
            "WARNING:tensorflow:Compiled the loaded model, but the compiled metrics have yet to be built. `model.compile_metrics` will be empty until you train or evaluate the model.\n",
            ">Saved: /content/drive/MyDrive/Project/CycleGAN/TrainedModels/C_Highboost_Real_Model/g_model_AtoB_000400.h5 and /content/drive/MyDrive/Project/CycleGAN/TrainedModels/C_Highboost_Real_Model/g_model_BtoA_000400.h5\n",
            "Iteration>401, dA[0.022,0.031] dB[0.034,0.011] g[17.173,16.682]\n",
            "Iteration>402, dA[0.056,0.028] dB[0.044,0.017] g[14.389,13.052]\n",
            "Iteration>403, dA[0.087,0.044] dB[0.010,0.008] g[13.404,11.529]\n",
            "Iteration>404, dA[0.093,0.057] dB[0.012,0.005] g[14.167,11.484]\n",
            "Iteration>405, dA[0.054,0.050] dB[0.007,0.009] g[16.637,15.453]\n",
            "Iteration>406, dA[0.064,0.022] dB[0.011,0.006] g[16.153,15.508]\n",
            "Iteration>407, dA[0.042,0.035] dB[0.014,0.003] g[14.108,12.380]\n",
            "Iteration>408, dA[0.029,0.050] dB[0.006,0.003] g[15.249,13.730]\n",
            "Iteration>409, dA[0.019,0.013] dB[0.010,0.043] g[15.716,14.846]\n",
            "Iteration>410, dA[0.120,0.054] dB[0.036,0.010] g[13.653,11.427]\n",
            "Iteration>411, dA[0.073,0.086] dB[0.009,0.017] g[15.822,13.737]\n",
            "Iteration>412, dA[0.134,0.158] dB[0.014,0.007] g[15.481,13.747]\n",
            "Iteration>413, dA[0.139,0.056] dB[0.008,0.011] g[15.096,13.833]\n",
            "Iteration>414, dA[0.097,0.040] dB[0.003,0.006] g[14.411,12.692]\n",
            "Iteration>415, dA[0.063,0.111] dB[0.054,0.078] g[14.719,12.513]\n",
            "Iteration>416, dA[0.037,0.036] dB[0.029,0.066] g[15.512,14.331]\n",
            "Iteration>417, dA[0.007,0.017] dB[0.099,0.034] g[17.812,16.933]\n",
            "Iteration>418, dA[0.021,0.071] dB[0.016,0.004] g[14.597,12.969]\n",
            "Iteration>419, dA[0.107,0.089] dB[0.015,0.020] g[14.769,12.497]\n",
            "Iteration>420, dA[0.016,0.025] dB[0.007,0.023] g[15.566,14.315]\n",
            "Iteration>421, dA[0.037,0.039] dB[0.037,0.013] g[14.787,13.579]\n",
            "Iteration>422, dA[0.027,0.015] dB[0.030,0.028] g[15.460,14.240]\n",
            "Iteration>423, dA[0.022,0.025] dB[0.016,0.014] g[15.309,14.107]\n",
            "Iteration>424, dA[0.010,0.034] dB[0.015,0.020] g[16.857,16.746]\n",
            "Iteration>425, dA[0.008,0.023] dB[0.076,0.046] g[15.087,13.765]\n",
            "Iteration>426, dA[0.021,0.017] dB[0.131,0.200] g[16.624,15.902]\n",
            "Iteration>427, dA[0.020,0.018] dB[0.137,0.017] g[15.324,13.932]\n",
            "Iteration>428, dA[0.022,0.025] dB[0.155,0.074] g[16.838,15.464]\n",
            "Iteration>429, dA[0.015,0.042] dB[0.073,0.017] g[16.346,15.278]\n",
            "Iteration>430, dA[0.045,0.025] dB[0.053,0.037] g[15.413,14.218]\n",
            "Iteration>431, dA[0.024,0.013] dB[0.041,0.018] g[15.501,14.409]\n",
            "Iteration>432, dA[0.276,0.548] dB[0.029,0.018] g[14.022,12.255]\n",
            "Iteration>433, dA[0.072,0.036] dB[0.032,0.010] g[13.897,12.188]\n",
            "Iteration>434, dA[0.106,0.056] dB[0.041,0.004] g[13.359,11.488]\n",
            "Iteration>435, dA[0.017,0.099] dB[0.010,0.005] g[14.786,12.443]\n",
            "Iteration>436, dA[0.090,0.025] dB[0.012,0.009] g[14.191,12.028]\n",
            "Iteration>437, dA[0.038,0.072] dB[0.009,0.004] g[15.994,15.164]\n",
            "Iteration>438, dA[0.074,0.028] dB[0.012,0.008] g[16.002,15.217]\n",
            "Iteration>439, dA[0.020,0.073] dB[0.006,0.004] g[16.690,15.841]\n",
            "Iteration>440, dA[0.047,0.034] dB[0.008,0.006] g[16.433,16.115]\n",
            "Iteration>441, dA[0.015,0.051] dB[0.030,0.034] g[15.037,13.767]\n",
            "Iteration>442, dA[0.205,0.036] dB[0.029,0.015] g[13.298,11.874]\n",
            "Iteration>443, dA[0.092,0.252] dB[0.024,0.025] g[15.606,13.611]\n",
            "Iteration>444, dA[0.499,0.019] dB[0.014,0.010] g[13.436,12.314]\n",
            "Iteration>445, dA[0.173,0.259] dB[0.008,0.012] g[12.757,10.450]\n",
            "Iteration>446, dA[0.050,0.075] dB[0.014,0.014] g[15.431,13.616]\n",
            "Iteration>447, dA[0.294,0.145] dB[0.037,0.007] g[13.716,12.049]\n",
            "Iteration>448, dA[0.049,0.017] dB[0.032,0.006] g[15.036,13.695]\n",
            "Iteration>449, dA[0.026,0.064] dB[0.038,0.047] g[16.424,15.381]\n",
            "Iteration>450, dA[0.026,0.008] dB[0.023,0.068] g[13.208,10.800]\n",
            "Iteration>451, dA[0.018,0.022] dB[0.033,0.013] g[14.207,12.229]\n",
            "Iteration>452, dA[0.018,0.124] dB[0.037,0.016] g[14.761,12.516]\n",
            "Iteration>453, dA[0.271,0.083] dB[0.019,0.009] g[13.530,12.291]\n",
            "Iteration>454, dA[0.073,0.020] dB[0.010,0.020] g[13.798,11.858]\n",
            "Iteration>455, dA[0.026,0.116] dB[0.011,0.049] g[14.707,13.025]\n",
            "Iteration>456, dA[0.053,0.139] dB[0.016,0.013] g[16.620,16.926]\n",
            "Iteration>457, dA[0.234,0.205] dB[0.033,0.008] g[15.564,14.305]\n",
            "Iteration>458, dA[0.068,0.119] dB[0.009,0.014] g[16.057,14.618]\n",
            "Iteration>459, dA[0.131,0.066] dB[0.024,0.007] g[14.197,12.782]\n",
            "Iteration>460, dA[0.129,0.155] dB[0.021,0.012] g[12.529,10.358]\n",
            "Iteration>461, dA[0.185,0.053] dB[0.021,0.008] g[13.360,11.810]\n",
            "Iteration>462, dA[0.019,0.013] dB[0.008,0.005] g[14.074,11.844]\n",
            "Iteration>463, dA[0.027,0.010] dB[0.007,0.002] g[14.273,12.267]\n",
            "Iteration>464, dA[0.096,0.080] dB[0.008,0.011] g[13.302,10.638]\n",
            "Iteration>465, dA[0.079,0.085] dB[0.004,0.015] g[16.181,14.784]\n",
            "Iteration>466, dA[0.144,0.034] dB[0.013,0.014] g[13.868,12.339]\n",
            "Iteration>467, dA[0.095,0.074] dB[0.023,0.020] g[13.934,12.271]\n",
            "Iteration>468, dA[0.035,0.024] dB[0.009,0.019] g[13.702,11.726]\n",
            "Iteration>469, dA[0.054,0.134] dB[0.026,0.018] g[12.881,11.155]\n",
            "Iteration>470, dA[0.062,0.017] dB[0.014,0.019] g[13.960,12.204]\n",
            "Iteration>471, dA[0.025,0.034] dB[0.016,0.002] g[15.124,13.904]\n",
            "Iteration>472, dA[0.031,0.029] dB[0.006,0.010] g[14.720,13.710]\n",
            "Iteration>473, dA[0.038,0.107] dB[0.010,0.015] g[14.984,13.210]\n",
            "Iteration>474, dA[0.382,0.027] dB[0.048,0.030] g[13.505,12.471]\n",
            "Iteration>475, dA[0.106,0.027] dB[0.010,0.009] g[13.554,10.655]\n",
            "Iteration>476, dA[0.020,0.046] dB[0.004,0.005] g[16.549,15.823]\n",
            "Iteration>477, dA[0.110,0.017] dB[0.004,0.007] g[15.655,14.572]\n",
            "Iteration>478, dA[0.024,0.036] dB[0.018,0.008] g[15.187,13.803]\n",
            "Iteration>479, dA[0.163,0.456] dB[0.024,0.034] g[13.469,11.500]\n",
            "Iteration>480, dA[0.139,0.057] dB[0.023,0.011] g[15.186,14.492]\n",
            "Iteration>481, dA[0.075,0.044] dB[0.016,0.006] g[15.929,15.332]\n",
            "Iteration>482, dA[0.033,0.106] dB[0.014,0.018] g[14.437,12.608]\n",
            "Iteration>483, dA[0.183,0.112] dB[0.005,0.012] g[13.650,11.282]\n",
            "Iteration>484, dA[0.135,0.182] dB[0.023,0.013] g[13.090,10.908]\n",
            "Iteration>485, dA[0.028,0.021] dB[0.006,0.019] g[13.500,11.050]\n",
            "Iteration>486, dA[0.071,0.072] dB[0.010,0.009] g[13.512,11.803]\n",
            "Iteration>487, dA[0.034,0.034] dB[0.011,0.004] g[16.627,15.783]\n",
            "Iteration>488, dA[0.028,0.039] dB[0.008,0.015] g[15.077,14.193]\n",
            "Iteration>489, dA[0.068,0.060] dB[0.016,0.019] g[12.370,10.247]\n",
            "Iteration>490, dA[0.041,0.063] dB[0.011,0.005] g[16.580,15.289]\n",
            "Iteration>491, dA[0.089,0.019] dB[0.008,0.011] g[15.511,14.517]\n",
            "Iteration>492, dA[0.183,0.637] dB[0.020,0.019] g[13.055,11.138]\n",
            "Iteration>493, dA[0.316,0.052] dB[0.020,0.012] g[14.309,13.835]\n",
            "Iteration>494, dA[0.139,0.131] dB[0.014,0.004] g[16.640,16.517]\n",
            "Iteration>495, dA[0.178,0.181] dB[0.009,0.018] g[13.901,12.258]\n",
            "Iteration>496, dA[0.046,0.157] dB[0.006,0.008] g[15.605,14.328]\n",
            "Iteration>497, dA[0.152,0.027] dB[0.004,0.077] g[15.185,15.056]\n",
            "Iteration>498, dA[0.227,0.291] dB[0.072,0.037] g[14.159,11.939]\n",
            "Iteration>499, dA[0.036,0.043] dB[0.017,0.009] g[16.563,15.676]\n",
            "Iteration>500, dA[0.086,0.058] dB[0.026,0.008] g[15.955,14.873]\n",
            "Iteration>501, dA[0.022,0.043] dB[0.005,0.012] g[14.804,12.928]\n",
            "Iteration>502, dA[0.025,0.014] dB[0.029,0.036] g[14.726,13.285]\n",
            "Iteration>503, dA[0.044,0.068] dB[0.030,0.010] g[16.799,16.197]\n",
            "Iteration>504, dA[0.020,0.033] dB[0.012,0.031] g[16.889,16.304]\n",
            "Iteration>505, dA[0.046,0.014] dB[0.024,0.010] g[15.453,14.176]\n",
            "Iteration>506, dA[0.014,0.082] dB[0.014,0.003] g[15.348,13.410]\n",
            "Iteration>507, dA[0.016,0.031] dB[0.010,0.003] g[16.941,16.781]\n",
            "Iteration>508, dA[0.203,0.230] dB[0.008,0.006] g[13.496,12.178]\n",
            "Iteration>509, dA[0.011,0.008] dB[0.025,0.037] g[15.235,13.961]\n",
            "Iteration>510, dA[0.260,0.078] dB[0.019,0.004] g[14.940,14.282]\n",
            "Iteration>511, dA[0.023,0.035] dB[0.023,0.020] g[14.577,13.015]\n",
            "Iteration>512, dA[0.068,0.106] dB[0.014,0.007] g[13.235,11.376]\n",
            "Iteration>513, dA[0.034,0.016] dB[0.008,0.025] g[14.784,13.527]\n",
            "Iteration>514, dA[0.031,0.057] dB[0.018,0.006] g[14.884,13.516]\n",
            "Iteration>515, dA[0.011,0.025] dB[0.011,0.016] g[14.920,13.833]\n",
            "Iteration>516, dA[0.033,0.024] dB[0.012,0.005] g[13.900,12.026]\n",
            "Iteration>517, dA[0.028,0.070] dB[0.007,0.015] g[15.273,13.848]\n",
            "Iteration>518, dA[0.038,0.050] dB[0.037,0.032] g[15.312,14.900]\n",
            "Iteration>519, dA[0.006,0.025] dB[0.047,0.007] g[15.018,14.244]\n",
            "Iteration>520, dA[0.008,0.033] dB[0.022,0.023] g[15.670,14.007]\n",
            "Iteration>521, dA[0.150,0.022] dB[0.025,0.015] g[13.393,11.777]\n",
            "Iteration>522, dA[0.072,0.121] dB[0.007,0.008] g[16.687,15.055]\n",
            "Iteration>523, dA[0.217,0.088] dB[0.020,0.007] g[13.643,11.905]\n",
            "Iteration>524, dA[0.020,0.095] dB[0.009,0.014] g[15.843,14.852]\n",
            "Iteration>525, dA[0.088,0.009] dB[0.008,0.004] g[14.845,13.715]\n",
            "Iteration>526, dA[0.388,0.405] dB[0.013,0.011] g[12.628,10.478]\n",
            "Iteration>527, dA[0.029,0.030] dB[0.015,0.013] g[16.533,15.538]\n",
            "Iteration>528, dA[0.110,0.044] dB[0.019,0.007] g[14.294,12.673]\n",
            "Iteration>529, dA[0.029,0.033] dB[0.014,0.044] g[14.060,11.565]\n",
            "Iteration>530, dA[0.105,0.367] dB[0.043,0.017] g[12.325,9.454]\n",
            "Iteration>531, dA[0.210,0.040] dB[0.006,0.008] g[14.829,13.643]\n",
            "Iteration>532, dA[0.018,0.056] dB[0.011,0.017] g[14.462,13.305]\n",
            "Iteration>533, dA[0.030,0.013] dB[0.013,0.029] g[13.430,11.261]\n",
            "Iteration>534, dA[0.013,0.095] dB[0.019,0.015] g[16.117,14.758]\n",
            "Iteration>535, dA[0.085,0.008] dB[0.008,0.010] g[13.847,12.968]\n",
            "Iteration>536, dA[0.076,0.156] dB[0.023,0.025] g[12.737,10.868]\n",
            "Iteration>537, dA[0.026,0.027] dB[0.020,0.007] g[14.457,12.912]\n",
            "Iteration>538, dA[0.087,0.051] dB[0.016,0.035] g[13.790,12.870]\n",
            "Iteration>539, dA[0.045,0.046] dB[0.018,0.017] g[15.013,13.845]\n",
            "Iteration>540, dA[0.012,0.047] dB[0.077,0.025] g[16.547,16.008]\n",
            "Iteration>541, dA[0.017,0.037] dB[0.010,0.102] g[15.974,15.912]\n",
            "Iteration>542, dA[0.041,0.034] dB[0.078,0.016] g[17.320,17.413]\n",
            "Iteration>543, dA[0.127,0.030] dB[0.017,0.011] g[13.689,12.245]\n",
            "Iteration>544, dA[0.089,0.019] dB[0.024,0.098] g[14.321,13.215]\n",
            "Iteration>545, dA[0.053,0.013] dB[0.044,0.132] g[13.015,11.543]\n",
            "Iteration>546, dA[0.012,0.022] dB[0.163,0.115] g[14.297,10.808]\n",
            "Iteration>547, dA[0.009,0.122] dB[0.235,0.139] g[15.481,13.292]\n",
            "Iteration>548, dA[0.121,0.048] dB[0.099,0.219] g[14.303,13.150]\n",
            "Iteration>549, dA[0.045,0.195] dB[0.157,0.036] g[13.721,11.402]\n",
            "Iteration>550, dA[0.044,0.090] dB[0.074,0.128] g[16.338,16.552]\n",
            "Iteration>551, dA[0.023,0.022] dB[0.065,0.042] g[17.437,17.915]\n",
            "Iteration>552, dA[0.035,0.046] dB[0.060,0.097] g[14.669,13.817]\n",
            "Iteration>553, dA[0.041,0.054] dB[0.067,0.044] g[13.948,12.489]\n",
            "Iteration>554, dA[0.043,0.017] dB[0.053,0.214] g[13.643,12.094]\n",
            "Iteration>555, dA[0.113,0.174] dB[0.115,0.033] g[13.371,10.239]\n",
            "Iteration>556, dA[0.037,0.016] dB[0.018,0.115] g[13.317,11.296]\n",
            "Iteration>557, dA[0.012,0.028] dB[0.054,0.045] g[16.046,15.834]\n",
            "Iteration>558, dA[0.074,0.117] dB[0.027,0.158] g[13.820,11.787]\n",
            "Iteration>559, dA[0.014,0.059] dB[0.116,0.075] g[14.696,13.682]\n",
            "Iteration>560, dA[0.322,0.372] dB[0.044,0.048] g[13.084,11.385]\n",
            "Iteration>561, dA[0.033,0.009] dB[0.022,0.041] g[15.072,14.000]\n",
            "Iteration>562, dA[0.095,0.028] dB[0.041,0.020] g[15.433,14.717]\n",
            "Iteration>563, dA[0.006,0.013] dB[0.045,0.047] g[15.049,13.815]\n",
            "Iteration>564, dA[0.011,0.021] dB[0.018,0.015] g[15.258,14.230]\n",
            "Iteration>565, dA[0.031,0.038] dB[0.037,0.032] g[13.711,12.234]\n",
            "Iteration>566, dA[0.020,0.030] dB[0.010,0.015] g[16.286,15.832]\n",
            "Iteration>567, dA[0.070,0.034] dB[0.036,0.091] g[14.276,12.861]\n",
            "Iteration>568, dA[0.007,0.027] dB[0.096,0.017] g[16.105,15.478]\n",
            "Iteration>569, dA[0.064,0.022] dB[0.172,0.184] g[13.091,12.393]\n",
            "Iteration>570, dA[0.014,0.011] dB[0.091,0.095] g[15.201,13.568]\n",
            "Iteration>571, dA[0.020,0.015] dB[0.047,0.087] g[14.470,12.493]\n",
            "Iteration>572, dA[0.016,0.008] dB[0.110,0.031] g[13.376,11.156]\n",
            "Iteration>573, dA[0.016,0.013] dB[0.064,0.011] g[15.768,14.544]\n",
            "Iteration>574, dA[0.009,0.012] dB[0.025,0.017] g[14.333,12.951]\n",
            "Iteration>575, dA[0.017,0.017] dB[0.013,0.145] g[13.520,12.191]\n",
            "Iteration>576, dA[0.012,0.011] dB[0.180,0.034] g[15.609,12.983]\n",
            "Iteration>577, dA[0.032,0.014] dB[0.060,0.030] g[13.129,11.258]\n",
            "Iteration>578, dA[0.020,0.194] dB[0.028,0.036] g[15.018,13.637]\n",
            "Iteration>579, dA[0.227,0.039] dB[0.043,0.024] g[14.962,15.528]\n",
            "Iteration>580, dA[0.028,0.021] dB[0.024,0.015] g[15.270,14.873]\n",
            "Iteration>581, dA[0.344,0.141] dB[0.013,0.013] g[12.457,12.162]\n",
            "Iteration>582, dA[0.111,0.035] dB[0.019,0.222] g[13.582,12.187]\n",
            "Iteration>583, dA[0.042,0.015] dB[0.364,0.075] g[17.035,15.671]\n",
            "Iteration>584, dA[0.021,0.016] dB[0.025,0.040] g[17.276,16.987]\n",
            "Iteration>585, dA[0.020,0.026] dB[0.034,0.037] g[15.032,13.630]\n",
            "Iteration>586, dA[0.015,0.107] dB[0.055,0.014] g[14.638,12.921]\n",
            "Iteration>587, dA[0.097,0.036] dB[0.024,0.010] g[14.529,14.200]\n",
            "Iteration>588, dA[0.012,0.004] dB[0.013,0.005] g[15.932,14.994]\n",
            "Iteration>589, dA[0.011,0.011] dB[0.018,0.013] g[14.882,13.941]\n",
            "Iteration>590, dA[0.214,0.126] dB[0.013,0.023] g[14.555,13.372]\n",
            "Iteration>591, dA[0.061,0.029] dB[0.018,0.010] g[16.262,15.630]\n",
            "Iteration>592, dA[0.027,0.028] dB[0.018,0.036] g[15.742,15.397]\n",
            "Iteration>593, dA[0.057,0.018] dB[0.024,0.009] g[13.746,12.467]\n",
            "Iteration>594, dA[0.024,0.030] dB[0.037,0.109] g[15.546,14.865]\n",
            "Iteration>595, dA[0.296,0.436] dB[0.035,0.016] g[13.145,11.614]\n",
            "Iteration>596, dA[0.017,0.081] dB[0.032,0.011] g[14.820,14.141]\n",
            "Iteration>597, dA[0.028,0.026] dB[0.020,0.019] g[16.004,16.340]\n",
            "Iteration>598, dA[0.022,0.032] dB[0.018,0.010] g[13.892,13.084]\n",
            "Iteration>599, dA[0.018,0.025] dB[0.012,0.004] g[13.566,11.693]\n",
            "Iteration>600, dA[0.030,0.043] dB[0.006,0.022] g[12.765,11.081]\n",
            "Iteration>601, dA[0.017,0.030] dB[0.016,0.015] g[14.529,12.970]\n",
            "Iteration>602, dA[0.007,0.064] dB[0.027,0.034] g[13.757,12.595]\n",
            "Iteration>603, dA[0.031,0.030] dB[0.025,0.018] g[16.518,16.482]\n",
            "Iteration>604, dA[0.009,0.014] dB[0.043,0.017] g[14.674,13.464]\n",
            "Iteration>605, dA[0.027,0.083] dB[0.022,0.016] g[16.503,16.513]\n",
            "Iteration>606, dA[0.080,0.023] dB[0.013,0.075] g[14.001,13.774]\n",
            "Iteration>607, dA[0.132,0.164] dB[0.044,0.020] g[13.928,12.467]\n",
            "Iteration>608, dA[0.019,0.016] dB[0.013,0.100] g[15.876,15.040]\n",
            "Iteration>609, dA[0.023,0.021] dB[0.036,0.040] g[16.273,15.770]\n",
            "Iteration>610, dA[0.091,0.164] dB[0.032,0.111] g[13.147,11.586]\n",
            "Iteration>611, dA[0.024,0.011] dB[0.051,0.037] g[14.089,11.936]\n",
            "Iteration>612, dA[0.039,0.032] dB[0.032,0.019] g[13.022,11.443]\n",
            "Iteration>613, dA[0.318,0.433] dB[0.021,0.053] g[13.029,11.334]\n",
            "Iteration>614, dA[0.028,0.030] dB[0.052,0.015] g[16.930,16.394]\n",
            "Iteration>615, dA[0.115,0.074] dB[0.025,0.014] g[15.361,14.974]\n",
            "Iteration>616, dA[0.020,0.018] dB[0.029,0.142] g[14.213,13.084]\n",
            "Iteration>617, dA[0.015,0.049] dB[0.025,0.037] g[14.975,13.508]\n",
            "Iteration>618, dA[0.033,0.006] dB[0.056,0.012] g[14.178,12.965]\n",
            "Iteration>619, dA[0.017,0.023] dB[0.088,0.063] g[13.891,12.818]\n",
            "Iteration>620, dA[0.008,0.022] dB[0.011,0.091] g[14.398,13.245]\n",
            "Iteration>621, dA[0.068,0.024] dB[0.031,0.081] g[13.173,12.038]\n",
            "Iteration>622, dA[0.011,0.070] dB[0.091,0.015] g[13.939,11.836]\n",
            "Iteration>623, dA[0.014,0.007] dB[0.040,0.014] g[16.908,16.592]\n",
            "Iteration>624, dA[0.168,0.280] dB[0.023,0.013] g[13.640,12.315]\n",
            "Iteration>625, dA[0.066,0.023] dB[0.025,0.013] g[13.930,12.492]\n",
            "Iteration>626, dA[0.025,0.006] dB[0.011,0.015] g[14.442,13.806]\n",
            "Iteration>627, dA[0.025,0.011] dB[0.028,0.011] g[15.852,15.590]\n",
            "Iteration>628, dA[0.005,0.019] dB[0.027,0.039] g[15.235,13.686]\n",
            "Iteration>629, dA[0.038,0.014] dB[0.029,0.009] g[14.447,12.609]\n",
            "Iteration>630, dA[0.032,0.014] dB[0.014,0.013] g[15.201,13.975]\n",
            "Iteration>631, dA[0.015,0.018] dB[0.033,0.023] g[14.779,13.584]\n",
            "Iteration>632, dA[0.030,0.020] dB[0.027,0.006] g[13.190,11.588]\n",
            "Iteration>633, dA[0.030,0.048] dB[0.016,0.027] g[13.905,12.203]\n",
            "Iteration>634, dA[0.016,0.061] dB[0.012,0.087] g[15.285,14.215]\n",
            "Iteration>635, dA[0.006,0.028] dB[0.084,0.020] g[15.256,13.437]\n",
            "Iteration>636, dA[0.026,0.032] dB[0.023,0.038] g[13.437,11.760]\n",
            "Iteration>637, dA[0.012,0.013] dB[0.021,0.006] g[15.562,14.603]\n",
            "Iteration>638, dA[0.040,0.027] dB[0.041,0.011] g[13.496,12.166]\n",
            "Iteration>639, dA[0.037,0.013] dB[0.020,0.006] g[14.355,13.082]\n",
            "Iteration>640, dA[0.014,0.009] dB[0.012,0.012] g[15.888,15.295]\n",
            "Iteration>641, dA[0.040,0.027] dB[0.017,0.010] g[13.878,12.742]\n",
            "Iteration>642, dA[0.055,0.064] dB[0.009,0.008] g[13.440,11.396]\n",
            "Iteration>643, dA[0.102,0.011] dB[0.007,0.007] g[13.320,11.945]\n",
            "Iteration>644, dA[0.011,0.011] dB[0.006,0.006] g[14.953,13.641]\n",
            "Iteration>645, dA[0.364,0.567] dB[0.010,0.013] g[12.529,10.628]\n",
            "Iteration>646, dA[0.034,0.032] dB[0.014,0.017] g[13.313,11.814]\n",
            "Iteration>647, dA[0.035,0.017] dB[0.006,0.006] g[15.547,14.935]\n",
            "Iteration>648, dA[0.007,0.008] dB[0.006,0.005] g[15.995,15.451]\n",
            "Iteration>649, dA[0.174,0.301] dB[0.007,0.003] g[12.974,10.841]\n",
            "Iteration>650, dA[0.018,0.046] dB[0.012,0.002] g[14.519,13.138]\n",
            "Iteration>651, dA[0.029,0.022] dB[0.007,0.001] g[14.502,13.694]\n",
            "Iteration>652, dA[0.043,0.037] dB[0.011,0.008] g[13.930,12.766]\n",
            "Iteration>653, dA[0.014,0.014] dB[0.012,0.031] g[15.289,14.367]\n",
            "Iteration>654, dA[0.008,0.020] dB[0.022,0.006] g[15.088,14.117]\n",
            "Iteration>655, dA[0.015,0.023] dB[0.014,0.015] g[16.020,15.441]\n",
            "Iteration>656, dA[0.014,0.006] dB[0.004,0.005] g[14.530,13.392]\n",
            "Iteration>657, dA[0.011,0.028] dB[0.005,0.009] g[15.721,14.890]\n",
            "Iteration>658, dA[0.136,0.026] dB[0.005,0.010] g[13.353,11.373]\n",
            "Iteration>659, dA[0.041,0.135] dB[0.005,0.004] g[14.214,12.334]\n",
            "Iteration>660, dA[0.167,0.011] dB[0.005,0.007] g[13.227,12.004]\n",
            "Iteration>661, dA[0.027,0.010] dB[0.015,0.026] g[14.419,13.031]\n",
            "Iteration>662, dA[0.017,0.048] dB[0.005,0.014] g[13.412,11.141]\n",
            "Iteration>663, dA[0.085,0.015] dB[0.003,0.004] g[13.145,11.388]\n",
            "Iteration>664, dA[0.028,0.103] dB[0.020,0.029] g[12.700,10.128]\n",
            "Iteration>665, dA[0.034,0.079] dB[0.006,0.008] g[14.662,13.909]\n",
            "Iteration>666, dA[0.033,0.083] dB[0.006,0.007] g[14.408,13.166]\n",
            "Iteration>667, dA[0.019,0.010] dB[0.008,0.002] g[14.074,12.497]\n",
            "Iteration>668, dA[0.036,0.014] dB[0.010,0.007] g[14.102,12.986]\n",
            "Iteration>669, dA[0.063,0.057] dB[0.005,0.001] g[13.037,11.499]\n",
            "Iteration>670, dA[0.019,0.014] dB[0.008,0.008] g[14.478,13.385]\n",
            "Iteration>671, dA[0.053,0.049] dB[0.011,0.014] g[13.085,11.541]\n",
            "Iteration>672, dA[0.036,0.036] dB[0.019,0.002] g[15.196,13.878]\n",
            "Iteration>673, dA[0.021,0.036] dB[0.013,0.012] g[15.216,14.517]\n",
            "Iteration>674, dA[0.019,0.017] dB[0.007,0.001] g[16.015,15.403]\n",
            "Iteration>675, dA[0.033,0.016] dB[0.006,0.056] g[12.851,11.214]\n",
            "Iteration>676, dA[0.027,0.157] dB[0.058,0.017] g[14.414,12.885]\n",
            "Iteration>677, dA[0.129,0.034] dB[0.030,0.056] g[13.267,12.917]\n",
            "Iteration>678, dA[0.009,0.014] dB[0.011,0.030] g[14.265,13.059]\n",
            "Iteration>679, dA[0.022,0.014] dB[0.018,0.011] g[15.466,14.300]\n",
            "Iteration>680, dA[0.030,0.027] dB[0.027,0.012] g[13.119,11.578]\n",
            "Iteration>681, dA[0.017,0.011] dB[0.002,0.003] g[14.219,12.608]\n",
            "Iteration>682, dA[0.019,0.039] dB[0.008,0.003] g[15.597,14.654]\n",
            "Iteration>683, dA[0.019,0.029] dB[0.003,0.002] g[12.689,10.290]\n",
            "Iteration>684, dA[0.016,0.088] dB[0.004,0.002] g[13.162,10.832]\n",
            "Iteration>685, dA[0.060,0.029] dB[0.015,0.008] g[12.726,11.331]\n",
            "Iteration>686, dA[0.009,0.009] dB[0.009,0.008] g[15.918,15.446]\n",
            "Iteration>687, dA[0.012,0.023] dB[0.015,0.015] g[14.269,12.925]\n",
            "Iteration>688, dA[0.018,0.018] dB[0.026,0.010] g[14.714,14.062]\n",
            "Iteration>689, dA[0.052,0.011] dB[0.008,0.004] g[13.582,12.161]\n",
            "Iteration>690, dA[0.022,0.154] dB[0.014,0.023] g[14.413,12.738]\n",
            "Iteration>691, dA[0.080,0.109] dB[0.024,0.011] g[14.091,13.655]\n",
            "Iteration>692, dA[0.061,0.059] dB[0.020,0.006] g[13.642,12.736]\n",
            "Iteration>693, dA[0.022,0.050] dB[0.016,0.015] g[14.734,13.425]\n",
            "Iteration>694, dA[0.010,0.016] dB[0.009,0.006] g[14.479,13.706]\n",
            "Iteration>695, dA[0.040,0.030] dB[0.009,0.004] g[14.248,13.438]\n",
            "Iteration>696, dA[0.015,0.057] dB[0.006,0.013] g[13.682,11.659]\n",
            "Iteration>697, dA[0.015,0.006] dB[0.014,0.004] g[14.120,12.680]\n",
            "Iteration>698, dA[0.014,0.026] dB[0.019,0.007] g[15.750,15.653]\n",
            "Iteration>699, dA[0.013,0.007] dB[0.015,0.003] g[14.480,13.195]\n",
            "Iteration>700, dA[0.007,0.010] dB[0.003,0.004] g[15.339,14.649]\n",
            "Iteration>701, dA[0.303,0.445] dB[0.004,0.001] g[12.147,10.477]\n",
            "Iteration>702, dA[0.019,0.015] dB[0.003,0.001] g[14.195,12.661]\n",
            "Iteration>703, dA[0.484,0.217] dB[0.005,0.006] g[11.688,9.935]\n",
            "Iteration>704, dA[0.026,0.083] dB[0.003,0.003] g[13.783,12.600]\n",
            "Iteration>705, dA[0.212,0.065] dB[0.003,0.001] g[14.332,13.057]\n",
            "Iteration>706, dA[0.027,0.012] dB[0.008,0.002] g[14.648,13.597]\n",
            "Iteration>707, dA[0.128,0.045] dB[0.007,0.003] g[12.588,10.931]\n",
            "Iteration>708, dA[0.032,0.181] dB[0.007,0.003] g[13.491,11.085]\n",
            "Iteration>709, dA[0.124,0.032] dB[0.006,0.002] g[14.425,13.651]\n",
            "Iteration>710, dA[0.010,0.038] dB[0.006,0.015] g[14.514,13.414]\n",
            "Iteration>711, dA[0.047,0.015] dB[0.021,0.003] g[14.177,12.601]\n",
            "Iteration>712, dA[0.016,0.012] dB[0.006,0.016] g[12.808,10.915]\n",
            "Iteration>713, dA[0.013,0.028] dB[0.022,0.003] g[14.847,13.773]\n",
            "Iteration>714, dA[0.017,0.028] dB[0.011,0.008] g[14.275,13.500]\n",
            "Iteration>715, dA[0.004,0.010] dB[0.008,0.020] g[14.359,13.133]\n",
            "Iteration>716, dA[0.012,0.012] dB[0.002,0.021] g[14.105,13.017]\n",
            "Iteration>717, dA[0.039,0.007] dB[0.024,0.003] g[13.043,11.697]\n",
            "Iteration>718, dA[0.020,0.015] dB[0.008,0.018] g[14.351,12.726]\n",
            "Iteration>719, dA[0.015,0.067] dB[0.011,0.001] g[12.424,10.198]\n",
            "Iteration>720, dA[0.045,0.028] dB[0.011,0.016] g[15.203,15.075]\n",
            "Iteration>721, dA[0.077,0.202] dB[0.007,0.016] g[14.497,13.285]\n",
            "Iteration>722, dA[0.013,0.016] dB[0.012,0.008] g[13.342,11.776]\n",
            "Iteration>723, dA[0.436,0.317] dB[0.009,0.005] g[12.046,10.256]\n",
            "Iteration>724, dA[0.104,0.113] dB[0.026,0.035] g[11.923,9.721]\n",
            "Iteration>725, dA[0.033,0.072] dB[0.012,0.033] g[12.558,10.832]\n",
            "Iteration>726, dA[0.012,0.014] dB[0.030,0.006] g[14.568,13.210]\n",
            "Iteration>727, dA[0.056,0.011] dB[0.034,0.014] g[12.871,11.175]\n",
            "Iteration>728, dA[0.052,0.107] dB[0.011,0.005] g[15.810,14.744]\n",
            "Iteration>729, dA[0.051,0.033] dB[0.004,0.007] g[15.234,15.077]\n",
            "Iteration>730, dA[0.010,0.009] dB[0.009,0.008] g[15.571,15.529]\n",
            "Iteration>731, dA[0.012,0.007] dB[0.036,0.003] g[13.164,11.385]\n",
            "Iteration>732, dA[0.006,0.011] dB[0.025,0.024] g[14.534,13.430]\n",
            "Iteration>733, dA[0.010,0.006] dB[0.002,0.023] g[14.876,14.153]\n",
            "Iteration>734, dA[0.033,0.028] dB[0.035,0.016] g[14.051,12.888]\n",
            "Iteration>735, dA[0.007,0.005] dB[0.012,0.009] g[13.546,12.190]\n",
            "Iteration>736, dA[0.183,0.316] dB[0.024,0.024] g[12.457,10.704]\n",
            "Iteration>737, dA[0.085,0.077] dB[0.011,0.005] g[12.942,11.059]\n",
            "Iteration>738, dA[0.242,0.318] dB[0.042,0.027] g[12.371,10.114]\n",
            "Iteration>739, dA[0.021,0.054] dB[0.017,0.027] g[14.427,12.715]\n",
            "Iteration>740, dA[0.069,0.134] dB[0.038,0.007] g[15.544,15.025]\n",
            "Iteration>741, dA[0.137,0.091] dB[0.009,0.004] g[14.202,13.237]\n",
            "Iteration>742, dA[0.214,0.272] dB[0.006,0.005] g[13.319,11.635]\n",
            "Iteration>743, dA[0.137,0.114] dB[0.005,0.019] g[12.616,10.847]\n",
            "Iteration>744, dA[0.327,0.499] dB[0.033,0.020] g[11.954,10.839]\n",
            "Iteration>745, dA[0.157,0.033] dB[0.008,0.022] g[14.989,13.614]\n",
            "Iteration>746, dA[0.157,0.116] dB[0.027,0.003] g[14.443,13.670]\n",
            "Iteration>747, dA[0.074,0.157] dB[0.008,0.010] g[13.930,11.677]\n",
            "Iteration>748, dA[0.115,0.053] dB[0.006,0.003] g[13.398,11.846]\n",
            "Iteration>749, dA[0.052,0.071] dB[0.009,0.004] g[13.300,10.862]\n",
            "Iteration>750, dA[0.034,0.117] dB[0.005,0.009] g[12.768,10.518]\n",
            "Iteration>751, dA[0.040,0.068] dB[0.006,0.004] g[14.552,13.222]\n",
            "Iteration>752, dA[0.043,0.050] dB[0.020,0.023] g[15.160,14.972]\n",
            "Iteration>753, dA[0.021,0.086] dB[0.031,0.071] g[13.806,12.721]\n",
            "Iteration>754, dA[0.074,0.045] dB[0.059,0.012] g[14.479,13.422]\n",
            "Iteration>755, dA[0.230,0.443] dB[0.074,0.030] g[12.362,10.773]\n",
            "Iteration>756, dA[0.116,0.079] dB[0.023,0.060] g[14.759,14.429]\n",
            "Iteration>757, dA[0.071,0.094] dB[0.031,0.004] g[15.507,15.270]\n",
            "Iteration>758, dA[0.170,0.069] dB[0.022,0.023] g[12.788,11.286]\n",
            "Iteration>759, dA[0.058,0.035] dB[0.008,0.004] g[14.544,12.845]\n",
            "Iteration>760, dA[0.089,0.112] dB[0.005,0.002] g[14.322,12.705]\n",
            "Iteration>761, dA[0.148,0.099] dB[0.007,0.008] g[12.579,10.668]\n",
            "Iteration>762, dA[0.086,0.122] dB[0.006,0.013] g[12.658,10.998]\n",
            "Iteration>763, dA[0.012,0.031] dB[0.007,0.010] g[15.836,14.987]\n",
            "Iteration>764, dA[0.064,0.027] dB[0.009,0.011] g[13.314,12.594]\n",
            "Iteration>765, dA[0.050,0.010] dB[0.009,0.004] g[13.852,12.392]\n",
            "Iteration>766, dA[0.019,0.046] dB[0.003,0.001] g[15.462,14.984]\n",
            "Iteration>767, dA[0.018,0.016] dB[0.005,0.009] g[14.483,13.651]\n",
            "Iteration>768, dA[0.043,0.055] dB[0.007,0.002] g[13.656,11.754]\n",
            "Iteration>769, dA[0.024,0.029] dB[0.007,0.004] g[12.413,10.671]\n",
            "Iteration>770, dA[0.015,0.047] dB[0.009,0.014] g[14.265,13.209]\n",
            "Iteration>771, dA[0.008,0.029] dB[0.015,0.005] g[15.127,14.618]\n",
            "Iteration>772, dA[0.037,0.042] dB[0.010,0.005] g[13.119,12.154]\n",
            "Iteration>773, dA[0.036,0.018] dB[0.005,0.009] g[14.643,13.340]\n",
            "Iteration>774, dA[0.064,0.028] dB[0.006,0.007] g[13.459,12.055]\n",
            "Iteration>775, dA[0.010,0.021] dB[0.020,0.016] g[15.295,14.981]\n",
            "Iteration>776, dA[0.022,0.020] dB[0.007,0.022] g[13.544,12.455]\n",
            "Iteration>777, dA[0.016,0.038] dB[0.034,0.014] g[13.369,11.345]\n",
            "Iteration>778, dA[0.136,0.362] dB[0.014,0.063] g[11.977,10.241]\n",
            "Iteration>779, dA[0.126,0.054] dB[0.026,0.016] g[12.434,10.754]\n",
            "Iteration>780, dA[0.028,0.272] dB[0.028,0.005] g[13.911,12.102]\n",
            "Iteration>781, dA[0.139,0.029] dB[0.006,0.003] g[13.747,12.733]\n",
            "Iteration>782, dA[0.149,0.101] dB[0.012,0.016] g[12.078,10.154]\n",
            "Iteration>783, dA[0.022,0.144] dB[0.005,0.013] g[12.001,9.242]\n",
            "Iteration>784, dA[0.262,0.030] dB[0.012,0.002] g[13.171,11.045]\n",
            "Iteration>785, dA[0.025,0.066] dB[0.004,0.004] g[14.424,13.098]\n",
            "Iteration>786, dA[0.296,0.316] dB[0.008,0.007] g[12.911,11.088]\n",
            "Iteration>787, dA[0.285,0.082] dB[0.004,0.007] g[12.464,9.878]\n",
            "Iteration>788, dA[0.019,0.070] dB[0.010,0.006] g[15.495,14.595]\n",
            "Iteration>789, dA[0.090,0.130] dB[0.013,0.009] g[13.911,12.758]\n",
            "Iteration>790, dA[0.127,0.050] dB[0.011,0.006] g[14.615,13.388]\n",
            "Iteration>791, dA[0.150,0.163] dB[0.012,0.007] g[12.180,9.926]\n",
            "Iteration>792, dA[0.021,0.023] dB[0.010,0.007] g[14.880,13.749]\n",
            "Iteration>793, dA[0.239,0.103] dB[0.005,0.019] g[12.784,10.836]\n",
            "Iteration>794, dA[0.043,0.108] dB[0.014,0.003] g[13.926,12.336]\n",
            "Iteration>795, dA[0.482,0.154] dB[0.014,0.028] g[12.944,12.454]\n",
            "Iteration>796, dA[0.047,0.015] dB[0.014,0.003] g[15.659,15.021]\n",
            "Iteration>797, dA[0.155,0.123] dB[0.026,0.020] g[12.722,11.439]\n",
            "Iteration>798, dA[0.033,0.005] dB[0.007,0.022] g[14.150,12.685]\n",
            "Iteration>799, dA[0.093,0.038] dB[0.014,0.017] g[13.452,11.895]\n",
            "Iteration>800, dA[0.015,0.014] dB[0.031,0.008] g[13.511,12.108]\n",
            "WARNING:tensorflow:Compiled the loaded model, but the compiled metrics have yet to be built. `model.compile_metrics` will be empty until you train or evaluate the model.\n",
            "WARNING:tensorflow:Compiled the loaded model, but the compiled metrics have yet to be built. `model.compile_metrics` will be empty until you train or evaluate the model.\n",
            ">Saved: /content/drive/MyDrive/Project/CycleGAN/TrainedModels/C_Highboost_Real_Model/g_model_AtoB_000800.h5 and /content/drive/MyDrive/Project/CycleGAN/TrainedModels/C_Highboost_Real_Model/g_model_BtoA_000800.h5\n",
            "Iteration>801, dA[0.034,0.012] dB[0.014,0.010] g[13.816,12.225]\n",
            "Iteration>802, dA[0.012,0.010] dB[0.007,0.006] g[12.290,10.716]\n",
            "Iteration>803, dA[0.023,0.007] dB[0.008,0.008] g[14.026,12.935]\n",
            "Iteration>804, dA[0.016,0.009] dB[0.016,0.032] g[12.130,10.265]\n",
            "Iteration>805, dA[0.021,0.114] dB[0.044,0.007] g[14.706,13.174]\n",
            "Iteration>806, dA[0.124,0.095] dB[0.074,0.061] g[13.878,14.249]\n",
            "Iteration>807, dA[0.023,0.040] dB[0.017,0.051] g[15.163,14.968]\n",
            "Iteration>808, dA[0.064,0.017] dB[0.059,0.008] g[13.532,12.112]\n",
            "Iteration>809, dA[0.021,0.017] dB[0.010,0.021] g[13.996,12.776]\n",
            "Iteration>810, dA[0.005,0.010] dB[0.005,0.005] g[14.598,13.759]\n",
            "Iteration>811, dA[0.004,0.011] dB[0.004,0.004] g[13.795,12.503]\n",
            "Iteration>812, dA[0.024,0.052] dB[0.006,0.004] g[13.663,12.789]\n",
            "Iteration>813, dA[0.012,0.039] dB[0.012,0.012] g[14.349,13.705]\n",
            "Iteration>814, dA[0.007,0.014] dB[0.008,0.002] g[15.401,15.092]\n",
            "Iteration>815, dA[0.012,0.028] dB[0.007,0.004] g[14.187,12.543]\n",
            "Iteration>816, dA[0.022,0.009] dB[0.004,0.002] g[14.173,13.095]\n",
            "Iteration>817, dA[0.015,0.019] dB[0.004,0.004] g[13.631,12.779]\n",
            "Iteration>818, dA[0.017,0.012] dB[0.007,0.002] g[12.925,11.204]\n",
            "Iteration>819, dA[0.219,0.549] dB[0.014,0.017] g[11.322,10.121]\n",
            "Iteration>820, dA[0.197,0.031] dB[0.015,0.005] g[12.680,10.894]\n",
            "Iteration>821, dA[0.297,0.214] dB[0.007,0.002] g[11.369,9.493]\n",
            "Iteration>822, dA[0.036,0.011] dB[0.017,0.011] g[12.448,10.114]\n",
            "Iteration>823, dA[0.216,0.050] dB[0.032,0.024] g[12.545,10.592]\n",
            "Iteration>824, dA[0.029,0.081] dB[0.017,0.008] g[12.799,10.742]\n",
            "Iteration>825, dA[0.074,0.014] dB[0.018,0.003] g[13.917,13.062]\n",
            "Iteration>826, dA[0.206,0.153] dB[0.003,0.007] g[12.920,11.485]\n",
            "Iteration>827, dA[0.058,0.299] dB[0.007,0.004] g[15.217,13.792]\n",
            "Iteration>828, dA[0.343,0.037] dB[0.007,0.025] g[14.604,14.247]\n",
            "Iteration>829, dA[0.262,0.496] dB[0.020,0.003] g[11.865,9.768]\n",
            "Iteration>830, dA[0.176,0.044] dB[0.010,0.010] g[12.404,10.940]\n",
            "Iteration>831, dA[0.142,0.235] dB[0.005,0.002] g[11.866,9.106]\n",
            "Iteration>832, dA[0.016,0.137] dB[0.006,0.007] g[15.706,14.929]\n",
            "Iteration>833, dA[0.359,0.067] dB[0.012,0.007] g[13.009,11.658]\n",
            "Iteration>834, dA[0.032,0.258] dB[0.008,0.010] g[14.308,12.628]\n",
            "Iteration>835, dA[0.128,0.020] dB[0.011,0.004] g[15.385,15.037]\n",
            "Iteration>836, dA[0.088,0.067] dB[0.007,0.016] g[14.233,13.898]\n",
            "Iteration>837, dA[0.014,0.087] dB[0.015,0.011] g[15.245,14.651]\n",
            "Iteration>838, dA[0.020,0.050] dB[0.019,0.007] g[15.520,15.018]\n",
            "Iteration>839, dA[0.192,0.130] dB[0.013,0.007] g[12.685,11.692]\n",
            "Iteration>840, dA[0.090,0.226] dB[0.008,0.005] g[12.370,10.391]\n",
            "Iteration>841, dA[0.022,0.034] dB[0.013,0.010] g[14.649,14.388]\n",
            "Iteration>842, dA[0.059,0.033] dB[0.006,0.003] g[13.823,13.055]\n",
            "Iteration>843, dA[0.036,0.019] dB[0.003,0.008] g[15.416,14.336]\n",
            "Iteration>844, dA[0.107,0.059] dB[0.005,0.002] g[12.840,10.641]\n",
            "Iteration>845, dA[0.020,0.146] dB[0.005,0.002] g[14.016,12.833]\n",
            "Iteration>846, dA[0.108,0.035] dB[0.004,0.001] g[13.890,12.614]\n",
            "Iteration>847, dA[0.043,0.129] dB[0.009,0.014] g[12.219,10.555]\n",
            "Iteration>848, dA[0.025,0.077] dB[0.005,0.003] g[13.924,12.803]\n",
            "Iteration>849, dA[0.056,0.050] dB[0.005,0.002] g[13.004,11.471]\n",
            "Iteration>850, dA[0.031,0.062] dB[0.005,0.008] g[12.815,10.925]\n",
            "Iteration>851, dA[0.080,0.011] dB[0.009,0.014] g[12.207,10.746]\n",
            "Iteration>852, dA[0.025,0.042] dB[0.034,0.004] g[14.915,13.796]\n",
            "Iteration>853, dA[0.338,0.523] dB[0.019,0.017] g[11.355,9.971]\n",
            "Iteration>854, dA[0.126,0.036] dB[0.007,0.008] g[11.715,10.152]\n",
            "Iteration>855, dA[0.031,0.019] dB[0.016,0.003] g[13.391,11.884]\n",
            "Iteration>856, dA[0.007,0.019] dB[0.008,0.003] g[13.560,12.467]\n",
            "Iteration>857, dA[0.024,0.026] dB[0.007,0.006] g[14.034,12.638]\n",
            "Iteration>858, dA[0.056,0.009] dB[0.015,0.033] g[14.685,13.418]\n",
            "Iteration>859, dA[0.025,0.038] dB[0.026,0.011] g[14.996,14.281]\n",
            "Iteration>860, dA[0.065,0.191] dB[0.062,0.035] g[12.572,10.891]\n",
            "Iteration>861, dA[0.010,0.025] dB[0.011,0.038] g[15.240,14.609]\n",
            "Iteration>862, dA[0.165,0.032] dB[0.036,0.011] g[13.152,11.780]\n",
            "Iteration>863, dA[0.025,0.077] dB[0.012,0.020] g[12.971,11.277]\n",
            "Iteration>864, dA[0.023,0.017] dB[0.029,0.008] g[14.644,13.919]\n",
            "Iteration>865, dA[0.121,0.084] dB[0.022,0.019] g[13.227,12.140]\n",
            "Iteration>866, dA[0.180,0.450] dB[0.007,0.002] g[12.549,11.077]\n",
            "Iteration>867, dA[0.013,0.115] dB[0.004,0.005] g[14.179,13.546]\n",
            "Iteration>868, dA[0.100,0.129] dB[0.004,0.010] g[13.824,12.575]\n",
            "Iteration>869, dA[0.166,0.200] dB[0.005,0.006] g[12.311,10.448]\n",
            "Iteration>870, dA[0.016,0.014] dB[0.018,0.016] g[13.810,12.707]\n",
            "Iteration>871, dA[0.114,0.098] dB[0.009,0.007] g[12.601,10.965]\n",
            "Iteration>872, dA[0.012,0.030] dB[0.023,0.018] g[14.471,13.959]\n",
            "Iteration>873, dA[0.022,0.047] dB[0.007,0.013] g[14.804,14.126]\n",
            "Iteration>874, dA[0.035,0.069] dB[0.031,0.011] g[13.598,12.048]\n",
            "Iteration>875, dA[0.064,0.014] dB[0.004,0.005] g[12.473,10.886]\n",
            "Iteration>876, dA[0.013,0.086] dB[0.006,0.003] g[12.038,9.966]\n",
            "Iteration>877, dA[0.007,0.054] dB[0.002,0.001] g[14.933,14.876]\n",
            "Iteration>878, dA[0.007,0.006] dB[0.001,0.002] g[14.585,14.402]\n",
            "Iteration>879, dA[0.024,0.010] dB[0.002,0.004] g[14.822,14.410]\n",
            "Iteration>880, dA[0.382,0.174] dB[0.005,0.003] g[12.222,10.674]\n",
            "Iteration>881, dA[0.047,0.054] dB[0.006,0.010] g[13.571,11.815]\n",
            "Iteration>882, dA[0.194,0.070] dB[0.008,0.033] g[12.544,11.168]\n",
            "Iteration>883, dA[0.020,0.025] dB[0.048,0.002] g[13.858,12.496]\n",
            "Iteration>884, dA[0.015,0.017] dB[0.056,0.080] g[13.999,13.021]\n",
            "Iteration>885, dA[0.006,0.018] dB[0.049,0.017] g[14.157,12.697]\n",
            "Iteration>886, dA[0.022,0.006] dB[0.060,0.030] g[15.480,15.450]\n",
            "Iteration>887, dA[0.174,0.269] dB[0.017,0.009] g[11.849,10.532]\n",
            "Iteration>888, dA[0.024,0.020] dB[0.012,0.004] g[13.668,12.328]\n",
            "Iteration>889, dA[0.032,0.055] dB[0.005,0.002] g[16.378,16.373]\n",
            "Iteration>890, dA[0.068,0.077] dB[0.010,0.012] g[12.772,11.431]\n",
            "Iteration>891, dA[0.012,0.031] dB[0.004,0.011] g[13.675,12.250]\n",
            "Iteration>892, dA[0.013,0.036] dB[0.024,0.006] g[14.902,14.958]\n",
            "Iteration>893, dA[0.056,0.068] dB[0.006,0.002] g[13.059,12.137]\n",
            "Iteration>894, dA[0.030,0.038] dB[0.004,0.003] g[12.044,10.182]\n",
            "Iteration>895, dA[0.010,0.044] dB[0.004,0.002] g[13.714,12.738]\n",
            "Iteration>896, dA[0.036,0.103] dB[0.003,0.002] g[12.968,11.706]\n",
            "Iteration>897, dA[0.009,0.007] dB[0.002,0.001] g[13.217,11.417]\n",
            "Iteration>898, dA[0.011,0.035] dB[0.004,0.003] g[14.097,13.162]\n",
            "Iteration>899, dA[0.005,0.013] dB[0.002,0.002] g[14.909,14.244]\n",
            "Iteration>900, dA[0.034,0.019] dB[0.004,0.004] g[12.515,10.811]\n",
            "Iteration>901, dA[0.014,0.017] dB[0.002,0.005] g[12.183,10.010]\n",
            "Iteration>902, dA[0.044,0.015] dB[0.008,0.002] g[15.131,14.247]\n",
            "Iteration>903, dA[0.067,0.023] dB[0.007,0.003] g[13.265,12.126]\n",
            "Iteration>904, dA[0.033,0.007] dB[0.003,0.005] g[14.149,12.596]\n",
            "Iteration>905, dA[0.003,0.005] dB[0.007,0.003] g[13.548,12.154]\n",
            "Iteration>906, dA[0.006,0.039] dB[0.004,0.002] g[13.555,12.222]\n",
            "Iteration>907, dA[0.115,0.035] dB[0.006,0.008] g[12.931,11.465]\n",
            "Iteration>908, dA[0.060,0.189] dB[0.002,0.012] g[13.524,11.777]\n",
            "Iteration>909, dA[0.298,0.049] dB[0.016,0.004] g[12.268,11.578]\n",
            "Iteration>910, dA[0.043,0.042] dB[0.016,0.007] g[13.323,11.062]\n",
            "Iteration>911, dA[0.047,0.133] dB[0.004,0.002] g[12.421,10.296]\n",
            "Iteration>912, dA[0.105,0.061] dB[0.002,0.006] g[11.825,10.186]\n",
            "Iteration>913, dA[0.036,0.086] dB[0.003,0.015] g[14.166,12.706]\n",
            "Iteration>914, dA[0.087,0.051] dB[0.026,0.008] g[13.893,13.167]\n",
            "Iteration>915, dA[0.093,0.187] dB[0.006,0.017] g[13.324,11.617]\n",
            "Iteration>916, dA[0.113,0.045] dB[0.023,0.009] g[12.404,11.190]\n",
            "Iteration>917, dA[0.030,0.021] dB[0.005,0.007] g[13.489,12.424]\n",
            "Iteration>918, dA[0.010,0.032] dB[0.003,0.005] g[14.869,14.207]\n",
            "Iteration>919, dA[0.027,0.101] dB[0.022,0.014] g[12.765,11.266]\n",
            "Iteration>920, dA[0.058,0.031] dB[0.005,0.006] g[12.249,10.771]\n",
            "Iteration>921, dA[0.057,0.035] dB[0.005,0.004] g[13.649,11.978]\n",
            "Iteration>922, dA[0.158,0.122] dB[0.004,0.003] g[11.968,10.236]\n",
            "Iteration>923, dA[0.035,0.144] dB[0.003,0.001] g[13.093,11.265]\n",
            "Iteration>924, dA[0.468,0.109] dB[0.005,0.004] g[12.390,11.364]\n",
            "Iteration>925, dA[0.042,0.071] dB[0.005,0.006] g[12.434,10.859]\n",
            "Iteration>926, dA[0.045,0.028] dB[0.013,0.017] g[13.146,11.862]\n",
            "Iteration>927, dA[0.043,0.144] dB[0.011,0.006] g[12.367,10.717]\n",
            "Iteration>928, dA[0.085,0.025] dB[0.023,0.047] g[13.640,12.776]\n",
            "Iteration>929, dA[0.025,0.119] dB[0.046,0.009] g[12.669,10.975]\n",
            "Iteration>930, dA[0.022,0.047] dB[0.053,0.098] g[15.107,14.794]\n",
            "Iteration>931, dA[0.305,0.508] dB[0.059,0.006] g[12.456,11.212]\n",
            "Iteration>932, dA[0.130,1.018] dB[0.017,0.030] g[13.628,13.699]\n",
            "Iteration>933, dA[0.393,2.959] dB[0.021,0.009] g[12.705,13.926]\n",
            "Iteration>934, dA[0.473,0.241] dB[0.010,0.037] g[14.647,14.317]\n",
            "Iteration>935, dA[0.385,0.363] dB[0.030,0.002] g[14.592,14.572]\n",
            "Iteration>936, dA[0.052,0.103] dB[0.013,0.019] g[15.975,15.941]\n",
            "Iteration>937, dA[0.246,0.216] dB[0.024,0.016] g[12.448,11.176]\n",
            "Iteration>938, dA[0.028,0.105] dB[0.011,0.007] g[13.505,11.977]\n",
            "Iteration>939, dA[0.063,0.076] dB[0.019,0.009] g[13.706,12.211]\n",
            "Iteration>940, dA[0.113,0.081] dB[0.009,0.004] g[13.024,11.507]\n",
            "Iteration>941, dA[0.017,0.048] dB[0.007,0.003] g[15.037,14.095]\n",
            "Iteration>942, dA[0.043,0.019] dB[0.008,0.002] g[13.448,12.756]\n",
            "Iteration>943, dA[0.028,0.039] dB[0.007,0.003] g[14.427,13.902]\n",
            "Iteration>944, dA[0.025,0.031] dB[0.003,0.007] g[13.983,12.733]\n",
            "Iteration>945, dA[0.144,0.153] dB[0.008,0.032] g[12.052,9.873]\n",
            "Iteration>946, dA[0.020,0.059] dB[0.067,0.012] g[13.497,11.095]\n",
            "Iteration>947, dA[0.091,0.043] dB[0.041,0.055] g[12.905,12.614]\n",
            "Iteration>948, dA[0.087,0.068] dB[0.035,0.019] g[13.168,11.721]\n",
            "Iteration>949, dA[0.030,0.058] dB[0.031,0.018] g[12.494,10.734]\n",
            "Iteration>950, dA[0.161,0.032] dB[0.006,0.014] g[12.325,11.171]\n",
            "Iteration>951, dA[0.019,0.084] dB[0.005,0.014] g[13.303,11.806]\n",
            "Iteration>952, dA[0.083,0.035] dB[0.012,0.004] g[12.465,11.043]\n",
            "Iteration>953, dA[0.012,0.145] dB[0.009,0.005] g[13.626,12.029]\n",
            "Iteration>954, dA[0.105,0.025] dB[0.005,0.002] g[13.194,12.457]\n",
            "Iteration>955, dA[0.010,0.032] dB[0.006,0.003] g[13.111,11.638]\n",
            "Iteration>956, dA[0.141,0.327] dB[0.010,0.005] g[12.340,10.745]\n",
            "Iteration>957, dA[0.045,0.066] dB[0.004,0.016] g[13.703,12.977]\n",
            "Iteration>958, dA[0.311,0.190] dB[0.019,0.004] g[11.729,10.724]\n",
            "Iteration>959, dA[0.020,0.069] dB[0.023,0.028] g[12.160,10.134]\n",
            "Iteration>960, dA[0.249,0.181] dB[0.027,0.012] g[11.609,9.888]\n",
            "Iteration>961, dA[0.017,0.031] dB[0.040,0.037] g[14.860,14.293]\n",
            "Iteration>962, dA[0.156,0.074] dB[0.028,0.006] g[13.363,12.354]\n",
            "Iteration>963, dA[0.046,0.120] dB[0.035,0.011] g[12.144,9.777]\n",
            "Iteration>964, dA[0.096,0.048] dB[0.008,0.005] g[11.839,10.400]\n",
            "Iteration>965, dA[0.086,0.081] dB[0.012,0.004] g[12.040,10.495]\n",
            "Iteration>966, dA[0.020,0.034] dB[0.007,0.012] g[13.735,12.629]\n",
            "Iteration>967, dA[0.169,0.113] dB[0.006,0.006] g[12.017,10.626]\n",
            "Iteration>968, dA[0.044,0.044] dB[0.019,0.012] g[13.037,11.637]\n",
            "Iteration>969, dA[0.234,0.113] dB[0.032,0.066] g[11.243,9.694]\n",
            "Iteration>970, dA[0.081,0.032] dB[0.039,0.013] g[11.135,9.758]\n",
            "Iteration>971, dA[0.032,0.060] dB[0.040,0.023] g[11.675,9.238]\n",
            "Iteration>972, dA[0.007,0.019] dB[0.015,0.004] g[12.411,11.361]\n",
            "Iteration>973, dA[0.020,0.007] dB[0.010,0.016] g[13.522,12.436]\n",
            "Iteration>974, dA[0.051,0.041] dB[0.017,0.005] g[14.225,13.098]\n",
            "Iteration>975, dA[0.032,0.026] dB[0.005,0.005] g[12.765,11.632]\n",
            "Iteration>976, dA[0.176,0.187] dB[0.010,0.003] g[10.889,8.903]\n",
            "Iteration>977, dA[0.009,0.070] dB[0.003,0.001] g[14.570,13.532]\n",
            "Iteration>978, dA[0.132,0.026] dB[0.009,0.011] g[13.076,12.857]\n",
            "Iteration>979, dA[0.014,0.051] dB[0.009,0.001] g[12.877,11.804]\n",
            "Iteration>980, dA[0.180,0.190] dB[0.007,0.007] g[11.207,9.155]\n",
            "Iteration>981, dA[0.119,0.032] dB[0.014,0.016] g[10.862,8.772]\n",
            "Iteration>982, dA[0.035,0.023] dB[0.005,0.022] g[13.150,11.885]\n",
            "Iteration>983, dA[0.187,0.117] dB[0.034,0.014] g[11.552,9.678]\n",
            "Iteration>984, dA[0.106,0.044] dB[0.014,0.028] g[11.615,9.453]\n",
            "Iteration>985, dA[0.029,0.016] dB[0.008,0.011] g[15.367,15.800]\n",
            "Iteration>986, dA[0.108,0.053] dB[0.035,0.031] g[12.878,12.239]\n",
            "Iteration>987, dA[0.006,0.023] dB[0.011,0.014] g[12.938,11.763]\n",
            "Iteration>988, dA[0.095,0.070] dB[0.020,0.005] g[11.594,9.656]\n",
            "Iteration>989, dA[0.043,0.084] dB[0.008,0.010] g[14.232,13.038]\n",
            "Iteration>990, dA[0.041,0.088] dB[0.009,0.002] g[14.159,13.504]\n",
            "Iteration>991, dA[0.193,0.089] dB[0.004,0.009] g[11.156,8.929]\n",
            "Iteration>992, dA[0.085,0.060] dB[0.015,0.013] g[11.715,10.101]\n",
            "Iteration>993, dA[0.026,0.051] dB[0.010,0.025] g[13.372,12.559]\n",
            "Iteration>994, dA[0.038,0.009] dB[0.028,0.009] g[14.109,13.146]\n",
            "Iteration>995, dA[0.026,0.191] dB[0.018,0.012] g[11.554,9.534]\n",
            "Iteration>996, dA[0.340,0.105] dB[0.010,0.018] g[12.043,11.323]\n",
            "Iteration>997, dA[0.013,0.031] dB[0.011,0.006] g[13.116,12.291]\n",
            "Iteration>998, dA[0.068,0.044] dB[0.009,0.010] g[12.154,10.773]\n",
            "Iteration>999, dA[0.061,0.067] dB[0.008,0.006] g[11.607,9.693]\n",
            "Iteration>1000, dA[0.028,0.035] dB[0.014,0.005] g[12.082,10.765]\n",
            "Iteration>1001, dA[0.067,0.209] dB[0.005,0.006] g[10.778,8.414]\n",
            "Iteration>1002, dA[0.153,0.051] dB[0.005,0.008] g[12.304,11.130]\n",
            "Iteration>1003, dA[0.039,0.073] dB[0.007,0.004] g[10.490,8.505]\n",
            "Iteration>1004, dA[0.020,0.013] dB[0.009,0.007] g[11.634,9.819]\n",
            "Iteration>1005, dA[0.047,0.065] dB[0.011,0.008] g[10.500,8.599]\n",
            "Iteration>1006, dA[0.020,0.025] dB[0.005,0.016] g[12.583,10.901]\n",
            "Iteration>1007, dA[0.020,0.017] dB[0.015,0.005] g[10.825,8.502]\n",
            "Iteration>1008, dA[0.032,0.012] dB[0.007,0.002] g[12.636,10.732]\n",
            "Iteration>1009, dA[0.033,0.030] dB[0.021,0.045] g[11.563,10.151]\n",
            "Iteration>1010, dA[0.014,0.024] dB[0.029,0.007] g[12.605,10.447]\n",
            "Iteration>1011, dA[0.061,0.065] dB[0.028,0.036] g[11.687,10.311]\n",
            "Iteration>1012, dA[0.011,0.022] dB[0.014,0.010] g[13.594,12.531]\n",
            "Iteration>1013, dA[0.008,0.012] dB[0.019,0.008] g[14.038,13.279]\n",
            "Iteration>1014, dA[0.027,0.013] dB[0.004,0.004] g[11.555,9.955]\n",
            "Iteration>1015, dA[0.023,0.033] dB[0.005,0.003] g[13.139,11.825]\n",
            "Iteration>1016, dA[0.075,0.035] dB[0.005,0.008] g[11.220,9.588]\n",
            "Iteration>1017, dA[0.012,0.065] dB[0.007,0.005] g[14.955,14.102]\n",
            "Iteration>1018, dA[0.085,0.027] dB[0.003,0.023] g[12.156,11.758]\n",
            "Iteration>1019, dA[0.035,0.035] dB[0.020,0.006] g[14.995,14.513]\n",
            "Iteration>1020, dA[0.035,0.033] dB[0.028,0.025] g[13.501,12.109]\n",
            "Iteration>1021, dA[0.062,0.049] dB[0.017,0.013] g[12.657,11.628]\n",
            "Iteration>1022, dA[0.009,0.018] dB[0.011,0.009] g[13.326,11.992]\n",
            "Iteration>1023, dA[0.016,0.062] dB[0.029,0.033] g[11.994,10.626]\n",
            "Iteration>1024, dA[0.046,0.018] dB[0.012,0.013] g[12.800,12.134]\n",
            "Iteration>1025, dA[0.137,0.085] dB[0.037,0.031] g[11.963,10.248]\n",
            "Iteration>1026, dA[0.077,0.067] dB[0.013,0.009] g[11.496,9.526]\n",
            "Iteration>1027, dA[0.012,0.035] dB[0.018,0.017] g[13.523,12.078]\n",
            "Iteration>1028, dA[0.012,0.026] dB[0.010,0.037] g[13.538,12.800]\n",
            "Iteration>1029, dA[0.005,0.016] dB[0.030,0.007] g[13.550,12.365]\n",
            "Iteration>1030, dA[0.133,0.152] dB[0.046,0.105] g[12.636,13.037]\n",
            "Iteration>1031, dA[0.064,0.014] dB[0.067,0.009] g[14.067,13.161]\n",
            "Iteration>1032, dA[0.032,0.018] dB[0.028,0.067] g[13.233,12.860]\n",
            "Iteration>1033, dA[0.016,0.018] dB[0.048,0.011] g[12.422,10.775]\n",
            "Iteration>1034, dA[0.011,0.022] dB[0.004,0.011] g[13.131,11.831]\n",
            "Iteration>1035, dA[0.012,0.062] dB[0.007,0.002] g[18.319,10.227]\n",
            "Iteration>1036, dA[0.052,0.014] dB[0.006,0.002] g[12.648,11.334]\n",
            "Iteration>1037, dA[0.012,0.018] dB[0.004,0.003] g[12.457,11.028]\n",
            "Iteration>1038, dA[0.010,0.020] dB[0.009,0.005] g[13.086,11.023]\n",
            "Iteration>1039, dA[0.005,0.033] dB[0.010,0.014] g[14.383,13.734]\n",
            "Iteration>1040, dA[0.028,0.023] dB[0.011,0.006] g[12.742,11.271]\n",
            "Iteration>1041, dA[0.023,0.026] dB[0.016,0.016] g[13.025,11.217]\n",
            "Iteration>1042, dA[0.039,0.031] dB[0.012,0.003] g[13.925,12.801]\n",
            "Iteration>1043, dA[0.020,0.175] dB[0.013,0.036] g[11.762,10.568]\n",
            "Iteration>1044, dA[0.047,0.048] dB[0.020,0.010] g[12.844,11.075]\n",
            "Iteration>1045, dA[0.016,0.025] dB[0.043,0.036] g[14.826,14.874]\n",
            "Iteration>1046, dA[0.006,0.022] dB[0.010,0.008] g[12.908,11.213]\n",
            "Iteration>1047, dA[0.018,0.007] dB[0.029,0.039] g[12.284,10.506]\n",
            "Iteration>1048, dA[0.005,0.014] dB[0.024,0.007] g[13.096,11.832]\n",
            "Iteration>1049, dA[0.011,0.019] dB[0.012,0.015] g[14.111,13.083]\n",
            "Iteration>1050, dA[0.020,0.006] dB[0.015,0.007] g[14.754,14.160]\n",
            "Iteration>1051, dA[0.005,0.009] dB[0.008,0.002] g[13.845,12.990]\n",
            "Iteration>1052, dA[0.015,0.008] dB[0.007,0.013] g[12.889,11.439]\n",
            "Iteration>1053, dA[0.021,0.062] dB[0.007,0.001] g[13.622,12.631]\n",
            "Iteration>1054, dA[0.023,0.050] dB[0.008,0.003] g[13.195,11.987]\n",
            "Iteration>1055, dA[0.010,0.016] dB[0.004,0.001] g[14.168,13.646]\n",
            "Iteration>1056, dA[0.022,0.049] dB[0.004,0.007] g[13.575,12.141]\n",
            "Iteration>1057, dA[0.008,0.022] dB[0.008,0.001] g[13.178,12.119]\n",
            "Iteration>1058, dA[0.228,0.287] dB[0.004,0.003] g[11.525,10.667]\n",
            "Iteration>1059, dA[0.014,0.071] dB[0.005,0.005] g[12.417,10.399]\n",
            "Iteration>1060, dA[0.095,0.195] dB[0.010,0.007] g[14.432,14.552]\n",
            "Iteration>1061, dA[0.086,0.294] dB[0.004,0.017] g[12.051,10.001]\n",
            "Iteration>1062, dA[0.041,0.105] dB[0.019,0.006] g[14.525,13.577]\n",
            "Iteration>1063, dA[0.157,0.046] dB[0.004,0.008] g[12.891,11.954]\n",
            "Iteration>1064, dA[0.030,0.124] dB[0.015,0.009] g[11.837,9.855]\n",
            "Iteration>1065, dA[0.074,0.008] dB[0.022,0.002] g[11.728,9.986]\n",
            "Iteration>1066, dA[0.024,0.042] dB[0.012,0.003] g[12.686,11.124]\n",
            "Iteration>1067, dA[0.013,0.012] dB[0.008,0.004] g[12.954,11.896]\n",
            "Iteration>1068, dA[0.016,0.007] dB[0.003,0.004] g[12.823,12.156]\n",
            "Iteration>1069, dA[0.020,0.028] dB[0.003,0.001] g[14.652,14.303]\n",
            "Iteration>1070, dA[0.015,0.049] dB[0.001,0.004] g[12.121,11.027]\n",
            "Iteration>1071, dA[0.018,0.021] dB[0.006,0.001] g[13.168,11.989]\n",
            "Iteration>1072, dA[0.016,0.009] dB[0.004,0.003] g[13.003,11.970]\n",
            "Iteration>1073, dA[0.073,0.018] dB[0.005,0.012] g[11.854,10.857]\n",
            "Iteration>1074, dA[0.038,0.027] dB[0.008,0.004] g[12.389,10.438]\n",
            "Iteration>1075, dA[0.020,0.015] dB[0.011,0.008] g[11.822,10.216]\n",
            "Iteration>1076, dA[0.039,0.016] dB[0.018,0.011] g[12.283,11.235]\n",
            "Iteration>1077, dA[0.028,0.012] dB[0.017,0.003] g[12.245,10.843]\n",
            "Iteration>1078, dA[0.023,0.046] dB[0.007,0.008] g[13.171,12.076]\n",
            "Iteration>1079, dA[0.024,0.027] dB[0.007,0.003] g[12.546,11.342]\n",
            "Iteration>1080, dA[0.027,0.002] dB[0.007,0.024] g[11.644,10.230]\n",
            "Iteration>1081, dA[0.012,0.013] dB[0.024,0.008] g[12.352,10.618]\n",
            "Iteration>1082, dA[0.038,0.002] dB[0.039,0.017] g[11.670,10.108]\n",
            "Iteration>1083, dA[0.012,0.042] dB[0.012,0.030] g[13.012,11.342]\n",
            "Iteration>1084, dA[0.031,0.015] dB[0.025,0.024] g[12.380,10.092]\n",
            "Iteration>1085, dA[0.020,0.005] dB[0.027,0.058] g[12.450,11.792]\n",
            "Iteration>1086, dA[0.016,0.007] dB[0.012,0.022] g[13.475,12.224]\n",
            "Iteration>1087, dA[0.021,0.032] dB[0.022,0.021] g[13.904,13.012]\n",
            "Iteration>1088, dA[0.022,0.014] dB[0.010,0.014] g[12.277,10.834]\n",
            "Iteration>1089, dA[0.048,0.029] dB[0.043,0.030] g[12.037,10.250]\n",
            "Iteration>1090, dA[0.211,0.436] dB[0.010,0.010] g[11.989,11.186]\n",
            "Iteration>1091, dA[0.313,0.096] dB[0.027,0.022] g[12.173,11.123]\n",
            "Iteration>1092, dA[0.062,0.190] dB[0.004,0.017] g[13.191,11.330]\n",
            "Iteration>1093, dA[0.154,0.268] dB[0.014,0.008] g[11.401,9.496]\n",
            "Iteration>1094, dA[0.302,0.063] dB[0.018,0.010] g[12.695,11.984]\n",
            "Iteration>1095, dA[0.139,0.119] dB[0.012,0.005] g[11.869,9.968]\n",
            "Iteration>1096, dA[0.042,0.154] dB[0.009,0.005] g[12.438,10.530]\n",
            "Iteration>1097, dA[0.078,0.104] dB[0.015,0.022] g[13.277,11.705]\n",
            "Iteration>1098, dA[0.510,0.100] dB[0.012,0.004] g[11.291,10.226]\n",
            "Iteration>1099, dA[0.117,0.240] dB[0.014,0.006] g[12.006,9.705]\n",
            "Iteration>1100, dA[0.042,0.018] dB[0.005,0.007] g[14.206,13.028]\n",
            "Iteration>1101, dA[0.044,0.025] dB[0.005,0.001] g[14.228,13.133]\n",
            "Iteration>1102, dA[0.261,0.388] dB[0.008,0.018] g[11.496,9.725]\n",
            "Iteration>1103, dA[0.040,0.090] dB[0.006,0.003] g[12.836,10.948]\n",
            "Iteration>1104, dA[0.105,0.141] dB[0.008,0.008] g[12.952,11.877]\n",
            "Iteration>1105, dA[0.040,0.024] dB[0.007,0.004] g[15.162,15.174]\n",
            "Iteration>1106, dA[0.415,0.133] dB[0.002,0.003] g[11.678,9.900]\n",
            "Iteration>1107, dA[0.094,0.227] dB[0.016,0.012] g[14.665,13.987]\n",
            "Iteration>1108, dA[0.471,0.058] dB[0.020,0.028] g[12.719,11.313]\n",
            "Iteration>1109, dA[0.154,0.091] dB[0.028,0.013] g[12.514,11.439]\n",
            "Iteration>1110, dA[0.009,0.123] dB[0.009,0.040] g[13.348,11.502]\n",
            "Iteration>1111, dA[0.050,0.032] dB[0.067,0.030] g[14.949,14.678]\n",
            "Iteration>1112, dA[0.278,0.240] dB[0.018,0.053] g[12.286,11.116]\n",
            "Iteration>1113, dA[0.329,0.145] dB[0.035,0.008] g[10.509,9.413]\n",
            "Iteration>1114, dA[0.016,0.190] dB[0.013,0.041] g[12.647,10.863]\n",
            "Iteration>1115, dA[0.312,0.097] dB[0.046,0.015] g[12.196,10.953]\n",
            "Iteration>1116, dA[0.046,0.104] dB[0.005,0.011] g[12.794,11.504]\n",
            "Iteration>1117, dA[0.174,0.360] dB[0.013,0.004] g[10.658,9.319]\n",
            "Iteration>1118, dA[0.357,0.028] dB[0.008,0.005] g[12.611,11.525]\n",
            "Iteration>1119, dA[0.025,0.052] dB[0.003,0.004] g[14.027,13.116]\n",
            "Iteration>1120, dA[0.048,0.174] dB[0.005,0.003] g[12.396,10.568]\n",
            "Iteration>1121, dA[0.129,0.052] dB[0.004,0.029] g[12.413,11.500]\n",
            "Iteration>1122, dA[0.059,0.165] dB[0.019,0.013] g[12.344,10.783]\n",
            "Iteration>1123, dA[0.084,0.037] dB[0.042,0.018] g[13.604,12.516]\n",
            "Iteration>1124, dA[0.080,0.073] dB[0.006,0.003] g[13.157,12.126]\n",
            "Iteration>1125, dA[0.014,0.018] dB[0.006,0.003] g[12.956,11.553]\n",
            "Iteration>1126, dA[0.016,0.021] dB[0.007,0.014] g[12.799,12.014]\n",
            "Iteration>1127, dA[0.007,0.013] dB[0.009,0.022] g[13.714,13.118]\n",
            "Iteration>1128, dA[0.067,0.070] dB[0.006,0.022] g[12.372,11.034]\n",
            "Iteration>1129, dA[0.056,0.125] dB[0.010,0.012] g[12.023,10.290]\n",
            "Iteration>1130, dA[0.501,0.253] dB[0.009,0.023] g[10.745,9.733]\n",
            "Iteration>1131, dA[0.039,0.037] dB[0.055,0.043] g[13.975,13.124]\n",
            "Iteration>1132, dA[0.220,0.131] dB[0.006,0.032] g[12.298,11.496]\n",
            "Iteration>1133, dA[0.027,0.030] dB[0.042,0.018] g[12.728,11.053]\n",
            "Iteration>1134, dA[0.101,0.267] dB[0.004,0.006] g[10.585,8.192]\n",
            "Iteration>1135, dA[0.190,0.132] dB[0.008,0.004] g[11.449,9.664]\n",
            "Iteration>1136, dA[0.124,0.088] dB[0.006,0.004] g[11.571,9.866]\n",
            "Iteration>1137, dA[0.046,0.010] dB[0.008,0.012] g[13.090,12.024]\n",
            "Iteration>1138, dA[0.019,0.080] dB[0.009,0.007] g[11.467,9.095]\n",
            "Iteration>1139, dA[0.023,0.023] dB[0.005,0.005] g[13.186,11.749]\n",
            "Iteration>1140, dA[0.102,0.121] dB[0.007,0.005] g[11.857,10.048]\n",
            "Iteration>1141, dA[0.011,0.016] dB[0.010,0.010] g[12.409,10.583]\n",
            "Iteration>1142, dA[0.027,0.084] dB[0.005,0.001] g[13.414,12.471]\n",
            "Iteration>1143, dA[0.261,0.321] dB[0.005,0.004] g[11.342,9.629]\n",
            "Iteration>1144, dA[0.089,0.034] dB[0.006,0.003] g[12.772,11.889]\n",
            "Iteration>1145, dA[0.016,0.051] dB[0.002,0.001] g[12.852,11.725]\n",
            "Iteration>1146, dA[0.172,0.149] dB[0.004,0.007] g[11.720,10.688]\n",
            "Iteration>1147, dA[0.335,0.266] dB[0.004,0.007] g[11.452,9.603]\n",
            "Iteration>1148, dA[0.044,0.154] dB[0.009,0.003] g[10.683,7.847]\n",
            "Iteration>1149, dA[0.289,0.101] dB[0.003,0.001] g[10.698,8.995]\n",
            "Iteration>1150, dA[0.026,0.045] dB[0.002,0.001] g[13.260,11.527]\n",
            "Iteration>1151, dA[0.222,0.121] dB[0.003,0.004] g[12.663,12.097]\n",
            "Iteration>1152, dA[0.034,0.063] dB[0.003,0.001] g[12.648,11.375]\n",
            "Iteration>1153, dA[0.111,0.032] dB[0.004,0.002] g[12.481,11.546]\n",
            "Iteration>1154, dA[0.024,0.046] dB[0.007,0.009] g[12.616,11.102]\n",
            "Iteration>1155, dA[0.037,0.037] dB[0.004,0.011] g[12.294,11.151]\n",
            "Iteration>1156, dA[0.055,0.018] dB[0.020,0.012] g[12.691,11.751]\n",
            "Iteration>1157, dA[0.010,0.073] dB[0.006,0.005] g[12.427,10.928]\n",
            "Iteration>1158, dA[0.088,0.057] dB[0.014,0.017] g[13.093,12.737]\n",
            "Iteration>1159, dA[0.048,0.069] dB[0.018,0.004] g[12.364,11.195]\n",
            "Iteration>1160, dA[0.075,0.033] dB[0.011,0.006] g[12.580,12.035]\n",
            "Iteration>1161, dA[0.422,0.420] dB[0.007,0.027] g[10.365,8.940]\n",
            "Iteration>1162, dA[0.162,0.089] dB[0.014,0.008] g[11.557,9.694]\n",
            "Iteration>1163, dA[0.270,0.148] dB[0.019,0.021] g[11.797,10.041]\n",
            "Iteration>1164, dA[0.027,0.074] dB[0.006,0.007] g[12.087,10.273]\n",
            "Iteration>1165, dA[0.041,0.019] dB[0.023,0.008] g[14.304,13.419]\n",
            "Iteration>1166, dA[0.295,0.045] dB[0.005,0.019] g[11.512,9.828]\n",
            "Iteration>1167, dA[0.049,0.113] dB[0.004,0.007] g[13.042,11.435]\n",
            "Iteration>1168, dA[0.332,0.101] dB[0.004,0.008] g[10.806,8.782]\n",
            "Iteration>1169, dA[0.079,0.162] dB[0.022,0.008] g[11.724,9.788]\n",
            "Iteration>1170, dA[0.035,0.017] dB[0.012,0.039] g[12.854,11.503]\n",
            "Iteration>1171, dA[0.008,0.030] dB[0.033,0.010] g[14.160,13.372]\n",
            "Iteration>1172, dA[0.197,0.161] dB[0.002,0.019] g[11.456,9.728]\n",
            "Iteration>1173, dA[0.043,0.079] dB[0.030,0.009] g[11.906,10.163]\n",
            "Iteration>1174, dA[0.038,0.077] dB[0.007,0.017] g[13.216,12.562]\n",
            "Iteration>1175, dA[0.012,0.023] dB[0.024,0.011] g[14.095,13.496]\n",
            "Iteration>1176, dA[0.063,0.037] dB[0.007,0.003] g[13.211,12.379]\n",
            "Iteration>1177, dA[0.017,0.017] dB[0.010,0.004] g[12.333,11.427]\n",
            "Iteration>1178, dA[0.267,0.396] dB[0.005,0.006] g[11.081,9.184]\n",
            "Iteration>1179, dA[0.017,0.042] dB[0.007,0.004] g[13.830,13.622]\n",
            "Iteration>1180, dA[0.096,0.030] dB[0.012,0.018] g[13.994,13.586]\n",
            "Iteration>1181, dA[0.070,0.127] dB[0.010,0.010] g[11.849,10.857]\n",
            "Iteration>1182, dA[0.114,0.074] dB[0.028,0.005] g[11.465,9.574]\n",
            "Iteration>1183, dA[0.009,0.041] dB[0.010,0.006] g[12.164,10.499]\n",
            "Iteration>1184, dA[0.078,0.013] dB[0.004,0.003] g[12.010,10.745]\n",
            "Iteration>1185, dA[0.025,0.017] dB[0.003,0.006] g[12.833,11.378]\n",
            "Iteration>1186, dA[0.077,0.062] dB[0.004,0.013] g[11.738,10.325]\n",
            "Iteration>1187, dA[0.024,0.035] dB[0.025,0.007] g[11.509,9.251]\n",
            "Iteration>1188, dA[0.009,0.028] dB[0.029,0.036] g[12.580,11.006]\n",
            "Iteration>1189, dA[0.043,0.018] dB[0.022,0.006] g[11.080,9.067]\n",
            "Iteration>1190, dA[0.063,0.031] dB[0.045,0.040] g[12.568,10.728]\n",
            "Iteration>1191, dA[0.415,0.365] dB[0.006,0.033] g[10.993,9.139]\n",
            "Iteration>1192, dA[0.019,0.042] dB[0.032,0.008] g[12.217,10.797]\n",
            "Iteration>1193, dA[0.175,0.098] dB[0.011,0.009] g[12.289,11.456]\n",
            "Iteration>1194, dA[0.023,0.021] dB[0.014,0.006] g[12.763,11.373]\n",
            "Iteration>1195, dA[0.106,0.135] dB[0.010,0.005] g[10.931,9.463]\n",
            "Iteration>1196, dA[0.022,0.037] dB[0.007,0.015] g[13.946,13.253]\n",
            "Iteration>1197, dA[0.031,0.020] dB[0.019,0.003] g[13.647,13.388]\n",
            "Iteration>1198, dA[0.008,0.009] dB[0.015,0.039] g[13.908,13.435]\n",
            "Iteration>1199, dA[0.147,0.112] dB[0.064,0.066] g[11.343,9.752]\n",
            "Iteration>1200, dA[0.032,0.051] dB[0.010,0.079] g[14.375,13.111]\n",
            "WARNING:tensorflow:Compiled the loaded model, but the compiled metrics have yet to be built. `model.compile_metrics` will be empty until you train or evaluate the model.\n",
            "WARNING:tensorflow:Compiled the loaded model, but the compiled metrics have yet to be built. `model.compile_metrics` will be empty until you train or evaluate the model.\n",
            ">Saved: /content/drive/MyDrive/Project/CycleGAN/TrainedModels/C_Highboost_Real_Model/g_model_AtoB_001200.h5 and /content/drive/MyDrive/Project/CycleGAN/TrainedModels/C_Highboost_Real_Model/g_model_BtoA_001200.h5\n",
            "Iteration>1201, dA[0.115,0.097] dB[0.094,0.069] g[12.084,10.453]\n",
            "Iteration>1202, dA[0.032,0.008] dB[0.052,0.012] g[12.376,11.834]\n",
            "Iteration>1203, dA[0.030,0.044] dB[0.033,0.113] g[11.736,10.381]\n",
            "Iteration>1204, dA[0.065,0.125] dB[0.066,0.004] g[10.979,9.315]\n",
            "Iteration>1205, dA[0.514,0.103] dB[0.033,0.024] g[10.494,9.044]\n",
            "Iteration>1206, dA[0.029,0.128] dB[0.011,0.012] g[11.696,9.900]\n",
            "Iteration>1207, dA[0.316,0.086] dB[0.015,0.010] g[11.967,10.548]\n",
            "Iteration>1208, dA[0.055,0.282] dB[0.010,0.011] g[10.032,7.670]\n",
            "Iteration>1209, dA[0.015,0.045] dB[0.011,0.013] g[14.134,13.746]\n",
            "Iteration>1210, dA[0.343,0.049] dB[0.018,0.011] g[12.285,11.353]\n",
            "Iteration>1211, dA[0.060,0.173] dB[0.024,0.064] g[11.622,9.801]\n",
            "Iteration>1212, dA[0.066,0.017] dB[0.048,0.005] g[12.748,11.787]\n",
            "Iteration>1213, dA[0.012,0.009] dB[0.037,0.054] g[13.630,13.430]\n",
            "Iteration>1214, dA[0.011,0.022] dB[0.042,0.015] g[12.363,11.207]\n",
            "Iteration>1215, dA[0.006,0.013] dB[0.013,0.032] g[12.288,11.020]\n",
            "Iteration>1216, dA[0.338,0.219] dB[0.024,0.006] g[11.043,8.985]\n",
            "Iteration>1217, dA[0.062,0.008] dB[0.010,0.010] g[12.270,10.698]\n",
            "Iteration>1218, dA[0.213,0.048] dB[0.027,0.003] g[11.470,10.190]\n",
            "Iteration>1219, dA[0.012,0.192] dB[0.018,0.020] g[12.042,10.044]\n",
            "Iteration>1220, dA[0.143,0.014] dB[0.012,0.011] g[11.560,9.799]\n",
            "Iteration>1221, dA[0.048,0.083] dB[0.007,0.006] g[10.487,7.952]\n",
            "Iteration>1222, dA[0.046,0.100] dB[0.011,0.018] g[14.329,13.792]\n",
            "Iteration>1223, dA[0.186,0.012] dB[0.014,0.005] g[11.831,10.856]\n",
            "Iteration>1224, dA[0.019,0.022] dB[0.006,0.005] g[12.387,11.024]\n",
            "Iteration>1225, dA[0.015,0.145] dB[0.010,0.007] g[10.959,8.800]\n",
            "Iteration>1226, dA[0.076,0.019] dB[0.005,0.013] g[12.093,12.283]\n",
            "Iteration>1227, dA[0.010,0.006] dB[0.014,0.003] g[13.114,12.550]\n",
            "Iteration>1228, dA[0.040,0.016] dB[0.002,0.004] g[11.831,10.630]\n",
            "Iteration>1229, dA[0.073,0.020] dB[0.009,0.008] g[12.228,10.601]\n",
            "Iteration>1230, dA[0.060,0.029] dB[0.010,0.013] g[11.506,10.778]\n",
            "Iteration>1231, dA[0.034,0.081] dB[0.010,0.059] g[12.742,11.321]\n",
            "Iteration>1232, dA[0.048,0.061] dB[0.078,0.017] g[12.942,12.067]\n",
            "Iteration>1233, dA[0.010,0.052] dB[0.031,0.076] g[12.252,11.346]\n",
            "Iteration>1234, dA[0.030,0.013] dB[0.071,0.010] g[13.671,12.928]\n",
            "Iteration>1235, dA[0.079,0.038] dB[0.006,0.013] g[12.131,11.762]\n",
            "Iteration>1236, dA[0.012,0.010] dB[0.004,0.020] g[11.149,9.416]\n",
            "Iteration>1237, dA[0.238,0.460] dB[0.039,0.022] g[10.487,8.620]\n",
            "Iteration>1238, dA[0.192,0.136] dB[0.032,0.092] g[10.720,8.625]\n",
            "Iteration>1239, dA[0.018,0.047] dB[0.091,0.012] g[12.067,10.247]\n",
            "Iteration>1240, dA[0.024,0.031] dB[0.025,0.058] g[12.527,11.969]\n",
            "Iteration>1241, dA[0.005,0.012] dB[0.045,0.016] g[13.538,12.848]\n",
            "Iteration>1242, dA[0.017,0.015] dB[0.007,0.010] g[12.027,11.016]\n",
            "Iteration>1243, dA[0.012,0.012] dB[0.016,0.008] g[12.249,11.156]\n",
            "Iteration>1244, dA[0.008,0.015] dB[0.004,0.005] g[12.035,10.923]\n",
            "Iteration>1245, dA[0.005,0.007] dB[0.003,0.002] g[13.761,13.372]\n",
            "Iteration>1246, dA[0.016,0.037] dB[0.002,0.003] g[12.272,11.391]\n",
            "Iteration>1247, dA[0.310,0.294] dB[0.005,0.001] g[10.427,8.891]\n",
            "Iteration>1248, dA[0.068,0.119] dB[0.003,0.005] g[10.516,8.663]\n",
            "Iteration>1249, dA[0.091,0.019] dB[0.006,0.005] g[12.320,11.548]\n",
            "Iteration>1250, dA[0.092,0.102] dB[0.006,0.003] g[10.819,8.962]\n",
            "Iteration>1251, dA[0.020,0.037] dB[0.008,0.005] g[12.764,11.549]\n",
            "Iteration>1252, dA[0.011,0.007] dB[0.009,0.012] g[12.432,11.726]\n",
            "Iteration>1253, dA[0.053,0.074] dB[0.005,0.009] g[11.265,9.417]\n",
            "Iteration>1254, dA[0.033,0.021] dB[0.005,0.007] g[13.446,12.760]\n",
            "Iteration>1255, dA[0.102,0.143] dB[0.002,0.005] g[11.366,9.730]\n",
            "Iteration>1256, dA[0.008,0.023] dB[0.008,0.002] g[12.241,10.804]\n",
            "Iteration>1257, dA[0.094,0.062] dB[0.008,0.014] g[11.518,10.496]\n",
            "Iteration>1258, dA[0.025,0.046] dB[0.011,0.005] g[11.855,9.873]\n",
            "Iteration>1259, dA[0.068,0.033] dB[0.026,0.009] g[12.767,12.712]\n",
            "Iteration>1260, dA[0.028,0.008] dB[0.004,0.010] g[13.290,12.542]\n",
            "Iteration>1261, dA[0.021,0.012] dB[0.009,0.002] g[11.879,10.928]\n",
            "Iteration>1262, dA[0.023,0.022] dB[0.006,0.001] g[11.996,10.943]\n",
            "Iteration>1263, dA[0.058,0.143] dB[0.009,0.003] g[10.784,9.014]\n",
            "Iteration>1264, dA[0.007,0.060] dB[0.005,0.006] g[13.163,13.208]\n",
            "Iteration>1265, dA[0.085,0.014] dB[0.006,0.001] g[12.791,12.012]\n",
            "Iteration>1266, dA[0.104,0.044] dB[0.006,0.008] g[12.719,11.304]\n",
            "Iteration>1267, dA[0.322,0.195] dB[0.004,0.002] g[10.713,8.885]\n",
            "Iteration>1268, dA[0.035,0.023] dB[0.004,0.003] g[12.399,10.639]\n",
            "Iteration>1269, dA[0.055,0.018] dB[0.007,0.022] g[11.890,10.952]\n",
            "Iteration>1270, dA[0.037,0.034] dB[0.010,0.004] g[12.028,10.350]\n",
            "Iteration>1271, dA[0.017,0.020] dB[0.008,0.005] g[12.357,10.934]\n",
            "Iteration>1272, dA[0.009,0.014] dB[0.003,0.003] g[13.855,13.054]\n",
            "Iteration>1273, dA[0.103,0.067] dB[0.003,0.002] g[10.999,9.553]\n",
            "Iteration>1274, dA[0.009,0.015] dB[0.011,0.006] g[12.432,11.468]\n",
            "Iteration>1275, dA[0.030,0.015] dB[0.007,0.004] g[12.578,11.681]\n",
            "Iteration>1276, dA[0.012,0.011] dB[0.012,0.025] g[12.017,10.817]\n",
            "Iteration>1277, dA[0.028,0.020] dB[0.014,0.003] g[12.518,11.008]\n",
            "Iteration>1278, dA[0.019,0.038] dB[0.005,0.006] g[12.409,11.961]\n",
            "Iteration>1279, dA[0.369,0.506] dB[0.014,0.024] g[10.586,9.300]\n",
            "Iteration>1280, dA[0.033,0.034] dB[0.011,0.010] g[11.475,9.625]\n",
            "Iteration>1281, dA[0.037,0.008] dB[0.013,0.008] g[12.692,12.276]\n",
            "Iteration>1282, dA[0.026,0.003] dB[0.005,0.007] g[12.330,11.599]\n",
            "Iteration>1283, dA[0.011,0.008] dB[0.015,0.007] g[13.574,12.664]\n",
            "Iteration>1284, dA[0.016,0.021] dB[0.020,0.007] g[12.370,11.224]\n",
            "Iteration>1285, dA[0.003,0.019] dB[0.005,0.007] g[12.056,10.961]\n",
            "Iteration>1286, dA[0.006,0.025] dB[0.005,0.002] g[14.336,14.391]\n",
            "Iteration>1287, dA[0.133,0.105] dB[0.005,0.003] g[11.499,10.239]\n",
            "Iteration>1288, dA[0.015,0.007] dB[0.002,0.002] g[11.379,9.826]\n",
            "Iteration>1289, dA[0.004,0.011] dB[0.003,0.002] g[12.033,11.196]\n",
            "Iteration>1290, dA[0.018,0.004] dB[0.008,0.026] g[12.072,11.049]\n",
            "Iteration>1291, dA[0.104,0.653] dB[0.019,0.007] g[10.483,8.249]\n",
            "Iteration>1292, dA[0.118,0.071] dB[0.033,0.030] g[11.437,10.734]\n",
            "Iteration>1293, dA[0.127,0.104] dB[0.007,0.018] g[11.482,10.283]\n",
            "Iteration>1294, dA[0.051,0.014] dB[0.044,0.048] g[11.999,10.624]\n",
            "Iteration>1295, dA[0.021,0.012] dB[0.008,0.014] g[13.192,13.047]\n",
            "Iteration>1296, dA[0.013,0.009] dB[0.026,0.015] g[11.740,10.781]\n",
            "Iteration>1297, dA[0.032,0.044] dB[0.013,0.012] g[10.724,8.794]\n",
            "Iteration>1298, dA[0.036,0.023] dB[0.017,0.020] g[12.307,10.813]\n",
            "Iteration>1299, dA[0.140,0.077] dB[0.010,0.047] g[10.279,9.037]\n",
            "Iteration>1300, dA[0.018,0.024] dB[0.048,0.004] g[11.691,10.299]\n",
            "Iteration>1301, dA[0.003,0.019] dB[0.017,0.028] g[11.872,10.929]\n",
            "Iteration>1302, dA[0.012,0.030] dB[0.028,0.010] g[13.392,12.541]\n",
            "Iteration>1303, dA[0.049,0.080] dB[0.007,0.037] g[11.268,10.016]\n",
            "Iteration>1304, dA[0.013,0.026] dB[0.029,0.012] g[13.125,12.497]\n",
            "Iteration>1305, dA[0.019,0.007] dB[0.018,0.038] g[12.025,11.164]\n",
            "Iteration>1306, dA[0.163,0.362] dB[0.021,0.006] g[10.862,9.648]\n",
            "Iteration>1307, dA[0.014,0.059] dB[0.019,0.005] g[12.043,11.388]\n",
            "Iteration>1308, dA[0.026,0.011] dB[0.007,0.004] g[12.175,11.780]\n",
            "Iteration>1309, dA[0.019,0.022] dB[0.015,0.004] g[13.023,12.351]\n",
            "Iteration>1310, dA[0.452,0.182] dB[0.005,0.023] g[11.159,12.169]\n",
            "Iteration>1311, dA[0.098,0.249] dB[0.012,0.007] g[10.929,8.496]\n",
            "Iteration>1312, dA[0.161,0.122] dB[0.018,0.009] g[10.782,8.381]\n",
            "Iteration>1313, dA[0.041,0.041] dB[0.004,0.018] g[14.589,14.840]\n",
            "Iteration>1314, dA[0.201,0.081] dB[0.018,0.006] g[11.309,10.181]\n",
            "Iteration>1315, dA[0.016,0.011] dB[0.001,0.007] g[12.800,11.409]\n",
            "Iteration>1316, dA[0.017,0.025] dB[0.012,0.014] g[11.805,10.525]\n",
            "Iteration>1317, dA[0.020,0.007] dB[0.013,0.005] g[11.874,10.822]\n",
            "Iteration>1318, dA[0.013,0.022] dB[0.006,0.012] g[14.420,14.225]\n",
            "Iteration>1319, dA[0.015,0.014] dB[0.027,0.029] g[12.217,11.355]\n",
            "Iteration>1320, dA[0.028,0.043] dB[0.005,0.003] g[11.246,9.389]\n",
            "Iteration>1321, dA[0.005,0.024] dB[0.007,0.006] g[11.882,10.076]\n",
            "Iteration>1322, dA[0.012,0.015] dB[0.005,0.001] g[12.225,11.613]\n",
            "Iteration>1323, dA[0.009,0.005] dB[0.005,0.002] g[14.402,14.279]\n",
            "Iteration>1324, dA[0.005,0.007] dB[0.007,0.015] g[13.062,12.367]\n",
            "Iteration>1325, dA[0.006,0.011] dB[0.011,0.002] g[12.061,11.168]\n",
            "Iteration>1326, dA[0.008,0.006] dB[0.005,0.004] g[11.245,10.170]\n",
            "Iteration>1327, dA[0.050,0.086] dB[0.015,0.022] g[11.314,9.240]\n",
            "Iteration>1328, dA[0.044,0.013] dB[0.005,0.007] g[12.295,10.298]\n",
            "Iteration>1329, dA[0.046,0.067] dB[0.009,0.006] g[11.198,9.738]\n",
            "Iteration>1330, dA[0.081,0.026] dB[0.002,0.004] g[11.379,9.868]\n",
            "Iteration>1331, dA[0.320,0.456] dB[0.005,0.002] g[10.263,8.923]\n",
            "Iteration>1332, dA[0.015,0.014] dB[0.006,0.013] g[12.412,11.182]\n",
            "Iteration>1333, dA[0.309,0.129] dB[0.009,0.002] g[10.906,9.884]\n",
            "Iteration>1334, dA[0.049,0.074] dB[0.014,0.017] g[12.537,11.281]\n",
            "Iteration>1335, dA[0.064,0.030] dB[0.008,0.005] g[12.185,11.090]\n",
            "Iteration>1336, dA[0.020,0.018] dB[0.011,0.004] g[13.822,13.302]\n",
            "Iteration>1337, dA[0.012,0.016] dB[0.006,0.006] g[12.362,11.390]\n",
            "Iteration>1338, dA[0.021,0.019] dB[0.006,0.004] g[12.338,11.396]\n",
            "Iteration>1339, dA[0.105,0.248] dB[0.003,0.006] g[11.004,9.197]\n",
            "Iteration>1340, dA[0.008,0.020] dB[0.005,0.001] g[11.756,10.891]\n",
            "Iteration>1341, dA[0.308,0.082] dB[0.005,0.004] g[11.221,10.103]\n",
            "Iteration>1342, dA[0.114,0.013] dB[0.010,0.015] g[11.170,9.522]\n",
            "Iteration>1343, dA[0.105,0.073] dB[0.012,0.005] g[10.925,9.593]\n",
            "Iteration>1344, dA[0.013,0.008] dB[0.024,0.012] g[11.735,10.753]\n",
            "Iteration>1345, dA[0.260,0.066] dB[0.019,0.008] g[10.431,8.971]\n",
            "Iteration>1346, dA[0.102,0.112] dB[0.013,0.013] g[12.329,10.508]\n",
            "Iteration>1347, dA[0.051,0.025] dB[0.012,0.009] g[11.666,10.528]\n",
            "Iteration>1348, dA[0.168,0.134] dB[0.010,0.013] g[10.569,9.550]\n",
            "Iteration>1349, dA[0.181,0.302] dB[0.007,0.008] g[10.197,8.757]\n",
            "Iteration>1350, dA[0.065,0.014] dB[0.008,0.006] g[11.108,9.493]\n",
            "Iteration>1351, dA[0.073,0.026] dB[0.006,0.005] g[11.394,10.428]\n",
            "Iteration>1352, dA[0.011,0.072] dB[0.004,0.012] g[12.091,10.743]\n",
            "Iteration>1353, dA[0.020,0.031] dB[0.025,0.009] g[11.800,10.982]\n",
            "Iteration>1354, dA[0.008,0.020] dB[0.009,0.020] g[11.763,10.679]\n",
            "Iteration>1355, dA[0.007,0.006] dB[0.012,0.003] g[13.624,13.032]\n",
            "Iteration>1356, dA[0.538,0.175] dB[0.018,0.047] g[10.299,9.826]\n",
            "Iteration>1357, dA[0.216,0.079] dB[0.024,0.014] g[12.962,11.038]\n",
            "Iteration>1358, dA[0.211,0.015] dB[0.075,0.080] g[10.681,8.843]\n",
            "Iteration>1359, dA[0.024,0.225] dB[0.030,0.018] g[11.941,10.186]\n",
            "Iteration>1360, dA[0.130,0.030] dB[0.074,0.045] g[12.573,12.293]\n",
            "Iteration>1361, dA[0.034,0.011] dB[0.036,0.015] g[13.101,12.820]\n",
            "Iteration>1362, dA[0.040,0.048] dB[0.031,0.056] g[10.477,8.585]\n",
            "Iteration>1363, dA[0.026,0.010] dB[0.103,0.063] g[12.642,10.805]\n",
            "Iteration>1364, dA[0.007,0.040] dB[0.019,0.061] g[12.884,12.051]\n",
            "Iteration>1365, dA[0.104,0.023] dB[0.079,0.017] g[11.393,10.181]\n",
            "Iteration>1366, dA[0.033,0.122] dB[0.012,0.005] g[11.085,9.370]\n",
            "Iteration>1367, dA[0.193,0.024] dB[0.017,0.009] g[10.051,8.905]\n",
            "Iteration>1368, dA[0.033,0.047] dB[0.011,0.020] g[11.399,9.522]\n",
            "Iteration>1369, dA[0.025,0.012] dB[0.015,0.007] g[11.198,9.711]\n",
            "Iteration>1370, dA[0.008,0.080] dB[0.038,0.036] g[11.875,10.575]\n",
            "Iteration>1371, dA[0.279,0.037] dB[0.010,0.012] g[10.712,9.794]\n",
            "Iteration>1372, dA[0.054,0.195] dB[0.033,0.027] g[9.757,7.348]\n",
            "Iteration>1373, dA[0.065,0.009] dB[0.006,0.019] g[10.686,9.512]\n",
            "Iteration>1374, dA[0.038,0.081] dB[0.010,0.005] g[10.341,8.632]\n",
            "Iteration>1375, dA[0.010,0.014] dB[0.019,0.016] g[10.960,9.345]\n",
            "Iteration>1376, dA[0.007,0.033] dB[0.006,0.022] g[11.419,10.850]\n",
            "Iteration>1377, dA[0.029,0.013] dB[0.022,0.007] g[11.298,10.289]\n",
            "Iteration>1378, dA[0.019,0.008] dB[0.014,0.018] g[10.859,9.132]\n",
            "Iteration>1379, dA[0.102,0.217] dB[0.013,0.015] g[10.294,8.064]\n",
            "Iteration>1380, dA[0.055,0.026] dB[0.018,0.013] g[12.723,11.610]\n",
            "Iteration>1381, dA[0.154,0.110] dB[0.013,0.011] g[12.004,11.776]\n",
            "Iteration>1382, dA[0.014,0.014] dB[0.011,0.007] g[12.705,11.708]\n",
            "Iteration>1383, dA[0.176,0.348] dB[0.007,0.010] g[10.753,9.303]\n",
            "Iteration>1384, dA[0.042,0.040] dB[0.023,0.007] g[11.933,11.061]\n",
            "Iteration>1385, dA[0.127,0.058] dB[0.033,0.049] g[11.432,10.053]\n",
            "Iteration>1386, dA[0.025,0.026] dB[0.022,0.005] g[13.554,12.932]\n",
            "Iteration>1387, dA[0.029,0.025] dB[0.017,0.014] g[12.030,11.464]\n",
            "Iteration>1388, dA[0.037,0.015] dB[0.013,0.006] g[13.289,13.031]\n",
            "Iteration>1389, dA[0.014,0.012] dB[0.023,0.015] g[11.034,9.774]\n",
            "Iteration>1390, dA[0.015,0.016] dB[0.044,0.053] g[11.890,10.924]\n",
            "Iteration>1391, dA[0.302,0.278] dB[0.029,0.026] g[9.970,8.741]\n",
            "Iteration>1392, dA[0.019,0.018] dB[0.030,0.010] g[12.889,12.112]\n",
            "Iteration>1393, dA[0.235,0.128] dB[0.015,0.015] g[12.113,11.220]\n",
            "Iteration>1394, dA[0.032,0.019] dB[0.005,0.008] g[12.729,12.285]\n",
            "Iteration>1395, dA[0.137,0.095] dB[0.005,0.008] g[10.741,9.180]\n",
            "Iteration>1396, dA[0.004,0.034] dB[0.013,0.005] g[12.054,10.769]\n",
            "Iteration>1397, dA[0.016,0.011] dB[0.016,0.016] g[11.997,10.873]\n",
            "Iteration>1398, dA[0.034,0.036] dB[0.029,0.018] g[11.235,10.169]\n",
            "Iteration>1399, dA[0.020,0.065] dB[0.011,0.015] g[12.202,10.559]\n",
            "Iteration>1400, dA[0.024,0.015] dB[0.022,0.001] g[11.164,10.585]\n",
            "Iteration>1401, dA[0.008,0.012] dB[0.014,0.004] g[12.155,10.698]\n",
            "Iteration>1402, dA[0.061,0.071] dB[0.003,0.005] g[10.687,9.101]\n",
            "Iteration>1403, dA[0.008,0.017] dB[0.009,0.016] g[11.733,10.447]\n",
            "Iteration>1404, dA[0.012,0.021] dB[0.011,0.014] g[11.486,10.287]\n",
            "Iteration>1405, dA[0.014,0.045] dB[0.017,0.003] g[10.402,8.739]\n",
            "Iteration>1406, dA[0.027,0.029] dB[0.015,0.006] g[12.775,11.637]\n",
            "Iteration>1407, dA[0.024,0.032] dB[0.005,0.009] g[11.380,10.527]\n",
            "Iteration>1408, dA[0.275,0.444] dB[0.014,0.007] g[9.688,8.075]\n",
            "Iteration>1409, dA[0.013,0.020] dB[0.015,0.011] g[10.564,9.044]\n",
            "Iteration>1410, dA[0.343,0.191] dB[0.008,0.004] g[10.727,9.424]\n",
            "Iteration>1411, dA[0.043,0.049] dB[0.004,0.001] g[11.953,10.601]\n",
            "Iteration>1412, dA[0.229,0.045] dB[0.004,0.001] g[10.749,10.060]\n",
            "Iteration>1413, dA[0.020,0.028] dB[0.003,0.002] g[11.335,9.745]\n",
            "Iteration>1414, dA[0.019,0.014] dB[0.005,0.005] g[12.180,11.786]\n",
            "Iteration>1415, dA[0.012,0.011] dB[0.006,0.002] g[13.137,12.831]\n",
            "Iteration>1416, dA[0.008,0.016] dB[0.002,0.001] g[12.155,11.115]\n",
            "Iteration>1417, dA[0.047,0.028] dB[0.002,0.002] g[10.992,9.512]\n",
            "Iteration>1418, dA[0.017,0.008] dB[0.002,0.001] g[11.190,9.584]\n",
            "Iteration>1419, dA[0.008,0.015] dB[0.012,0.012] g[12.117,11.143]\n",
            "Iteration>1420, dA[0.024,0.012] dB[0.018,0.010] g[12.184,10.777]\n",
            "Iteration>1421, dA[0.031,0.018] dB[0.022,0.004] g[11.414,10.343]\n",
            "Iteration>1422, dA[0.118,0.229] dB[0.006,0.005] g[10.252,8.546]\n",
            "Iteration>1423, dA[0.016,0.097] dB[0.003,0.004] g[11.309,9.083]\n",
            "Iteration>1424, dA[0.060,0.076] dB[0.006,0.013] g[13.115,13.818]\n",
            "Iteration>1425, dA[0.106,0.096] dB[0.022,0.011] g[11.264,10.209]\n",
            "Iteration>1426, dA[0.062,0.387] dB[0.021,0.039] g[9.895,7.411]\n",
            "Iteration>1427, dA[0.007,0.194] dB[0.022,0.005] g[11.918,11.284]\n",
            "Iteration>1428, dA[0.098,0.061] dB[0.014,0.017] g[11.577,11.181]\n",
            "Iteration>1429, dA[0.051,0.034] dB[0.013,0.007] g[12.183,11.204]\n",
            "Iteration>1430, dA[0.028,0.021] dB[0.024,0.010] g[11.925,10.986]\n",
            "Iteration>1431, dA[0.034,0.012] dB[0.009,0.008] g[11.490,10.724]\n",
            "Iteration>1432, dA[0.086,0.036] dB[0.012,0.004] g[10.496,8.710]\n",
            "Iteration>1433, dA[0.048,0.065] dB[0.009,0.038] g[12.179,11.010]\n",
            "Iteration>1434, dA[0.068,0.046] dB[0.024,0.006] g[11.358,10.866]\n",
            "Iteration>1435, dA[0.086,0.081] dB[0.030,0.047] g[10.368,9.290]\n",
            "Iteration>1436, dA[0.056,0.048] dB[0.014,0.013] g[12.229,10.652]\n",
            "Iteration>1437, dA[0.032,0.033] dB[0.030,0.025] g[11.689,10.632]\n",
            "Iteration>1438, dA[0.009,0.061] dB[0.003,0.012] g[13.141,12.471]\n",
            "Iteration>1439, dA[0.071,0.077] dB[0.019,0.009] g[10.906,9.339]\n",
            "Iteration>1440, dA[0.077,0.086] dB[0.003,0.001] g[10.279,8.538]\n",
            "Iteration>1441, dA[0.396,0.204] dB[0.002,0.003] g[10.407,8.428]\n",
            "Iteration>1442, dA[0.031,0.062] dB[0.004,0.010] g[10.525,9.157]\n",
            "Iteration>1443, dA[0.017,0.014] dB[0.008,0.003] g[11.379,10.258]\n",
            "Iteration>1444, dA[0.404,0.168] dB[0.007,0.008] g[10.326,9.241]\n",
            "Iteration>1445, dA[0.078,0.028] dB[0.013,0.034] g[11.005,9.328]\n",
            "Iteration>1446, dA[0.358,0.143] dB[0.013,0.006] g[9.664,8.172]\n",
            "Iteration>1447, dA[0.026,0.031] dB[0.008,0.007] g[12.849,12.189]\n",
            "Iteration>1448, dA[0.109,0.085] dB[0.008,0.004] g[10.434,9.097]\n",
            "Iteration>1449, dA[0.019,0.009] dB[0.018,0.028] g[12.939,12.432]\n",
            "Iteration>1450, dA[0.345,0.354] dB[0.021,0.006] g[9.666,8.112]\n",
            "Iteration>1451, dA[0.020,0.014] dB[0.018,0.007] g[12.946,11.978]\n",
            "Iteration>1452, dA[0.122,0.036] dB[0.007,0.004] g[12.091,11.677]\n",
            "Iteration>1453, dA[0.091,0.058] dB[0.005,0.008] g[10.624,8.949]\n",
            "Iteration>1454, dA[0.021,0.052] dB[0.004,0.009] g[12.415,11.682]\n",
            "Iteration>1455, dA[0.020,0.089] dB[0.009,0.007] g[11.496,10.676]\n",
            "Iteration>1456, dA[0.088,0.033] dB[0.005,0.006] g[11.189,9.817]\n",
            "Iteration>1457, dA[0.017,0.061] dB[0.002,0.002] g[10.452,8.566]\n",
            "Iteration>1458, dA[0.013,0.018] dB[0.007,0.011] g[11.777,10.635]\n",
            "Iteration>1459, dA[0.006,0.017] dB[0.008,0.002] g[12.661,12.477]\n",
            "Iteration>1460, dA[0.009,0.014] dB[0.013,0.040] g[11.509,10.730]\n",
            "Iteration>1461, dA[0.005,0.014] dB[0.023,0.011] g[13.056,12.328]\n",
            "Iteration>1462, dA[0.007,0.018] dB[0.051,0.065] g[12.026,11.104]\n",
            "Iteration>1463, dA[0.005,0.009] dB[0.018,0.018] g[12.373,11.725]\n",
            "Iteration>1464, dA[0.024,0.064] dB[0.056,0.072] g[10.782,9.716]\n",
            "Iteration>1465, dA[0.070,0.013] dB[0.040,0.045] g[10.810,9.026]\n",
            "Iteration>1466, dA[0.045,0.014] dB[0.081,0.017] g[10.921,9.588]\n",
            "Iteration>1467, dA[0.005,0.068] dB[0.038,0.019] g[11.873,10.439]\n",
            "Iteration>1468, dA[0.033,0.014] dB[0.010,0.008] g[11.068,10.529]\n",
            "Iteration>1469, dA[0.029,0.012] dB[0.014,0.009] g[11.036,9.842]\n",
            "Iteration>1470, dA[0.013,0.035] dB[0.008,0.010] g[11.737,10.618]\n",
            "Iteration>1471, dA[0.008,0.025] dB[0.009,0.013] g[12.717,12.795]\n",
            "Iteration>1472, dA[0.005,0.016] dB[0.016,0.007] g[12.668,12.416]\n",
            "Iteration>1473, dA[0.007,0.013] dB[0.010,0.056] g[11.843,11.261]\n",
            "Iteration>1474, dA[0.020,0.026] dB[0.038,0.006] g[11.946,11.182]\n",
            "Iteration>1475, dA[0.030,0.009] dB[0.027,0.011] g[11.342,10.640]\n",
            "Iteration>1476, dA[0.005,0.013] dB[0.050,0.037] g[11.194,10.654]\n",
            "Iteration>1477, dA[0.007,0.007] dB[0.013,0.060] g[11.567,10.092]\n",
            "Iteration>1478, dA[0.004,0.006] dB[0.037,0.019] g[11.625,10.472]\n",
            "Iteration>1479, dA[0.006,0.066] dB[0.040,0.013] g[11.126,9.323]\n",
            "Iteration>1480, dA[0.275,0.065] dB[0.020,0.015] g[10.184,9.606]\n",
            "Iteration>1481, dA[0.102,0.192] dB[0.015,0.008] g[11.135,8.830]\n",
            "Iteration>1482, dA[0.428,0.097] dB[0.007,0.036] g[9.819,9.350]\n",
            "Iteration>1483, dA[0.038,0.036] dB[0.057,0.025] g[12.467,12.008]\n",
            "Iteration>1484, dA[0.035,0.029] dB[0.020,0.042] g[10.864,9.582]\n",
            "Iteration>1485, dA[0.018,0.192] dB[0.036,0.036] g[13.305,11.865]\n",
            "Iteration>1486, dA[0.079,0.048] dB[0.020,0.008] g[12.337,13.052]\n",
            "Iteration>1487, dA[0.065,0.027] dB[0.024,0.023] g[11.171,11.215]\n",
            "Iteration>1488, dA[0.125,0.234] dB[0.020,0.010] g[10.616,9.710]\n",
            "Iteration>1489, dA[0.029,0.024] dB[0.007,0.004] g[10.775,9.434]\n",
            "Iteration>1490, dA[0.172,0.161] dB[0.010,0.035] g[10.699,9.077]\n",
            "Iteration>1491, dA[0.077,0.152] dB[0.024,0.037] g[12.665,12.598]\n",
            "Iteration>1492, dA[0.172,0.136] dB[0.010,0.045] g[12.586,13.099]\n",
            "Iteration>1493, dA[0.425,0.258] dB[0.072,0.105] g[10.153,9.017]\n",
            "Iteration>1494, dA[0.140,0.153] dB[0.061,0.057] g[12.435,10.331]\n",
            "Iteration>1495, dA[0.172,0.038] dB[0.108,0.082] g[10.822,10.562]\n",
            "Iteration>1496, dA[0.014,0.088] dB[0.065,0.142] g[13.097,12.647]\n",
            "Iteration>1497, dA[0.043,0.019] dB[0.092,0.028] g[12.448,10.605]\n",
            "Iteration>1498, dA[0.120,0.038] dB[0.027,0.111] g[10.819,10.335]\n",
            "Iteration>1499, dA[0.074,0.310] dB[0.096,0.018] g[10.490,7.897]\n",
            "Iteration>1500, dA[0.012,0.089] dB[0.054,0.092] g[12.527,12.231]\n",
            "Iteration>1501, dA[0.028,0.015] dB[0.079,0.009] g[12.961,12.404]\n",
            "Iteration>1502, dA[0.260,0.130] dB[0.034,0.036] g[10.881,9.842]\n",
            "Iteration>1503, dA[0.027,0.061] dB[0.041,0.024] g[11.211,10.259]\n",
            "Iteration>1504, dA[0.030,0.013] dB[0.075,0.137] g[10.974,10.730]\n",
            "Iteration>1505, dA[0.071,0.036] dB[0.129,0.006] g[11.410,10.125]\n",
            "Iteration>1506, dA[0.036,0.009] dB[0.092,0.089] g[11.346,11.192]\n",
            "Iteration>1507, dA[0.181,0.068] dB[0.028,0.114] g[9.833,10.023]\n",
            "Iteration>1508, dA[0.021,0.024] dB[0.097,0.073] g[11.354,10.169]\n",
            "Iteration>1509, dA[0.015,0.008] dB[0.030,0.040] g[12.888,12.391]\n",
            "Iteration>1510, dA[0.007,0.019] dB[0.009,0.097] g[11.125,10.688]\n",
            "Iteration>1511, dA[0.012,0.015] dB[0.074,0.017] g[11.411,10.047]\n",
            "Iteration>1512, dA[0.004,0.008] dB[0.019,0.034] g[12.849,12.345]\n",
            "Iteration>1513, dA[0.043,0.149] dB[0.028,0.112] g[11.254,9.105]\n",
            "Iteration>1514, dA[0.009,0.073] dB[0.062,0.050] g[11.891,10.920]\n",
            "Iteration>1515, dA[0.039,0.029] dB[0.076,0.174] g[11.769,10.836]\n",
            "Iteration>1516, dA[0.025,0.014] dB[0.216,0.056] g[11.997,9.832]\n",
            "Iteration>1517, dA[0.126,0.060] dB[0.197,0.056] g[10.394,9.343]\n",
            "Iteration>1518, dA[0.034,0.025] dB[0.219,0.258] g[10.612,7.964]\n",
            "Iteration>1519, dA[0.010,0.029] dB[0.114,0.222] g[10.110,8.763]\n",
            "Iteration>1520, dA[0.009,0.077] dB[0.194,0.019] g[11.962,10.009]\n",
            "Iteration>1521, dA[0.040,0.058] dB[0.135,0.081] g[11.815,12.202]\n",
            "Iteration>1522, dA[0.009,0.010] dB[0.084,0.161] g[11.601,11.753]\n",
            "Iteration>1523, dA[0.015,0.022] dB[0.194,0.022] g[11.217,10.199]\n",
            "Iteration>1524, dA[0.028,0.011] dB[0.107,0.075] g[11.540,11.119]\n",
            "Iteration>1525, dA[0.032,0.026] dB[0.036,0.060] g[10.970,10.564]\n",
            "Iteration>1526, dA[0.006,0.008] dB[0.083,0.033] g[11.969,10.713]\n",
            "Iteration>1527, dA[0.262,0.188] dB[0.034,0.031] g[10.235,8.487]\n",
            "Iteration>1528, dA[0.059,0.009] dB[0.018,0.015] g[10.842,8.978]\n",
            "Iteration>1529, dA[0.024,0.036] dB[0.022,0.006] g[11.415,10.396]\n",
            "Iteration>1530, dA[0.024,0.005] dB[0.014,0.015] g[12.141,11.216]\n",
            "Iteration>1531, dA[0.172,0.132] dB[0.014,0.014] g[10.867,9.383]\n",
            "Iteration>1532, dA[0.015,0.012] dB[0.011,0.018] g[10.575,9.058]\n",
            "Iteration>1533, dA[0.011,0.008] dB[0.008,0.009] g[11.296,10.689]\n",
            "Iteration>1534, dA[0.004,0.004] dB[0.018,0.040] g[11.194,10.493]\n",
            "Iteration>1535, dA[0.011,0.008] dB[0.032,0.009] g[12.704,12.326]\n",
            "Iteration>1536, dA[0.025,0.012] dB[0.029,0.009] g[11.686,10.870]\n",
            "Iteration>1537, dA[0.008,0.023] dB[0.025,0.015] g[10.901,10.204]\n",
            "Iteration>1538, dA[0.002,0.005] dB[0.026,0.021] g[11.666,10.578]\n",
            "Iteration>1539, dA[0.009,0.009] dB[0.036,0.043] g[12.989,12.648]\n",
            "Iteration>1540, dA[0.005,0.005] dB[0.030,0.057] g[11.167,10.438]\n",
            "Iteration>1541, dA[0.075,0.102] dB[0.055,0.023] g[10.845,8.929]\n",
            "Iteration>1542, dA[0.005,0.011] dB[0.036,0.020] g[10.744,9.471]\n",
            "Iteration>1543, dA[0.010,0.019] dB[0.042,0.060] g[12.642,12.537]\n",
            "Iteration>1544, dA[0.011,0.019] dB[0.018,0.063] g[11.928,11.876]\n",
            "Iteration>1545, dA[0.010,0.015] dB[0.244,0.034] g[12.736,10.867]\n",
            "Iteration>1546, dA[0.039,0.024] dB[0.170,0.031] g[10.038,9.170]\n",
            "Iteration>1547, dA[0.048,0.007] dB[0.151,0.343] g[11.597,10.618]\n",
            "Iteration>1548, dA[0.023,0.008] dB[0.306,0.052] g[12.061,11.084]\n",
            "Iteration>1549, dA[0.002,0.013] dB[0.279,0.560] g[12.105,12.446]\n",
            "Iteration>1550, dA[0.014,0.015] dB[0.501,0.235] g[14.088,10.280]\n",
            "Iteration>1551, dA[0.051,0.016] dB[0.141,0.299] g[15.014,13.803]\n",
            "Iteration>1552, dA[0.010,0.013] dB[0.102,0.029] g[11.702,11.082]\n",
            "Iteration>1553, dA[0.004,0.020] dB[0.044,0.024] g[12.121,11.305]\n",
            "Iteration>1554, dA[0.025,0.009] dB[0.031,0.024] g[11.171,10.195]\n",
            "Iteration>1555, dA[0.010,0.013] dB[0.014,0.007] g[11.946,10.691]\n",
            "Iteration>1556, dA[0.011,0.015] dB[0.013,0.011] g[11.488,10.108]\n",
            "Iteration>1557, dA[0.006,0.016] dB[0.008,0.014] g[12.773,12.024]\n",
            "Iteration>1558, dA[0.016,0.029] dB[0.009,0.012] g[11.677,10.507]\n",
            "Iteration>1559, dA[0.018,0.005] dB[0.008,0.007] g[11.662,11.097]\n",
            "Iteration>1560, dA[0.084,0.167] dB[0.039,0.033] g[11.017,10.469]\n",
            "Iteration>1561, dA[0.053,0.024] dB[0.027,0.041] g[10.523,9.143]\n",
            "Iteration>1562, dA[0.014,0.176] dB[0.118,0.030] g[13.382,12.939]\n",
            "Iteration>1563, dA[0.242,0.043] dB[0.139,0.149] g[10.956,10.564]\n",
            "Iteration>1564, dA[0.011,0.122] dB[0.031,0.093] g[10.716,9.129]\n",
            "Iteration>1565, dA[0.176,0.072] dB[0.017,0.054] g[10.866,9.512]\n",
            "Iteration>1566, dA[0.012,0.019] dB[0.048,0.013] g[11.412,10.705]\n",
            "Iteration>1567, dA[0.011,0.020] dB[0.011,0.011] g[12.980,12.640]\n",
            "Iteration>1568, dA[0.038,0.011] dB[0.016,0.012] g[11.449,10.617]\n",
            "Iteration>1569, dA[0.147,0.115] dB[0.008,0.011] g[10.524,9.043]\n",
            "Iteration>1570, dA[0.040,0.093] dB[0.028,0.032] g[10.050,7.562]\n",
            "Iteration>1571, dA[0.085,0.186] dB[0.015,0.019] g[10.027,9.644]\n",
            "Iteration>1572, dA[0.136,0.024] dB[0.017,0.026] g[11.637,11.491]\n",
            "Iteration>1573, dA[0.013,0.034] dB[0.011,0.016] g[12.191,11.979]\n",
            "Iteration>1574, dA[0.035,0.015] dB[0.009,0.007] g[10.720,9.569]\n",
            "Iteration>1575, dA[0.024,0.016] dB[0.007,0.004] g[11.535,10.650]\n",
            "Iteration>1576, dA[0.006,0.016] dB[0.023,0.006] g[11.159,10.817]\n",
            "Iteration>1577, dA[0.006,0.011] dB[0.006,0.006] g[11.277,10.703]\n",
            "Iteration>1578, dA[0.154,0.351] dB[0.007,0.004] g[10.062,9.281]\n",
            "Iteration>1579, dA[0.115,0.022] dB[0.015,0.088] g[10.905,11.168]\n",
            "Iteration>1580, dA[0.050,0.045] dB[0.171,0.021] g[11.774,10.830]\n",
            "Iteration>1581, dA[0.098,0.174] dB[0.037,0.023] g[9.894,8.608]\n",
            "Iteration>1582, dA[0.034,0.012] dB[0.009,0.012] g[12.409,11.965]\n",
            "Iteration>1583, dA[0.578,0.220] dB[0.011,0.011] g[10.237,9.831]\n",
            "Iteration>1584, dA[0.026,0.109] dB[0.016,0.040] g[10.230,8.344]\n",
            "Iteration>1585, dA[0.010,0.015] dB[0.007,0.041] g[11.016,10.525]\n",
            "Iteration>1586, dA[0.041,0.018] dB[0.028,0.007] g[11.652,11.046]\n",
            "Iteration>1587, dA[0.008,0.042] dB[0.016,0.029] g[10.840,9.998]\n",
            "Iteration>1588, dA[0.015,0.016] dB[0.007,0.016] g[11.085,10.368]\n",
            "Iteration>1589, dA[0.071,0.017] dB[0.009,0.054] g[10.254,9.404]\n",
            "Iteration>1590, dA[0.006,0.086] dB[0.024,0.011] g[10.690,8.712]\n",
            "Iteration>1591, dA[0.093,0.055] dB[0.013,0.021] g[10.343,9.728]\n",
            "Iteration>1592, dA[0.003,0.032] dB[0.032,0.014] g[11.974,11.312]\n",
            "Iteration>1593, dA[0.010,0.011] dB[0.011,0.013] g[11.131,11.193]\n",
            "Iteration>1594, dA[0.151,0.781] dB[0.013,0.016] g[9.577,8.772]\n",
            "Iteration>1595, dA[0.195,0.022] dB[0.009,0.006] g[10.161,8.595]\n",
            "Iteration>1596, dA[0.027,0.055] dB[0.005,0.007] g[11.493,10.516]\n",
            "Iteration>1597, dA[0.098,0.166] dB[0.032,0.059] g[10.334,8.971]\n",
            "Iteration>1598, dA[0.115,0.032] dB[0.026,0.007] g[10.608,9.063]\n",
            "Iteration>1599, dA[0.008,0.032] dB[0.020,0.007] g[11.117,9.734]\n",
            "Iteration>1600, dA[0.009,0.070] dB[0.004,0.006] g[11.264,9.803]\n",
            "WARNING:tensorflow:Compiled the loaded model, but the compiled metrics have yet to be built. `model.compile_metrics` will be empty until you train or evaluate the model.\n",
            "WARNING:tensorflow:Compiled the loaded model, but the compiled metrics have yet to be built. `model.compile_metrics` will be empty until you train or evaluate the model.\n",
            ">Saved: /content/drive/MyDrive/Project/CycleGAN/TrainedModels/C_Highboost_Real_Model/g_model_AtoB_001600.h5 and /content/drive/MyDrive/Project/CycleGAN/TrainedModels/C_Highboost_Real_Model/g_model_BtoA_001600.h5\n",
            "Iteration>1601, dA[0.040,0.058] dB[0.011,0.029] g[12.918,12.964]\n",
            "Iteration>1602, dA[0.069,0.037] dB[0.012,0.002] g[11.111,10.308]\n",
            "Iteration>1603, dA[0.039,0.028] dB[0.002,0.006] g[13.242,13.222]\n",
            "Iteration>1604, dA[0.026,0.026] dB[0.012,0.015] g[11.168,10.409]\n",
            "Iteration>1605, dA[0.009,0.005] dB[0.011,0.023] g[11.452,10.570]\n",
            "Iteration>1606, dA[0.102,0.189] dB[0.040,0.061] g[9.979,8.666]\n",
            "Iteration>1607, dA[0.008,0.018] dB[0.037,0.026] g[12.295,12.151]\n",
            "Iteration>1608, dA[0.340,0.105] dB[0.035,0.018] g[10.622,9.164]\n",
            "Iteration>1609, dA[0.052,0.045] dB[0.020,0.024] g[11.334,9.668]\n",
            "Iteration>1610, dA[0.021,0.033] dB[0.024,0.029] g[12.348,11.963]\n",
            "Iteration>1611, dA[0.020,0.034] dB[0.009,0.016] g[11.508,10.460]\n",
            "Iteration>1612, dA[0.017,0.014] dB[0.010,0.024] g[11.086,10.083]\n",
            "Iteration>1613, dA[0.011,0.017] dB[0.027,0.018] g[11.321,10.073]\n",
            "Iteration>1614, dA[0.010,0.056] dB[0.008,0.009] g[11.295,10.243]\n",
            "Iteration>1615, dA[0.035,0.025] dB[0.013,0.004] g[10.769,10.121]\n",
            "Iteration>1616, dA[0.052,0.104] dB[0.007,0.041] g[9.937,9.190]\n",
            "Iteration>1617, dA[0.025,0.033] dB[0.021,0.014] g[10.441,9.054]\n",
            "Iteration>1618, dA[0.024,0.021] dB[0.012,0.023] g[9.574,8.507]\n",
            "Iteration>1619, dA[0.012,0.008] dB[0.006,0.018] g[12.096,11.509]\n",
            "Iteration>1620, dA[0.003,0.007] dB[0.018,0.004] g[11.862,11.534]\n",
            "Iteration>1621, dA[0.009,0.008] dB[0.030,0.025] g[10.388,9.360]\n",
            "Iteration>1622, dA[0.005,0.016] dB[0.013,0.019] g[12.150,12.123]\n",
            "Iteration>1623, dA[0.012,0.014] dB[0.014,0.011] g[10.735,9.773]\n",
            "Iteration>1624, dA[0.310,0.199] dB[0.021,0.008] g[10.220,9.013]\n",
            "Iteration>1625, dA[0.095,0.087] dB[0.016,0.014] g[10.849,9.441]\n",
            "Iteration>1626, dA[0.405,0.114] dB[0.021,0.010] g[9.984,9.737]\n",
            "Iteration>1627, dA[0.031,0.030] dB[0.011,0.020] g[10.662,9.901]\n",
            "Iteration>1628, dA[0.044,0.111] dB[0.010,0.010] g[11.808,10.862]\n",
            "Iteration>1629, dA[0.068,0.016] dB[0.006,0.014] g[10.622,9.081]\n",
            "Iteration>1630, dA[0.027,0.152] dB[0.022,0.027] g[11.852,10.715]\n",
            "Iteration>1631, dA[0.681,0.086] dB[0.009,0.019] g[9.871,10.469]\n",
            "Iteration>1632, dA[0.030,0.048] dB[0.011,0.030] g[10.945,9.612]\n",
            "Iteration>1633, dA[0.020,0.041] dB[0.019,0.009] g[11.537,10.504]\n",
            "Iteration>1634, dA[0.076,0.010] dB[0.015,0.043] g[11.106,10.613]\n",
            "Iteration>1635, dA[0.043,0.020] dB[0.011,0.015] g[12.933,12.080]\n",
            "Iteration>1636, dA[0.007,0.040] dB[0.003,0.007] g[11.170,9.971]\n",
            "Iteration>1637, dA[0.198,0.242] dB[0.011,0.015] g[10.055,8.433]\n",
            "Iteration>1638, dA[0.009,0.023] dB[0.019,0.019] g[11.366,10.028]\n",
            "Iteration>1639, dA[0.151,0.045] dB[0.005,0.012] g[10.395,9.720]\n",
            "Iteration>1640, dA[0.045,0.036] dB[0.013,0.002] g[11.286,9.870]\n",
            "Iteration>1641, dA[0.015,0.055] dB[0.013,0.013] g[10.776,9.614]\n",
            "Iteration>1642, dA[0.054,0.038] dB[0.005,0.006] g[10.722,10.236]\n",
            "Iteration>1643, dA[0.015,0.044] dB[0.013,0.005] g[10.985,9.997]\n",
            "Iteration>1644, dA[0.015,0.030] dB[0.015,0.003] g[10.963,10.037]\n",
            "Iteration>1645, dA[0.236,0.264] dB[0.005,0.002] g[9.072,7.827]\n",
            "Iteration>1646, dA[0.052,0.024] dB[0.003,0.003] g[9.508,7.429]\n",
            "Iteration>1647, dA[0.110,0.014] dB[0.006,0.004] g[10.155,9.131]\n",
            "Iteration>1648, dA[0.009,0.305] dB[0.004,0.008] g[10.141,8.023]\n",
            "Iteration>1649, dA[0.246,0.069] dB[0.006,0.006] g[11.124,11.471]\n",
            "Iteration>1650, dA[0.113,0.341] dB[0.018,0.027] g[9.672,7.982]\n",
            "Iteration>1651, dA[0.019,0.006] dB[0.022,0.022] g[11.360,10.224]\n",
            "Iteration>1652, dA[0.419,0.146] dB[0.005,0.021] g[9.958,8.806]\n",
            "Iteration>1653, dA[0.045,0.131] dB[0.016,0.006] g[9.556,7.830]\n",
            "Iteration>1654, dA[0.028,0.022] dB[0.008,0.004] g[12.468,11.349]\n",
            "Iteration>1655, dA[0.059,0.014] dB[0.002,0.006] g[12.830,13.268]\n",
            "Iteration>1656, dA[0.033,0.012] dB[0.008,0.003] g[12.031,11.854]\n",
            "Iteration>1657, dA[0.012,0.012] dB[0.010,0.005] g[12.322,12.209]\n",
            "Iteration>1658, dA[0.016,0.007] dB[0.003,0.011] g[11.040,10.286]\n",
            "Iteration>1659, dA[0.007,0.004] dB[0.004,0.015] g[11.415,10.231]\n",
            "Iteration>1660, dA[0.083,0.060] dB[0.011,0.010] g[10.416,9.037]\n",
            "Iteration>1661, dA[0.020,0.013] dB[0.008,0.030] g[12.176,11.432]\n",
            "Iteration>1662, dA[0.007,0.013] dB[0.014,0.004] g[11.929,11.565]\n",
            "Iteration>1663, dA[0.092,0.116] dB[0.019,0.024] g[10.116,9.044]\n",
            "Iteration>1664, dA[0.015,0.012] dB[0.010,0.033] g[12.082,11.758]\n",
            "Iteration>1665, dA[0.013,0.010] dB[0.013,0.022] g[11.553,11.569]\n",
            "Iteration>1666, dA[0.272,0.179] dB[0.017,0.019] g[9.256,7.735]\n",
            "Iteration>1667, dA[0.142,0.041] dB[0.013,0.003] g[11.443,10.368]\n",
            "Iteration>1668, dA[0.027,0.055] dB[0.018,0.015] g[10.679,10.323]\n",
            "Iteration>1669, dA[0.034,0.011] dB[0.004,0.013] g[10.905,10.269]\n",
            "Iteration>1670, dA[0.029,0.011] dB[0.019,0.008] g[9.902,8.261]\n",
            "Iteration>1671, dA[0.009,0.060] dB[0.007,0.009] g[11.549,10.999]\n",
            "Iteration>1672, dA[0.030,0.024] dB[0.006,0.012] g[10.383,9.511]\n",
            "Iteration>1673, dA[0.019,0.008] dB[0.024,0.009] g[11.188,10.158]\n",
            "Iteration>1674, dA[0.019,0.025] dB[0.007,0.020] g[10.101,8.530]\n",
            "Iteration>1675, dA[0.008,0.039] dB[0.004,0.030] g[10.326,9.002]\n",
            "Iteration>1676, dA[0.013,0.015] dB[0.008,0.011] g[11.511,11.085]\n",
            "Iteration>1677, dA[0.496,0.237] dB[0.011,0.004] g[9.552,8.734]\n",
            "Iteration>1678, dA[0.037,0.172] dB[0.003,0.001] g[9.493,7.153]\n",
            "Iteration>1679, dA[0.033,0.074] dB[0.005,0.014] g[10.820,9.823]\n",
            "Iteration>1680, dA[0.021,0.027] dB[0.028,0.028] g[11.575,11.463]\n",
            "Iteration>1681, dA[0.046,0.062] dB[0.027,0.024] g[11.196,10.090]\n",
            "Iteration>1682, dA[0.053,0.026] dB[0.006,0.008] g[12.213,11.584]\n",
            "Iteration>1683, dA[0.014,0.033] dB[0.005,0.004] g[12.046,11.440]\n",
            "Iteration>1684, dA[0.021,0.008] dB[0.017,0.006] g[11.291,10.455]\n",
            "Iteration>1685, dA[0.216,0.111] dB[0.005,0.012] g[9.549,7.934]\n",
            "Iteration>1686, dA[0.071,0.255] dB[0.003,0.005] g[12.058,10.970]\n",
            "Iteration>1687, dA[0.177,0.160] dB[0.003,0.007] g[12.236,13.314]\n",
            "Iteration>1688, dA[0.179,0.078] dB[0.013,0.013] g[11.280,11.112]\n",
            "Iteration>1689, dA[0.059,0.048] dB[0.019,0.033] g[10.542,9.259]\n",
            "Iteration>1690, dA[0.028,0.018] dB[0.016,0.004] g[10.572,9.931]\n",
            "Iteration>1691, dA[0.015,0.018] dB[0.016,0.021] g[10.557,9.651]\n",
            "Iteration>1692, dA[0.008,0.007] dB[0.013,0.008] g[10.904,9.902]\n",
            "Iteration>1693, dA[0.005,0.009] dB[0.006,0.046] g[11.137,10.918]\n",
            "Iteration>1694, dA[0.010,0.005] dB[0.014,0.021] g[11.117,9.974]\n",
            "Iteration>1695, dA[0.010,0.008] dB[0.011,0.004] g[10.932,9.926]\n",
            "Iteration>1696, dA[0.072,0.030] dB[0.007,0.009] g[9.959,8.860]\n",
            "Iteration>1697, dA[0.027,0.008] dB[0.002,0.002] g[11.475,9.990]\n",
            "Iteration>1698, dA[0.019,0.017] dB[0.005,0.008] g[10.512,8.967]\n",
            "Iteration>1699, dA[0.035,0.118] dB[0.004,0.005] g[9.882,8.054]\n",
            "Iteration>1700, dA[0.005,0.018] dB[0.011,0.003] g[10.750,9.953]\n",
            "Iteration>1701, dA[0.012,0.016] dB[0.004,0.003] g[11.598,11.370]\n",
            "Iteration>1702, dA[0.373,0.312] dB[0.009,0.003] g[9.562,8.251]\n",
            "Iteration>1703, dA[0.093,0.013] dB[0.002,0.007] g[11.054,9.524]\n",
            "Iteration>1704, dA[0.330,0.024] dB[0.003,0.011] g[10.767,10.219]\n",
            "Iteration>1705, dA[0.048,0.066] dB[0.001,0.005] g[11.686,10.359]\n",
            "Iteration>1706, dA[0.036,0.022] dB[0.004,0.005] g[10.951,9.698]\n",
            "Iteration>1707, dA[0.191,0.148] dB[0.017,0.019] g[9.721,8.295]\n",
            "Iteration>1708, dA[0.010,0.046] dB[0.019,0.012] g[10.504,9.035]\n",
            "Iteration>1709, dA[0.011,0.034] dB[0.006,0.004] g[10.920,10.013]\n",
            "Iteration>1710, dA[0.061,0.111] dB[0.004,0.008] g[10.077,8.321]\n",
            "Iteration>1711, dA[0.045,0.016] dB[0.007,0.004] g[9.519,8.451]\n",
            "Iteration>1712, dA[0.011,0.012] dB[0.005,0.008] g[11.062,10.154]\n",
            "Iteration>1713, dA[0.147,0.282] dB[0.012,0.026] g[9.360,7.874]\n",
            "Iteration>1714, dA[0.005,0.014] dB[0.007,0.004] g[10.970,10.093]\n",
            "Iteration>1715, dA[0.388,0.186] dB[0.006,0.003] g[9.455,8.310]\n",
            "Iteration>1716, dA[0.012,0.037] dB[0.001,0.001] g[9.434,7.156]\n",
            "Iteration>1717, dA[0.051,0.082] dB[0.005,0.022] g[9.746,8.688]\n",
            "Iteration>1718, dA[0.046,0.007] dB[0.027,0.008] g[10.844,10.089]\n",
            "Iteration>1719, dA[0.075,0.205] dB[0.015,0.014] g[10.109,8.893]\n",
            "Iteration>1720, dA[0.008,0.030] dB[0.004,0.006] g[11.325,10.423]\n",
            "Iteration>1721, dA[0.009,0.008] dB[0.009,0.008] g[11.390,11.550]\n",
            "Iteration>1722, dA[0.010,0.009] dB[0.007,0.010] g[11.770,11.521]\n",
            "Iteration>1723, dA[0.102,0.039] dB[0.019,0.016] g[9.936,8.831]\n",
            "Iteration>1724, dA[0.019,0.122] dB[0.003,0.011] g[10.059,8.146]\n",
            "Iteration>1725, dA[0.063,0.042] dB[0.012,0.016] g[10.075,9.761]\n",
            "Iteration>1726, dA[0.008,0.021] dB[0.010,0.004] g[11.827,11.503]\n",
            "Iteration>1727, dA[0.008,0.034] dB[0.002,0.005] g[11.015,10.187]\n",
            "Iteration>1728, dA[0.004,0.027] dB[0.007,0.009] g[10.605,9.660]\n",
            "Iteration>1729, dA[0.012,0.020] dB[0.010,0.007] g[10.743,10.079]\n",
            "Iteration>1730, dA[0.038,0.042] dB[0.026,0.034] g[9.866,9.020]\n",
            "Iteration>1731, dA[0.011,0.006] dB[0.012,0.043] g[11.171,10.376]\n",
            "Iteration>1732, dA[0.012,0.003] dB[0.039,0.016] g[11.322,10.078]\n",
            "Iteration>1733, dA[0.004,0.011] dB[0.058,0.126] g[10.001,9.342]\n",
            "Iteration>1734, dA[0.013,0.007] dB[0.417,0.197] g[10.355,9.012]\n",
            "Iteration>1735, dA[0.017,0.032] dB[0.053,0.019] g[9.611,8.382]\n",
            "Iteration>1736, dA[0.017,0.032] dB[0.121,0.045] g[10.052,7.930]\n",
            "Iteration>1737, dA[0.014,0.006] dB[0.022,0.051] g[9.732,8.657]\n",
            "Iteration>1738, dA[0.010,0.006] dB[0.065,0.079] g[10.305,9.922]\n",
            "Iteration>1739, dA[0.005,0.028] dB[0.063,0.008] g[10.784,9.946]\n",
            "Iteration>1740, dA[0.014,0.049] dB[0.081,0.089] g[12.648,13.140]\n",
            "Iteration>1741, dA[0.011,0.008] dB[0.024,0.060] g[10.327,9.952]\n",
            "Iteration>1742, dA[0.007,0.008] dB[0.027,0.023] g[10.780,10.032]\n",
            "Iteration>1743, dA[0.061,0.016] dB[0.023,0.036] g[9.742,8.369]\n",
            "Iteration>1744, dA[0.017,0.115] dB[0.020,0.012] g[9.773,7.888]\n",
            "Iteration>1745, dA[0.027,0.150] dB[0.016,0.006] g[11.961,12.707]\n",
            "Iteration>1746, dA[0.276,0.342] dB[0.008,0.008] g[9.537,8.810]\n",
            "Iteration>1747, dA[0.059,0.061] dB[0.033,0.019] g[13.102,12.485]\n",
            "Iteration>1748, dA[0.183,0.084] dB[0.006,0.026] g[10.867,10.770]\n",
            "Iteration>1749, dA[0.014,0.109] dB[0.030,0.019] g[9.519,8.164]\n",
            "Iteration>1750, dA[0.089,0.011] dB[0.034,0.028] g[9.598,8.324]\n",
            "Iteration>1751, dA[0.032,0.006] dB[0.067,0.016] g[10.521,9.192]\n",
            "Iteration>1752, dA[0.013,0.009] dB[0.051,0.082] g[10.421,9.840]\n",
            "Iteration>1753, dA[0.006,0.029] dB[0.041,0.069] g[10.690,10.062]\n",
            "Iteration>1754, dA[0.015,0.025] dB[0.109,0.063] g[13.384,13.315]\n",
            "Iteration>1755, dA[0.021,0.018] dB[0.029,0.054] g[11.388,10.777]\n",
            "Iteration>1756, dA[0.016,0.052] dB[0.031,0.011] g[10.244,8.957]\n",
            "Iteration>1757, dA[0.077,0.025] dB[0.034,0.023] g[9.767,8.259]\n",
            "Iteration>1758, dA[0.014,0.020] dB[0.029,0.016] g[10.639,9.613]\n",
            "Iteration>1759, dA[0.018,0.009] dB[0.028,0.021] g[10.462,9.278]\n",
            "Iteration>1760, dA[0.007,0.042] dB[0.028,0.021] g[11.048,10.329]\n",
            "Iteration>1761, dA[0.017,0.050] dB[0.008,0.051] g[11.477,10.792]\n",
            "Iteration>1762, dA[0.027,0.042] dB[0.026,0.107] g[10.980,10.449]\n",
            "Iteration>1763, dA[0.004,0.013] dB[0.064,0.154] g[11.603,10.330]\n",
            "Iteration>1764, dA[0.011,0.009] dB[0.062,0.013] g[11.767,11.108]\n",
            "Iteration>1765, dA[0.349,0.474] dB[0.027,0.040] g[9.469,8.371]\n",
            "Iteration>1766, dA[0.014,0.031] dB[0.041,0.012] g[10.682,9.868]\n",
            "Iteration>1767, dA[0.347,0.111] dB[0.009,0.009] g[10.166,9.497]\n",
            "Iteration>1768, dA[0.054,0.056] dB[0.011,0.083] g[10.375,8.727]\n",
            "Iteration>1769, dA[0.018,0.031] dB[0.043,0.020] g[12.349,12.692]\n",
            "Iteration>1770, dA[0.132,0.104] dB[0.038,0.022] g[11.170,10.470]\n",
            "Iteration>1771, dA[0.046,0.014] dB[0.027,0.019] g[10.876,9.738]\n",
            "Iteration>1772, dA[0.040,0.029] dB[0.025,0.022] g[10.563,10.137]\n",
            "Iteration>1773, dA[0.005,0.009] dB[0.028,0.006] g[11.698,10.949]\n",
            "Iteration>1774, dA[0.008,0.014] dB[0.006,0.007] g[10.822,10.079]\n",
            "Iteration>1775, dA[0.005,0.019] dB[0.017,0.006] g[12.756,12.624]\n",
            "Iteration>1776, dA[0.003,0.013] dB[0.012,0.031] g[11.381,11.402]\n",
            "Iteration>1777, dA[0.009,0.011] dB[0.008,0.006] g[11.720,11.236]\n",
            "Iteration>1778, dA[0.006,0.003] dB[0.005,0.016] g[11.696,11.047]\n",
            "Iteration>1779, dA[0.128,0.165] dB[0.018,0.019] g[9.727,9.141]\n",
            "Iteration>1780, dA[0.021,0.015] dB[0.013,0.013] g[11.457,10.634]\n",
            "Iteration>1781, dA[0.079,0.022] dB[0.025,0.015] g[10.215,9.154]\n",
            "Iteration>1782, dA[0.007,0.042] dB[0.023,0.031] g[11.070,10.762]\n",
            "Iteration>1783, dA[0.006,0.016] dB[0.046,0.026] g[10.338,9.877]\n",
            "Iteration>1784, dA[0.008,0.010] dB[0.038,0.009] g[11.018,10.618]\n",
            "Iteration>1785, dA[0.006,0.024] dB[0.014,0.038] g[10.065,9.247]\n",
            "Iteration>1786, dA[0.014,0.005] dB[0.018,0.017] g[11.313,10.334]\n",
            "Iteration>1787, dA[0.086,0.106] dB[0.014,0.065] g[9.673,8.800]\n",
            "Iteration>1788, dA[0.007,0.011] dB[0.132,0.031] g[11.114,9.239]\n",
            "Iteration>1789, dA[0.016,0.010] dB[0.074,0.048] g[10.148,9.581]\n",
            "Iteration>1790, dA[0.017,0.003] dB[0.010,0.013] g[10.250,8.974]\n",
            "Iteration>1791, dA[0.020,0.005] dB[0.003,0.007] g[10.232,9.331]\n",
            "Iteration>1792, dA[0.004,0.003] dB[0.009,0.053] g[10.820,10.556]\n",
            "Iteration>1793, dA[0.046,0.155] dB[0.122,0.024] g[10.031,7.874]\n",
            "Iteration>1794, dA[0.256,0.349] dB[0.022,0.038] g[8.619,7.211]\n",
            "Iteration>1795, dA[0.018,0.045] dB[0.010,0.019] g[11.447,11.054]\n",
            "Iteration>1796, dA[0.306,0.131] dB[0.013,0.011] g[9.980,9.184]\n",
            "Iteration>1797, dA[0.078,0.057] dB[0.013,0.031] g[10.375,10.285]\n",
            "Iteration>1798, dA[0.046,0.007] dB[0.019,0.100] g[10.704,9.824]\n",
            "Iteration>1799, dA[0.013,0.018] dB[0.010,0.033] g[11.150,10.066]\n",
            "Iteration>1800, dA[0.008,0.026] dB[0.088,0.075] g[10.504,9.264]\n",
            "Iteration>1801, dA[0.049,0.010] dB[0.005,0.026] g[10.053,9.083]\n",
            "Iteration>1802, dA[0.021,0.007] dB[0.019,0.015] g[10.562,9.481]\n",
            "Iteration>1803, dA[0.013,0.006] dB[0.020,0.032] g[10.174,9.085]\n",
            "Iteration>1804, dA[0.007,0.003] dB[0.033,0.108] g[10.958,9.771]\n",
            "Iteration>1805, dA[0.030,0.158] dB[0.088,0.074] g[10.293,8.386]\n",
            "Iteration>1806, dA[0.499,0.162] dB[0.016,0.029] g[9.210,8.346]\n",
            "Iteration>1807, dA[0.023,0.008] dB[0.064,0.028] g[10.945,10.265]\n",
            "Iteration>1808, dA[0.211,0.054] dB[0.023,0.063] g[9.613,8.859]\n",
            "Iteration>1809, dA[0.015,0.115] dB[0.014,0.008] g[9.850,8.000]\n",
            "Iteration>1810, dA[0.048,0.026] dB[0.023,0.014] g[10.508,9.841]\n",
            "Iteration>1811, dA[0.012,0.004] dB[0.012,0.019] g[10.749,10.263]\n",
            "Iteration>1812, dA[0.014,0.004] dB[0.020,0.007] g[10.680,9.794]\n",
            "Iteration>1813, dA[0.006,0.011] dB[0.007,0.027] g[10.311,9.945]\n",
            "Iteration>1814, dA[0.036,0.031] dB[0.041,0.030] g[10.326,9.698]\n",
            "Iteration>1815, dA[0.006,0.013] dB[0.042,0.012] g[9.554,8.634]\n",
            "Iteration>1816, dA[0.031,0.019] dB[0.006,0.019] g[9.403,8.418]\n",
            "Iteration>1817, dA[0.032,0.016] dB[0.051,0.030] g[10.720,9.303]\n",
            "Iteration>1818, dA[0.061,0.009] dB[0.005,0.036] g[9.836,8.939]\n",
            "Iteration>1819, dA[0.011,0.014] dB[0.019,0.007] g[11.461,10.489]\n",
            "Iteration>1820, dA[0.007,0.007] dB[0.018,0.009] g[10.135,9.336]\n",
            "Iteration>1821, dA[0.056,0.301] dB[0.014,0.086] g[8.787,7.814]\n",
            "Iteration>1822, dA[0.026,0.103] dB[0.029,0.246] g[11.703,12.047]\n",
            "Iteration>1823, dA[0.400,0.318] dB[0.404,0.204] g[9.652,8.076]\n",
            "Iteration>1824, dA[0.070,0.024] dB[0.057,0.026] g[11.398,10.552]\n",
            "Iteration>1825, dA[0.095,0.043] dB[0.041,0.033] g[10.351,9.369]\n",
            "Iteration>1826, dA[0.026,0.024] dB[0.058,0.139] g[11.466,11.205]\n",
            "Iteration>1827, dA[0.010,0.018] dB[0.110,0.029] g[11.265,10.337]\n",
            "Iteration>1828, dA[0.379,0.499] dB[0.038,0.057] g[9.422,8.824]\n",
            "Iteration>1829, dA[0.080,0.021] dB[0.037,0.021] g[11.143,10.432]\n",
            "Iteration>1830, dA[0.089,0.025] dB[0.007,0.010] g[11.616,11.799]\n",
            "Iteration>1831, dA[0.107,0.044] dB[0.007,0.016] g[10.392,10.158]\n",
            "Iteration>1832, dA[0.023,0.046] dB[0.023,0.005] g[11.634,10.771]\n",
            "Iteration>1833, dA[0.035,0.024] dB[0.010,0.015] g[10.855,10.484]\n",
            "Iteration>1834, dA[0.032,0.024] dB[0.012,0.009] g[11.468,11.692]\n",
            "Iteration>1835, dA[0.018,0.021] dB[0.005,0.004] g[10.407,9.632]\n",
            "Iteration>1836, dA[0.223,0.295] dB[0.011,0.012] g[8.887,7.965]\n",
            "Iteration>1837, dA[0.042,0.009] dB[0.012,0.031] g[10.226,9.530]\n",
            "Iteration>1838, dA[0.053,0.009] dB[0.011,0.007] g[10.763,9.964]\n",
            "Iteration>1839, dA[0.013,0.010] dB[0.011,0.005] g[11.493,11.231]\n",
            "Iteration>1840, dA[0.006,0.007] dB[0.006,0.023] g[11.486,11.428]\n",
            "Iteration>1841, dA[0.395,0.231] dB[0.022,0.007] g[9.203,10.172]\n",
            "Iteration>1842, dA[0.025,0.091] dB[0.009,0.004] g[10.350,9.865]\n",
            "Iteration>1843, dA[0.009,0.026] dB[0.004,0.022] g[11.654,11.576]\n",
            "Iteration>1844, dA[0.437,0.173] dB[0.015,0.008] g[9.863,8.851]\n",
            "Iteration>1845, dA[0.065,0.054] dB[0.016,0.014] g[10.391,9.185]\n",
            "Iteration>1846, dA[0.018,0.012] dB[0.014,0.015] g[11.727,11.634]\n",
            "Iteration>1847, dA[0.043,0.015] dB[0.017,0.012] g[10.706,10.446]\n",
            "Iteration>1848, dA[0.014,0.022] dB[0.006,0.008] g[10.825,10.109]\n",
            "Iteration>1849, dA[0.267,0.151] dB[0.011,0.024] g[9.604,8.399]\n",
            "Iteration>1850, dA[0.060,0.018] dB[0.015,0.005] g[10.754,9.423]\n",
            "Iteration>1851, dA[0.399,0.175] dB[0.013,0.017] g[8.893,7.937]\n",
            "Iteration>1852, dA[0.034,0.037] dB[0.009,0.024] g[11.195,10.413]\n",
            "Iteration>1853, dA[0.053,0.038] dB[0.006,0.014] g[10.509,9.753]\n",
            "Iteration>1854, dA[0.015,0.024] dB[0.013,0.025] g[11.263,11.043]\n",
            "Iteration>1855, dA[0.012,0.017] dB[0.006,0.006] g[10.935,9.934]\n",
            "Iteration>1856, dA[0.009,0.014] dB[0.007,0.011] g[10.711,9.666]\n",
            "Iteration>1857, dA[0.229,0.140] dB[0.015,0.008] g[9.326,7.710]\n",
            "Iteration>1858, dA[0.020,0.015] dB[0.014,0.006] g[11.187,9.999]\n",
            "Iteration>1859, dA[0.061,0.014] dB[0.029,0.008] g[10.613,9.514]\n",
            "Iteration>1860, dA[0.005,0.076] dB[0.016,0.012] g[11.395,10.064]\n",
            "Iteration>1861, dA[0.452,0.052] dB[0.013,0.009] g[8.554,7.422]\n",
            "Iteration>1862, dA[0.038,0.037] dB[0.006,0.011] g[10.621,9.392]\n",
            "Iteration>1863, dA[0.130,0.146] dB[0.009,0.005] g[9.459,8.241]\n",
            "Iteration>1864, dA[0.255,0.064] dB[0.010,0.003] g[8.823,7.509]\n",
            "Iteration>1865, dA[0.036,0.317] dB[0.006,0.002] g[8.970,7.301]\n",
            "Iteration>1866, dA[0.173,0.031] dB[0.005,0.007] g[9.523,8.169]\n",
            "Iteration>1867, dA[0.026,0.041] dB[0.002,0.003] g[9.441,8.090]\n",
            "Iteration>1868, dA[0.008,0.038] dB[0.003,0.010] g[10.176,9.137]\n",
            "Iteration>1869, dA[0.194,0.075] dB[0.010,0.012] g[8.618,8.097]\n",
            "Iteration>1870, dA[0.014,0.114] dB[0.016,0.025] g[10.966,9.711]\n",
            "Iteration>1871, dA[0.113,0.069] dB[0.026,0.006] g[10.659,10.356]\n",
            "Iteration>1872, dA[0.262,0.296] dB[0.013,0.010] g[9.190,8.245]\n",
            "Iteration>1873, dA[0.061,0.081] dB[0.005,0.004] g[9.281,7.363]\n",
            "Iteration>1874, dA[0.203,0.162] dB[0.005,0.003] g[8.485,7.193]\n",
            "Iteration>1875, dA[0.077,0.059] dB[0.012,0.018] g[11.259,10.132]\n",
            "Iteration>1876, dA[0.257,0.079] dB[0.011,0.009] g[9.549,8.785]\n",
            "Iteration>1877, dA[0.033,0.028] dB[0.019,0.004] g[10.427,9.373]\n",
            "Iteration>1878, dA[0.058,0.057] dB[0.009,0.010] g[9.883,8.232]\n",
            "Iteration>1879, dA[0.023,0.015] dB[0.010,0.002] g[10.229,9.179]\n",
            "Iteration>1880, dA[0.050,0.124] dB[0.005,0.010] g[8.400,6.647]\n",
            "Iteration>1881, dA[0.029,0.055] dB[0.006,0.003] g[8.683,6.836]\n",
            "Iteration>1882, dA[0.043,0.027] dB[0.004,0.002] g[11.212,10.613]\n",
            "Iteration>1883, dA[0.017,0.018] dB[0.005,0.004] g[10.038,9.563]\n",
            "Iteration>1884, dA[0.006,0.016] dB[0.008,0.005] g[10.294,9.257]\n",
            "Iteration>1885, dA[0.018,0.116] dB[0.017,0.008] g[9.587,8.368]\n",
            "Iteration>1886, dA[0.340,0.125] dB[0.017,0.005] g[9.606,8.296]\n",
            "Iteration>1887, dA[0.028,0.051] dB[0.010,0.017] g[9.149,8.060]\n",
            "Iteration>1888, dA[0.025,0.043] dB[0.009,0.005] g[9.746,8.535]\n",
            "Iteration>1889, dA[0.035,0.024] dB[0.018,0.034] g[10.216,9.237]\n",
            "Iteration>1890, dA[0.019,0.017] dB[0.028,0.026] g[11.333,10.331]\n",
            "Iteration>1891, dA[0.027,0.012] dB[0.118,0.035] g[9.846,8.944]\n",
            "Iteration>1892, dA[0.020,0.017] dB[0.020,0.019] g[11.007,9.551]\n",
            "Iteration>1893, dA[0.025,0.032] dB[0.030,0.008] g[9.498,7.980]\n",
            "Iteration>1894, dA[0.018,0.017] dB[0.028,0.016] g[10.267,9.142]\n",
            "Iteration>1895, dA[0.093,0.173] dB[0.098,0.065] g[9.599,8.857]\n",
            "Iteration>1896, dA[0.029,0.119] dB[0.103,0.287] g[11.674,12.285]\n",
            "Iteration>1897, dA[0.545,0.287] dB[0.118,0.034] g[8.973,8.132]\n",
            "Iteration>1898, dA[0.109,0.086] dB[0.093,0.015] g[10.457,8.861]\n",
            "Iteration>1899, dA[0.411,0.112] dB[0.038,0.025] g[9.343,8.058]\n",
            "Iteration>1900, dA[0.179,0.156] dB[0.008,0.011] g[9.286,7.813]\n",
            "Iteration>1901, dA[0.032,0.026] dB[0.015,0.005] g[11.286,10.337]\n",
            "Iteration>1902, dA[0.371,0.106] dB[0.006,0.016] g[9.032,8.183]\n",
            "Iteration>1903, dA[0.025,0.098] dB[0.026,0.013] g[9.940,8.267]\n",
            "Iteration>1904, dA[0.022,0.022] dB[0.042,0.040] g[10.714,9.370]\n",
            "Iteration>1905, dA[0.080,0.022] dB[0.034,0.004] g[10.003,9.683]\n",
            "Iteration>1906, dA[0.025,0.005] dB[0.027,0.011] g[10.468,9.772]\n",
            "Iteration>1907, dA[0.009,0.009] dB[0.006,0.011] g[10.828,10.408]\n",
            "Iteration>1908, dA[0.419,0.128] dB[0.005,0.006] g[9.263,9.130]\n",
            "Iteration>1909, dA[0.113,0.030] dB[0.003,0.006] g[10.573,9.137]\n",
            "Iteration>1910, dA[0.041,0.012] dB[0.004,0.006] g[10.210,9.605]\n",
            "Iteration>1911, dA[0.043,0.043] dB[0.020,0.004] g[10.989,10.384]\n",
            "Iteration>1912, dA[0.033,0.054] dB[0.013,0.030] g[9.562,8.064]\n",
            "Iteration>1913, dA[0.007,0.009] dB[0.015,0.011] g[10.500,9.475]\n",
            "Iteration>1914, dA[0.066,0.006] dB[0.028,0.005] g[9.666,8.289]\n",
            "Iteration>1915, dA[0.029,0.060] dB[0.003,0.008] g[9.080,7.194]\n",
            "Iteration>1916, dA[0.022,0.045] dB[0.004,0.007] g[10.325,9.723]\n",
            "Iteration>1917, dA[0.295,0.342] dB[0.004,0.012] g[8.688,7.357]\n",
            "Iteration>1918, dA[0.018,0.027] dB[0.008,0.003] g[11.398,10.289]\n",
            "Iteration>1919, dA[0.067,0.022] dB[0.003,0.005] g[10.412,10.173]\n",
            "Iteration>1920, dA[0.014,0.038] dB[0.009,0.012] g[9.866,8.913]\n",
            "Iteration>1921, dA[0.006,0.010] dB[0.006,0.008] g[11.274,10.904]\n",
            "Iteration>1922, dA[0.008,0.010] dB[0.012,0.003] g[10.992,10.833]\n",
            "Iteration>1923, dA[0.003,0.020] dB[0.009,0.015] g[10.012,9.139]\n",
            "Iteration>1924, dA[0.342,0.219] dB[0.013,0.021] g[8.791,7.488]\n",
            "Iteration>1925, dA[0.048,0.049] dB[0.014,0.009] g[9.955,8.491]\n",
            "Iteration>1926, dA[0.109,0.022] dB[0.024,0.006] g[9.853,9.015]\n",
            "Iteration>1927, dA[0.027,0.094] dB[0.007,0.005] g[8.841,7.402]\n",
            "Iteration>1928, dA[0.015,0.016] dB[0.006,0.011] g[9.263,7.673]\n",
            "Iteration>1929, dA[0.004,0.005] dB[0.005,0.009] g[11.725,11.134]\n",
            "Iteration>1930, dA[0.090,0.139] dB[0.003,0.019] g[8.690,7.497]\n",
            "Iteration>1931, dA[0.006,0.023] dB[0.033,0.014] g[9.934,9.379]\n",
            "Iteration>1932, dA[0.047,0.024] dB[0.021,0.003] g[10.181,9.292]\n",
            "Iteration>1933, dA[0.018,0.013] dB[0.014,0.019] g[9.882,9.042]\n",
            "Iteration>1934, dA[0.114,0.322] dB[0.013,0.009] g[8.907,7.939]\n",
            "Iteration>1935, dA[0.006,0.042] dB[0.004,0.005] g[10.409,9.824]\n",
            "Iteration>1936, dA[0.042,0.019] dB[0.003,0.009] g[10.331,10.029]\n",
            "Iteration>1937, dA[0.250,0.219] dB[0.010,0.004] g[8.373,6.586]\n",
            "Iteration>1938, dA[0.122,0.026] dB[0.010,0.012] g[10.454,8.940]\n",
            "Iteration>1939, dA[0.072,0.014] dB[0.023,0.005] g[9.639,9.072]\n",
            "Iteration>1940, dA[0.076,0.163] dB[0.012,0.015] g[9.635,7.928]\n",
            "Iteration>1941, dA[0.010,0.013] dB[0.006,0.007] g[10.200,8.938]\n",
            "Iteration>1942, dA[0.006,0.005] dB[0.003,0.008] g[11.684,11.560]\n",
            "Iteration>1943, dA[0.050,0.050] dB[0.002,0.006] g[9.383,8.101]\n",
            "Iteration>1944, dA[0.013,0.013] dB[0.006,0.005] g[10.217,8.944]\n",
            "Iteration>1945, dA[0.010,0.016] dB[0.014,0.009] g[9.361,8.344]\n",
            "Iteration>1946, dA[0.002,0.013] dB[0.007,0.007] g[9.614,8.769]\n",
            "Iteration>1947, dA[0.003,0.006] dB[0.007,0.004] g[10.254,9.299]\n",
            "Iteration>1948, dA[0.011,0.022] dB[0.002,0.011] g[10.136,9.331]\n",
            "Iteration>1949, dA[0.007,0.011] dB[0.013,0.013] g[10.127,9.056]\n",
            "Iteration>1950, dA[0.161,0.087] dB[0.015,0.004] g[9.200,7.265]\n",
            "Iteration>1951, dA[0.063,0.105] dB[0.005,0.004] g[8.543,6.739]\n",
            "Iteration>1952, dA[0.477,0.050] dB[0.003,0.011] g[8.680,7.552]\n",
            "Iteration>1953, dA[0.008,0.266] dB[0.013,0.003] g[10.388,8.876]\n",
            "Iteration>1954, dA[0.232,0.022] dB[0.010,0.009] g[9.107,8.165]\n",
            "Iteration>1955, dA[0.069,0.014] dB[0.014,0.009] g[10.161,9.775]\n",
            "Iteration>1956, dA[0.015,0.112] dB[0.037,0.014] g[9.028,7.729]\n",
            "Iteration>1957, dA[0.279,0.038] dB[0.017,0.018] g[9.492,8.736]\n",
            "Iteration>1958, dA[0.035,0.183] dB[0.030,0.077] g[10.374,8.832]\n",
            "Iteration>1959, dA[0.109,0.019] dB[0.031,0.016] g[9.894,9.408]\n",
            "Iteration>1960, dA[0.028,0.024] dB[0.060,0.023] g[10.469,9.650]\n",
            "Iteration>1961, dA[0.003,0.006] dB[0.010,0.011] g[11.203,10.843]\n",
            "Iteration>1962, dA[0.007,0.005] dB[0.008,0.032] g[10.941,10.641]\n",
            "Iteration>1963, dA[0.131,0.124] dB[0.017,0.007] g[9.031,7.893]\n",
            "Iteration>1964, dA[0.005,0.037] dB[0.014,0.001] g[10.743,10.525]\n",
            "Iteration>1965, dA[0.039,0.010] dB[0.013,0.013] g[10.311,9.431]\n",
            "Iteration>1966, dA[0.282,0.322] dB[0.006,0.010] g[9.009,7.588]\n",
            "Iteration>1967, dA[0.035,0.006] dB[0.015,0.011] g[10.385,9.406]\n",
            "Iteration>1968, dA[0.206,0.078] dB[0.008,0.014] g[9.535,8.869]\n",
            "Iteration>1969, dA[0.029,0.044] dB[0.013,0.009] g[9.008,7.348]\n",
            "Iteration>1970, dA[0.083,0.038] dB[0.003,0.013] g[9.399,8.077]\n",
            "Iteration>1971, dA[0.268,0.254] dB[0.016,0.004] g[8.356,6.744]\n",
            "Iteration>1972, dA[0.017,0.013] dB[0.004,0.003] g[10.173,9.081]\n",
            "Iteration>1973, dA[0.032,0.005] dB[0.003,0.002] g[10.904,10.861]\n",
            "Iteration>1974, dA[0.096,0.077] dB[0.003,0.002] g[9.483,8.206]\n",
            "Iteration>1975, dA[0.043,0.019] dB[0.004,0.005] g[10.561,9.523]\n",
            "Iteration>1976, dA[0.019,0.013] dB[0.002,0.003] g[10.152,9.580]\n",
            "Iteration>1977, dA[0.007,0.009] dB[0.003,0.001] g[9.805,8.895]\n",
            "Iteration>1978, dA[0.010,0.016] dB[0.010,0.012] g[9.157,7.681]\n",
            "Iteration>1979, dA[0.011,0.093] dB[0.008,0.007] g[8.653,6.964]\n",
            "Iteration>1980, dA[0.018,0.014] dB[0.006,0.006] g[10.002,8.853]\n",
            "Iteration>1981, dA[0.061,0.012] dB[0.005,0.009] g[8.950,7.763]\n",
            "Iteration>1982, dA[0.053,0.093] dB[0.009,0.002] g[10.831,9.412]\n",
            "Iteration>1983, dA[0.059,0.063] dB[0.015,0.007] g[10.021,9.732]\n",
            "Iteration>1984, dA[0.043,0.045] dB[0.008,0.009] g[9.948,8.670]\n",
            "Iteration>1985, dA[0.011,0.013] dB[0.008,0.010] g[8.832,7.571]\n",
            "Iteration>1986, dA[0.033,0.013] dB[0.013,0.008] g[10.025,9.241]\n",
            "Iteration>1987, dA[0.063,0.066] dB[0.009,0.009] g[9.052,7.700]\n",
            "Iteration>1988, dA[0.095,0.320] dB[0.009,0.004] g[8.914,7.363]\n",
            "Iteration>1989, dA[0.045,0.026] dB[0.010,0.007] g[10.101,9.826]\n",
            "Iteration>1990, dA[0.323,0.243] dB[0.003,0.006] g[9.086,8.179]\n",
            "Iteration>1991, dA[0.029,0.008] dB[0.003,0.005] g[10.353,9.286]\n",
            "Iteration>1992, dA[0.062,0.015] dB[0.014,0.002] g[9.393,8.725]\n",
            "Iteration>1993, dA[0.010,0.017] dB[0.012,0.004] g[11.627,11.117]\n",
            "Iteration>1994, dA[0.086,0.122] dB[0.003,0.002] g[9.248,8.424]\n",
            "Iteration>1995, dA[0.011,0.018] dB[0.003,0.004] g[11.068,10.726]\n",
            "Iteration>1996, dA[0.005,0.004] dB[0.003,0.001] g[11.536,11.435]\n",
            "Iteration>1997, dA[0.016,0.011] dB[0.001,0.002] g[9.180,8.294]\n",
            "Iteration>1998, dA[0.019,0.033] dB[0.002,0.002] g[10.902,10.356]\n",
            "Iteration>1999, dA[0.005,0.013] dB[0.004,0.005] g[10.130,9.341]\n",
            "Iteration>2000, dA[0.246,0.440] dB[0.007,0.003] g[8.568,7.403]\n",
            "WARNING:tensorflow:Compiled the loaded model, but the compiled metrics have yet to be built. `model.compile_metrics` will be empty until you train or evaluate the model.\n",
            "WARNING:tensorflow:Compiled the loaded model, but the compiled metrics have yet to be built. `model.compile_metrics` will be empty until you train or evaluate the model.\n",
            ">Saved: /content/drive/MyDrive/Project/CycleGAN/TrainedModels/C_Highboost_Real_Model/g_model_AtoB_002000.h5 and /content/drive/MyDrive/Project/CycleGAN/TrainedModels/C_Highboost_Real_Model/g_model_BtoA_002000.h5\n",
            "Execution time is:  1:18:51.727201\n"
          ]
        }
      ],
      "source": [
        "from datetime import datetime \n",
        "start1 = datetime.now() \n",
        "# train models\n",
        "train(d_model_A, d_model_B, g_model_AtoB, g_model_BtoA, c_model_AtoB, c_model_BtoA, dataset, epochs=25)\n",
        "stop1 = datetime.now()\n",
        "\n",
        "#Execution time of the model \n",
        "execution_time = stop1-start1\n",
        "print(\"Execution time is: \", execution_time)"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "Oe8RwHXNIwgl"
      },
      "outputs": [],
      "source": [
        "from keras_contrib.layers.normalization.instancenormalization import InstanceNormalization\n",
        "from keras.models import load_model\n",
        "from matplotlib import pyplot\n",
        "from numpy.random import randint\n"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "2gPLwZLMI5D4"
      },
      "outputs": [],
      "source": [
        "def select_sample(dataset, n_samples):\n",
        "\t# choose random instances\n",
        "\tix = randint(0, dataset.shape[0], n_samples)\n",
        " \n",
        "\t# retrieve selected images\n",
        "\tX = dataset[ix]\n",
        "\treturn X"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "5fqxqTXuQKIq"
      },
      "outputs": [],
      "source": [
        "def show_plot_b(imagesX, imagesY1, imagesY2):\n",
        "\timages = vstack((imagesX, imagesY1, imagesY2))\n",
        "\ttitles = ['Real', 'Generated', 'Reconstructed']\n",
        "\t# scale from [-1,1] to [0,1]\n",
        "\timages = (images + 1) / 2.0\n",
        "\n",
        "\t# plot images row by row\n",
        "\tfor i in range(len(images)):\n",
        "\n",
        "\t\t# define subplot\n",
        "\t\tpyplot.subplot(1, len(images), 1 + i)\n",
        "\t\n",
        "\t\t# turn off axis\n",
        "\t\tpyplot.axis('off')\n",
        "\t\n",
        "\t\t# plot raw pixel data\n",
        "\t\tpyplot.imshow(images[i])\n",
        "\t\n",
        "\t\t# title\n",
        "\t\tpyplot.title(titles[i])\n",
        "\tpyplot.show()"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "yOGOZQNSQOKR"
      },
      "outputs": [],
      "source": [
        "A_data = resample(dataA_all, \n",
        "                 replace=False,     \n",
        "                 n_samples=50,    \n",
        "                 random_state=42) # reproducible results\n",
        "\n",
        "B_data = resample(dataB_all, \n",
        "                 replace=False,     \n",
        "                 n_samples=50,    \n",
        "                 random_state=42) # reproducible results\n",
        "\n",
        "A_data = (A_data - 127.5) / 127.5\n",
        "B_data = (B_data - 127.5) / 127.5"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "fGSJ-VSOQbEK"
      },
      "outputs": [],
      "source": [
        "cust = {'InstanceNormalization': InstanceNormalization}\n",
        "model_AtoB = load_model('/content/drive/MyDrive/WarpedDreams/CycleGAN/TrainedModels/C_Laplacian_Real_Model/g_model_AtoB_004000.h5', cust, compile=False)\n",
        "model_BtoA = load_model('/content/drive/MyDrive/WarpedDreams/CycleGAN/TrainedModels/C_Laplacian_Real_Model/g_model_BtoA_004000.h5', cust, compile=False)"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 273
        },
        "id": "5RMj8iw-QgMx",
        "outputId": "dfdb83cf-309f-490f-9d40-e4812ea0f0c4"
      },
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 432x288 with 3 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {
            "needs_background": "light"
          }
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 432x288 with 3 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {
            "needs_background": "light"
          }
        }
      ],
      "source": [
        "A_real = select_sample(A_data, 1)\n",
        "B_generated  = model_AtoB.predict(A_real)\n",
        "A_reconstructed = model_BtoA.predict(B_generated)\n",
        "show_plot_b(A_real, B_generated, A_reconstructed)\n",
        "\n",
        "# plot B->A->B (Sketch to Real to Sketch)\n",
        "B_real = select_sample(B_data, 1)\n",
        "A_generated  = model_BtoA.predict(B_real)\n",
        "B_reconstructed = model_AtoB.predict(A_generated)\n",
        "show_plot_b(B_real, A_generated, B_reconstructed)\n"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "nW_umxpWQnqB"
      },
      "outputs": [],
      "source": [
        "# test_image = load_img('/content/drive/MyDrive/WarpedDreams/CycleGAN/Dataset/Test_Train_Split/C_Unsharp_Real_Dataset/testA/n02958343_12557-1.png')\n",
        "# test_image = img_to_array(test_image)\n",
        "# test_image_input = np.array([test_image])  # Convert single image to a batch.\n",
        "# test_image_input = (test_image_input - 127.5) / 127.5\n",
        "\n",
        "# # plot B->A->B (Photo to Monet to Photo)\n",
        "# monet_generated  = model_BtoA.predict(test_image_input)\n",
        "# photo_reconstructed = model_AtoB.predict(monet_generated)\n",
        "# show_plot(test_image_input, monet_generated, photo_reconstructed)"
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "Real"
      ],
      "metadata": {
        "id": "V05HYmg6FEiB"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "modelh = load_model('/content/drive/MyDrive/WarpedDreams/CycleGAN/TrainedModels/C_Highboost_Real_Model/g_model_AtoB_002000.h5', cust, compile=False)\n",
        "modell=load_model('/content/drive/MyDrive/WarpedDreams/CycleGAN/TrainedModels/C_Laplacian_Real_Model/g_model_AtoB_004000.h5', cust, compile=False)\n",
        "modelu=load_model('/content/drive/MyDrive/WarpedDreams/CycleGAN/TrainedModels/C_Unsharp_Real_Model/g_model_AtoB_002400.h5', cust, compile=False)"
      ],
      "metadata": {
        "id": "OvB6PK4Pt6qc"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "from numpy import vstack\n",
        "def show_plot(imagesX, imagesY,imagesY1):\n",
        "\timages = vstack((imagesX,imagesY, imagesY1))\n",
        "\ttitles = ['Sketch','Real', 'Generated']\n",
        "\t# scale from [-1,1] to [0,1]\n",
        "\timages = (images + 1) / 2.0\n",
        "\n",
        "\t# plot images row by row\n",
        "\tfor i in range(len(images)):\n",
        "\n",
        "\t\t# define subplot\n",
        "\t\tpyplot.subplot(1, len(images), 1 + i)\n",
        "\t\n",
        "\t\t# turn off axis\n",
        "\t\tpyplot.axis('off')\n",
        "\t\n",
        "\t\t# plot raw pixel data\n",
        "\t\tpyplot.imshow(images[i])\n",
        "\t\n",
        "\t\t# title\n",
        "\t\tpyplot.title(titles[i])\n",
        "\tpyplot.show()"
      ],
      "metadata": {
        "id": "Rt-ECE0PbVvf"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "#unsharp\n",
        "from keras.preprocessing.image import img_to_array\n",
        "from keras.preprocessing.image import load_img\n",
        "test_image=cv2.imread('/content/drive/MyDrive/WarpedDreams/CycleGAN/Dataset/sketch_car/n02958343_1104-2.png')\n",
        "test_image=unsharp(test_image)\n",
        "test_image = img_to_array(test_image)\n",
        "test_image_input = np.array([test_image])  # Convert single image to a batch.\n",
        "test_image_input = (test_image_input - 127.5) / 127.5\n",
        "gen_image = modelu.predict(test_image_input)\n",
        "image1 = cv2.imread('/content/drive/MyDrive/WarpedDreams/CycleGAN/Dataset/realistic_car/n02958343_1104-2.jpg',1)\n",
        "image1 = img_to_array(image1)\n",
        "image1 = np.array([image1])  # Convert single image to a batch.\n",
        "image1 = (image1 - 127.5) / 127.5\n",
        "show_plot(test_image_input,image1,gen_image)"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 145
        },
        "id": "1QGMYENxbk3F",
        "outputId": "ab709bcc-2276-4116-9184-383df6efc540"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 432x288 with 3 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "#lap\n",
        "test_image=cv2.imread('/content/drive/MyDrive/sketch_car/n02958343_14365-3.png')\n",
        "test_image=lapfil(test_image)\n",
        "test_image = img_to_array(test_image)\n",
        "test_image_input = np.array([test_image])  # Convert single image to a batch.\n",
        "test_image_input = (test_image_input - 127.5) / 127.5\n",
        "gen_image = modell.predict(test_image_input)\n",
        "image1 = cv2.imread('/content/drive/MyDrive/realistic_car/n02958343_14365-3.jpg',1)\n",
        "image1 = img_to_array(image1)\n",
        "image1 = np.array([image1])  # Convert single image to a batch.\n",
        "image1 = (image1 - 127.5) / 127.5\n",
        "show_plot(test_image_input,image1,gen_image)"
      ],
      "metadata": {
        "id": "2p6nT-PejnO4"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "#Highboost\n",
        "test_image=cv2.imread('/content/drive/MyDrive/sketch_car/n02958343_14365-3.png')\n",
        "test_image=highboost(test_image)\n",
        "test_image = img_to_array(test_image)\n",
        "test_image_input = np.array([test_image])  # Convert single image to a batch.\n",
        "test_image_input = (test_image_input - 127.5) / 127.5\n",
        "gen_image = modelh.predict(test_image_input)\n",
        "image1 = cv2.imread('/content/drive/MyDrive/realistic_car/n02958343_14365-3.jpg',1)\n",
        "image1 = img_to_array(image1)\n",
        "image1 = np.array([image1])  # Convert single image to a batch.\n",
        "image1 = (image1 - 127.5) / 127.5\n",
        "show_plot(test_image_input,image1,gen_image)"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 145
        },
        "id": "PFXZkQv4312_",
        "outputId": "781a38c2-5463-4610-898f-8d66fc3414ce"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 432x288 with 3 Axes>"
            ],
            "image/png": "iVBORw0KGgoAAAANSUhEUgAAAV0AAACACAYAAACoX7ryAAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAALEgAACxIB0t1+/AAAADh0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uMy4yLjIsIGh0dHA6Ly9tYXRwbG90bGliLm9yZy+WH4yJAAAgAElEQVR4nOy9eZhkR3nm+/vibLnW3tVd1fui3iSQEJIRkhiQMGIVmDE2GIMBGw8MZjz24G3MfWxfj/G1jed6uXgB3XsBb9iW7TEGgQUCtAu0S91SS2r1vlR1LZlZlevZIuaPk1mVVV3VarW6q1rd+faT1V/GOREnzjl53vPFGxFfiDGGDjrooIMOlgZquSvQQQcddHAxoUO6HXTQQQdLiA7pdtBBBx0sITqk20EHHXSwhOiQbgcddNDBEqJDuh100EEHS4iXHemKyIdF5N5zfIwNImJExD6Xx+ng3ENE3iAiR5e7Hh2cv1gKTmnHeUu6InK9iNwvIlMiUhCR+0Tk6pdQ3pJe2A5ePETkoIjURaQiIqMi8iURyS13vS5WiMj7ROQHIlIVkbGm/QkRkeWuWztE5E4R+ehy1+N0cV6Sroh0AV8H/h+gD1gN/J+Av5z16mBJcLMxJgdcAbwK+O/LXJ+LEiLyKeBPgM8Cq4CVwMeB6wB3CetxwbU2z0vSBbYCGGO+YoyJjTF1Y8y3jDFPzt9RRD4rIveKSHfz8/+JyIiIHBOR3xERS0R2AH8JvLbpRZWaedMi8j9F5FDTo75XRNJtxf+kiBwWkQkR+fSSnHkHABhjRoHbScgXEbmm2fIpicgTIvKG1r4i8hER2SMiZRHZLyIfW6ZqXxAQkW7gt4FPGGP+yRhTNgkeM8b8pDHGFxFPRP6w+XycEJG/bD07LUlHRD7V9JBHROQjbeWfTt5fFZFR4Isi0isiXxeRcREpNu01zf0/A7wO+Fzz2f5cM327iHy72Up+VkR+vO34/SLybyIyLSIPApuX6trC+Uu6zwGxiHxZRN4qIr3zdxARJSK3AK8EbjLGTAFfAiJgC4mXdBPwUWPMHpK39APGmJwxpqdZzB8CrwauJfGofwXQbYe5HtgGvBH4jSZ5d7AEaD5UbwWeF5HVwG3A75Dcp18C/llEVjR3HwPeAXQBHwH+SESuXPpaXzB4LeABXz3FPr9H4hxdQfK8rQZ+o237KqC7mf4zwJ+1Pcenk7cPWA/8JxKe+mLz+zqgDnwOwBjzaeAe4JPNZ/uTIpIFvg38HTAIvA/4cxHZ2Sz/z4AGMAT8dPOzdDDGnJcfYAcJiR4lIdJ/I2nifBj4AfAPwD8DbnP/lSTyQ7qtjJ8Avte0Pwzc27ZNkdy8yxc49gbAAGva0h4E3rfc1+VC/gAHgQpQbl7/7wA9wK8Cfz1v39uBDy1Szr8C/7VpvwE4utzn9nL6AB8ARuel3Q+Ums/M64EqsLlt+2uBA23XvA7YbdvHgGsAOY28AZA6Rf2uAIpt3+8kca5a398L3DMvz+eB3wQsIAS2t2373XZuONef81YvMYl3+mFImgrA3wB/TPKwbQEuB37IGBM0s6wHHGCkTedXwJFFDjEApIB9p6jGaJtdAzqdOuceP2KMuUNEXk/iqQyQ3NsfE5Gb2/ZzgO8BiMhbSR6orST3PAPsWtJaX1iYBAZExDbGRADGmGsBmiNBVpJc40fanjUhIbSZMlp5m2g9PytOI++4MaYxs1EkA/wR8Bag5S3nRcQyxsQL1H898JqWjNiEDfx18/g2c3nh0CLX4ZzgfJUX5sAY8wyJ13tZM2kPSTPymyKyrZl2hMTTHTDG9DQ/XcaYS1vFzCt2gqSJsaR6TgenB2PMXST3/A9J7u1ft93XHmNM1hjzeyLikbR4/hBYaRLp6BskD3IHZ4YHSJ6ldy2yfYLEk7207X50m6QD9IVwOnnnP6ufIpH5XmOM6QL+QzNdFtn/CHDXvN9Lzhjzn4Fxkpbz2rb9151Gvc8azkvSbYrgn2oTy9eSSAXfb+1jjPkK8OvAHSKy2RgzAnwL+J8i0tXUfDc3PSaAE8AaEXGb+TXw/wP/t4gMNzvcXtt8iDs4P/DHwJtImrY3i8ibm/cp1exwWUPSk+7RfJiaXu9Ny1fllz+MMSWS0UJ/LiLvEZF883m6AsiS9HvcQqKdDwKIyGoRefNplH0mefMkRF0SkT6SVk07TgCb2r5/HdgqIh8UEaf5uVpEdjQ9438BfktEMk2d90MveFHOIs5L0iXR9F4D/EBEqiRku5vkjTcDY8yXSXpZvysiG4CfInkInwaKwD+RiOUA3wWeAkZFZKKZ9kskzdCHgALw+5y/1+SigzFmHPgr4OdJvK5fJyHXI8AvA8oYU25u/0eSe/5+Ev2/g5cAY8wfAP+NpHP5RPPzeRJ9/f7m/88D3xeRaeAOEm/0dPBi8/4xkCbxkr8P/Pu87X8CvKc5suFPm7+Jm0g60I6TyIS/T/JyBvgkidQxStKa+uJp1vusQJpCcgcddNBBB0uAjlfXQQcddLCE6JBuBx100MESokO6HXTQQQdLiA7pdtBBBx0sIU45OUJEOr1s5wmMMWdz3Gnnvp4/OKvjiV915ZXJvTVJwa0bvZDdPsh1vn2qfOe6jNPNd36UIQitS97cagSweezxHyx4b8/bGWkddNDBi4cxzLCGwTRtSewmrbTsGXY+yT5538Xss1/G6ec7P8pgzt+EdjVzQ7jMRYd0O+jgQoJJ/ogxINJkAZ3YLUZeyJYWUzdd5BnCprlvu32uy6DN9TzNMuYce6nLmFueAk41FLdDuh10cAFhxmdrjzPesmXmzwvYzLVPSuuUcVJaWz7TUhgWQacjrYMOLji0msoJXqx9pvnOtzKW49gASKL0LoYO6XbQwQWElpM21wGTk6xZW06yZ7cvZp+vZcwt70zzvZT6M2MvLi90SLeDDi40mIU8scXaxAulvVh7KcswC6affI6zqYllFkidm+9Ueyxcxvx8SZo5KW0uOqTbQQcXEEy7w9Xmg83ihUjvdLCcZczPtxj9tZOzacu1kG3m2YuVcSof2bR9n7t9Pjqk20EHFxBane4LE8hiaadjG5avjPn5zAJpL1TWmeJU9Wn/3uaJm1Mfu0O6HXRwAaH1vM+livlN8hcsoc0+E4/0bJfRymfa7Nm9Wn7qLAXOtVt5Frc5yV7YN57NZ2Y0XHMyvUpHXuigg4sGsqC88GL01fkdQqebbznKaMdccp85Z2n+kXk27bbM2NJmM7+Mk0tvq9N8ZbcjL3TQwcWBput3YcgL8zH3rMwCZNzyQA1gJHE5jTTT220Wtk3TBknSWmXMpLeObWZ83JOU8xeQFzqTIzro4AJCy9+bpzLSPtxpFi/W8zyTfC+mjJNlkDnUNTPTbVYGEFHNEpppMzPJmv6kosmWBpTMs5uHkza73bs1itk3WHKEhHhnp/i2T0ibsWQByaENHdLtoIMLCDKHsxKCOV0Vdflxck3bg8lIkyBFBBGFIIgxxEoDMbYyGN3cj2b8hKZkYIxpcqvAQnaTjEWSfWfzJUdHFMQJ6WoRbBQmFmKjZz1bmZUZThUqrEO68+A4Dv39/Wzfvh2lzr360mg0ePjhhwmC4IV37qCDF0LL85K2LzMbFvaBT89u78R6KfleqIx5Oc3MySBKASohUKUJiRhaPcglO7bT4+TRVkRQiVAOoATRCpRBGxANYhm0lsT5bRJ0y0YLIk1PNlaI0mgEMYIo0BosLWS60+iGYteTjzFyeJw6DYyYNu1BZt3fRdAhXSCdTrNp0yZe85rXsGLFCo4ePcquXbsol8vn/NhXXnklw8PDHDx48Jwfq4MLH3OpNbGWV154sfmgpZhCQniiBCUWmIhYh+S7Pd76tnewadMWLr/8KrZu3nrKjquzD8MX//IW/uhP/wwrZTHj3c6RJhavz0VNuuvWreOd73wntm3zzDPPsGfPHu666y6uuOIK3vjGN1Iqlc55HWzbZmpq6pwfp4OLA3Ni23CyfT4joVkz+9IQhWWBjgWciOEtQ3z0pz/G9rXbWL9+A67rzuSNwoh6vUaywnu7Ty1N59PMcqFhcbvdUZ1xt2VW6hUIA5+9R/diUlZyvXX7FV6oo3AuLjrSVUrxmte8hquvvhrXdfnWt77F8PAw27dvx/M89uzZwxNPPMHXv/51fN9f7up20IY5D8OcDfOSFmCZkx6B1sO20MPRJowu3q1znlLZnGv0UuWFFysNnGkZ807BJLO/lK0IIp8tGzfx0Z/7aa69+nV05bqR5v2J44iRY4cpTE7w3Tu/yz33fpfQjzCxRlIKIgE04lgYX4PSYNvga3AMiIUJYpQLxliYMEZ5gokEo2PEsyAAIwZxLGI/wnLBD6AwViFr24RxnMgL7Z696XSkzcG1117LK17xCm677TY2b97Mu9/9bh599FG+9KUvMTU1dco4mB2cG5h53+b4DWZ2n1bIPN0MDyvMfmZ6kduf7bYH2zDb5yy0Pe5tB28NC2r9BtRMelvjVRYZJnSe4OzKC2cqKbzYMmZrb0wiJ1i2wmhNb3cvf/D7f8TmLRtm9vIbDb5/79384Ilvc/vtd1KuBtjaxpIYYtDaYMSgEHTTfVWGWVtD3NwuZtbGGLQCyySdaS17Jl8MkRh0rFDKIjK08UUb8QpIR15IsGPHDt7+9rfzpS99iQ996EN873vf47Of/Sy1Wm25q3bRY8YHMic30Fo9yILBGI2FAR2BiTFhSFivo7XBclwQg1gKUWp2OJGlELFALJQo2okYhFhkhpjEyKyTKIAkPeTNXZuVkYU54zzAC8sLi3mYp+O9LoQX2zG3SCmtzn8EpQRtNNlsjl/6zV9nw6Z1yT465uEHvs1f/cPf8uDDT6AjhWMJVqAIifGNAKr1S0E1/zcGlCTkmbxwF7ZBMJGZuYYmmh1ZloxyUMm+rbfagg5aR16YwfDwMB/60If4i7/4Cz7ykY/w5S9/mX379i13tTpoYvbxlDleKUYjJkYR06gWqBbH8Csl6sUxasUJpiYL1MplQj9Aa02sNVEcI8pCieA4Lulclkw2TzrXSzbfRaa7m1Sui3RXF24qhUplUK6HKBsxFohD0oPTRhQzb4XkUX1hGlkmtFoG0qplK/GFiHZeAYvap5YGXriMRYi4pbuKIBagDZ/8+V/iba9/E6p5H06c2M9v/fZvcWRskr5UjroIdR+UsjEmblMwWrqwmdVpMbMLQ7S1ptrt1vCxVvVaw3eT+rWW8Gk/pUVaEh15AUSEd73rXXzlK1/h537u5/jWt77F5OQkPT09lMtl4jhe7ipe3JjnMbQ8WwMoHVEcH2Hv048RV8fpsiKkPoUKfVQQ0KNjutMG41poDX4jpFypEDRCfN8nMFANQ2Kt8SNNjIDtoMVCuR5d3d2s6l1B/6phMv2DRE6KzIpV9KwYws31YNvOzENvUMQm8X6UmdWDRc4f+p1LZQsR5PkiKcy7Zs2xsUopdOxz9VU3cfPb3jxz7QG6u4Z4980f5MFH7mP37mdbGcHEbS+beS8HWcReSGNedN+5EhacfFXnoCMvwObNm+nt7WXfvn3s2rWLrq4ubrzxRnp6esjn89xyyy0diWG5MTMg3aDQGKOZLhZ58pEHqU5PsHaoj5UrVlMZPcj40VFMGBMGmjiKkx7rWDc9DINBY9mCi4Nl2WjtoY0hkBjLdrAcD2MEbaDhVzi2f5zndj1CaARx0mS6+hlas57M+k2sWree7u5uBAVOmlzfChzbBWUt9xVbEC2uaBdQmLHnk/CZ2mejjAXqbgARdGB4801vJp3y5mxPZ3N89JM/x7smf5xP/uzHOHDoUOLNa6E17GC25Bfy2F9o+3x77susbU4c8oLlz8VFQbrXXXcd3/jGN/jZn/3ZmfGwQ0NDFAoFSqUSAwMDHD58eHkreZEjGbKjEWLCRpXnnnqSA3ufYc3QSjZcsgFlRYRhjVgbypUGJgjRWtDGYHSMiXVzJpEmiBvEcYzRoFSAiKBsG2W7xNrQKFebOp+FbdtIXw/dw6sQDZEfM12YYt+j36f22INku7vo7ekhVjZTgWbr5Vdx3ZvegaTzzZlR54+XC01KMy15oZV26uVjTi5hMftsljG/vERYEqWwlM2lmzbMpNfrVRqNBrlMFsdL09/bz4aNG9l9YB/unCb+vBbTIkc7tT3XA56VH9rtxY+z0FnPx0VBuo7joLVmZGSEO++8cyb9Pe95D4VCgTAMl69yLwOYee0qDTNjv5Oui9mNs3HzF3tAzcyecx+TpJk4OnKE7997J7mUxc6tG/FcC60D/HoNv1KiOlUiFsF2HXQYEDYCoigmjiLQIMoQ6YBYa0zcVIZFSKXThAHETXI2CFpHaOMTVyoYk4yZdhyHdH+e3uFBbK2pl2sURg6hEbr7V3Diud08nu3iqjfchG27c66NNI+1rJD5xnzCPb/khdn+yiTAjKUNmWwvfSsGADhy6Dn+r9/7FZ7df5gbrr2RX/6138V1HFwUtrISaQFmOjvbf3lzA42/mDOZ69MuZLfj5Ot76pfcBU+6PT09ZDIZtmzZwpNPPkmxWJzZ1mg0MMYQRdEy1vDlgLbemeby1NpojIkRbYjigLBeJWhUMFGAMhG2iRJPUCksy0IphVgWKBux7KTTynLASr5HQcRTTzzOrscfZse2S+jrzaPE0AiqhH6VcmEc8St0ZRUbXr0Nz7aJY0OxOMWhg0cZOT5GHJtkyJBRaN0cOmQ0Bk2gGxhtoXVTr1OqGVlqRtkgDgLCRoNqpYxRinQqQz4/QKZ/FSeOHMB2S2jf8MgD97Jy/QbWbroElIsSab54lr97rdVZNN/re+nSwGIdaWdaxjxLaA7fiti8bTtdAznA8LWv/S/uv/8p8mmP2772PX7qA+Os3TiMsQxKCcTS9OylrbylkEdOsd200hbGBU+6K1as4Mknn2TDhg3s3bt3uavzMkWzCScGTEhYLVMZO874of1MjR9nanKC+nSRuFFD6QhlIjwrQlkKy7axHQfHdXFcF8vN4KbSuKkMqUwWL5vBTeeo+iHF55/nDa/cTCbjIuJTr9eolUtUK1Pk0w4rVgySdsG1wEgICtLdA7hZl8MjR/F1DFqhmhGm4thgNGhjCIMIrcMknoYkGqBukm6oBEsEq+mfxMagtUb7AY1wknKpjAkj/FBz4NhRjjeOcGSqwQc+/NNs23nZvOu0vDg78sJiEsDZLyOpb5OgLEUcaXZu3Y7jOkRhwEOPP4FyMzRiFzEaHScOUhyHMyM0klbXyXLF4nU6nfqfzrkslJq8PU413P+CJ92BgQFEZEZi6ODMYDBE2ufgs7t5+I5/p3roIOFUAd+voMMQiUOIAiyTvOljrCQSlGp+LIVtK7KewfMcHDeF7bp4mTReOoOb8ujOZigfeIowk8V1UyjbJhMH9HfnyGRcHFdQSqPQYCu0MYhlsW//QUrTNVw3SxxppPlgJkPIzKw7a3QSvMSYuZ6uUkRRhGm2eCzXxXIcfB1htKFSruISUw+FQgOOjE1xpPgY5frn+fmf/3m2btsGcn4MI1s4ytj5KC/MTRYRFMl9WLNiEAEqxQIThw6hlWBpjTKGOIgwxhBHAcbo5jTclns/V1w46SBntf6nlh0uannhkksu4YEHHmDNmjWdoWFnCG0SvXXs2BH++vOfo3r0EBnfRwUhRhuMTkYbaB1htCbpz7WSt31z7GPS6QRKaWzbwnIsRAm24+CmXLpTFp7r4KZSpNIZ0pkc2VwWL+2RzrjU0h6pjIvtWdi2wvYcbCdFuVJn354jOGSJfBtjIvJdKTDCVGmKWIOOE+IVBcRxoklrTbMdjtIRUaOBiWJEhLjuk8nniR1F4PtEQUgmk6ZUDSg0oNQwGL/OE0/u4ktf+hKf+tR/o39gBcjShl1ZEE0P6/yJMrbYsdvqZcCIwSiDlbLYfOlWAOrVafxGBctYoIVYm5kpwpbrgiVInCyNbhbU0ufXeTH7LF+Di11eiKKIcrlMtVpd7qq8LGEAxBD5DXY9cB8nnt9HUCowXq8TNEKCSKNNnBCOUjiOg+vY5GwLWwTRBhtBNTuvImwgRiRGKYOIj2UJk7bgODaWpUh5LulMmkwmhec5pNMu6UyKTC5JS2dTpNMplLLwQ8OGofVMV31K0zUi7XPZK7Zj2w5P7drDyMgJtEmIV/TsygCt7j4AoggTGyyliONkGJoVBFi2i1/3EW0QsSiWG+hUD10rc+x7djfpdIo9e/Zw/3338bZ33IxlL//jtDi1vlRP9mx7wO11NohJZJ+M18/gumQGmh8E+KEGy8bEYFtJ+EUEHMtGsDDoeRNB2s/6XNZ/1p6/TnASxPwi9XRt22bt2rV4nnfeDe053zBDQAuIUdoYRo8e4ejeffTlepBcD5brYjdJMopiGr5PrVajUi4zPV2hUA6I/QbEEa4oUo5L2vVwxWCLYKlEyxPRxEqoiYUmwFGQ8mJS5RDHreLYgus6eCmHdNojlXbJpD1yGY9MxiOdydPfnae3p5fhIU09rNFoVHFsl7XrVjM9XWa6XE3mz+vmPHqgqQZigFgbLMfGVhYSaYyOqUcG148I/QClDUEYU6w2OOFPY+d66O7pJYqTsm77xm286qqrGF6zdlafbB1liX93L6cg5qYlQJNM/RUDa4ZX0tuXB6BcKxCKRmuDrQSjDZpkGrdtATpKSLhdNZSTjHOOk1VqTvodtOOCJt2NGzdy4MABbrjhBh5++OFF91uKYOUvBzQb3EjTTr4bTORz/MA+ntq9hzsfeJSiH0FgcNCIbbBsh3w2R08+w6r+Htat3cBQ/wpq9WkMMbVGg/FCiWJhmnq5go4iHFGkbBuv+TEqIYdQIIxiGj54LijRIH5Cim4Nx4FsxiGb8sh6KVLpKWw3kSjSmRSpTAYVWOhAEwea1UNryXfVKU1NU63ViHwfo5tjQg2JzmsiFBZBEKGUk/SIi8XUVAMr1DgI037ElIZG5NNlNO//8Z/ga1//ZyzPZWRynG999w7e//4P4NgOSajtZcJZlRfOZZOcprozmx7qBq/YcClZzwMMj+/eTaQilE4BFqBwXAcRId2TSoKat2b/vmh54VxJLMy97AvggibdoaEhHnvsMW666Sb27Nmz4D7Hjx9n586dnDhxYolrd/6hfaxjy1JA1KgwcfwIE+NjZPM5dEpTLZQx2hCLwg9D4mqVcnmKyfFRrrx0G5uHV+C4GfYeOEA9jEl19bJj3WayaQeFoTo9zdTEJBNjY0xOlwkjja0s0p5L2nFJWZIELNGCjmKswMdtCLEt6FrAtKomLwVJmqeWrbAchWc7eI5HNpcnnc3hZjJ0uTm6Bruoh3XqjTr1RoNarYYfBARBQIyL1hZBEOKlUmglaLHw/WlSGKyUR6UWUG0E1GuGvr5B3nDjG3jwoXuJoxBb2dx55938h9ffwIb161lKL2s+5lJrYp25vHBumuQzFNWczIKlUI7CTMXsvPJVQCItfO/uu4mNjbIUYjS9+R5yvV0gsG7tmuTcjG7+SpmZ0Tb3SrzU+p/6XJIjnSwvXLRBzLdt20axWMS27ZM60VzXZfPmzezevZtcLrdMNTyfMBs8uvVRGKKgxuFndvPEQw8wPV1k5YoBukNNPZUhRlMOAhqNBo5SxPUKr371K7ls83rqU0XWbNxEOYp4/Om9PP/041Rqdby0kMvmGOjrZWjFIJe8+iry2SxREFEqFhgbGWG6UKQ4PUU+m8VyXGyErIAnFp7tIigaQUhIjFIK21LYONjiNFstQr1Wp1atNYd/GWzXAcfCdh1SmTTp3j40hkqtTnHaR8TB9myUcoiNUK3XUWKjdUAoFpUwwo8C6uUak5NjGEvo6umiWBin4QdUteH273yXn/7gB1C2M7+dv2Q4f4KYn2IshzGgEulFUIjYaN1gy7adXHPjDwGgo4ipiQl0pNFWiNgRP/SGm+ju6wKgHpTRYYQlMivSz8HyygsXZUdaf38/fX195PP5mSm+IkIul+PSSy/l8ssv59lnn+WVr3wlt99++zLXdvmRhFRM3tkamr3EMeXCGPueeJRqYYJipcz44RF0nPgWdspDOR5dmQwZz2HHq3Zyxc7NqLBO98o+amHA4Jr1ZI4VcbPTWLFQ0zHENmMHR9n9/DHAkEml6M1nWL1qJWtXr2Xnq64i5Tg06nXGChOMHT1IFPlEJqTcCFB2miA2iNjYWGDZ2MrFsz2yuRRePtGareZ8fK1jtDY0goggipgulgjjGJTC9Tz68l1EWhDlEsZQbSQvErfZU14NYxoIUaxRkU9YKaL9MmsG+yiOH8GyFbEtPPz4Y7zljTeyfu1aFGp5+hGaz/rczqX5O5xreWEhWaOtek39KlnrDMQRgorhXe/5SQb7ewFwXJtrr7sS78ludmzZwNXXvZbrf+gmHMtCxzFH9o1hoZonajBmoSnZ50IeeaEymB3AsAguSNJ1XZePfexj3Hvvvfzoj/4oX/7yl3nb297G5ZdfTqVS4cknn+SOO+7AdV1+5Ed+hPHx8eWu8nmC5MfTaqXFoc+JIwc4uOcpdBSTynezOrcCKzLUG1X8KGC6XMGvTFMm5u3/4Wo8Iry0w9jUNFdc8UMUaoa+A6P0TU7hOC6FShVLOURR4ks7joPYHmNTVcaKz/PYU8+S8Vxy6Qxrhoe54rLNvOnad3Limd3UJsaoVHwmKhENIwwOr8GxLBq1KpUooFytYQU1KJOMoEinyGfSZNIp8qkUeVFEYUgUx8RxTBgG+A0fvzYFRjCWSxwZgmod3Shjuw627TIyXSF2PLTRpIhZaRse+fq/kCpNsC7tUNKamm2oV6e56+67+MBPvB+s5VF15z7+7fJCe9rp2i+1GT575JM6asWgLIWyhKhe4003vYkf+49vmyFO2/b4hV/8HyQxkK05hBo0Ghx+aj/aUkicTIiZe7RzK4+8oC3m4ooy5jgO73vf+3j44YepVqtUKhXWrl2LiPDNb36TrVu3ctVVV3HVVVfx9NNP88UvfrEzfrcNLWlBTEylMM7+p3dx4vBhBIvxySJT5ZC85eBlPFLZLMNdXThKSNuGtGfT25Ujl0uj8r1YqRwre3uo1r7L+NgJsp7H6v4kXOJQbzcN3yeMIqqNOpVYk/JS5NJpcpkMYgz1ICJvC8eKhG4AACAASURBVOWj+xiwQ1Ire6j0KQYjB2/9JVx5ww+j45hGtUqtPE15qkipOEm1UqdRrdColJkoTROPV0i5FpmcQ3c+TzaVRrQhtm2M6yESE0aaqh9QacS4XS4pK41jeVTqAYXpCvbKLBhNznNYk7I5et+dpFybLscincvjpz1Kcciep/cwMT7B6qGhZbl/c+UF02a3cKadSO32Qt7fSTU5KY80K2jQCArLEvwwYPulO/jP//W/kE7PjSpmWbPf4yjCD+rseupB7vjOHewZP4Ql9sxy6XOrcTrncjaux2JltP9/Mi4o0nVdl49//OPs2rWLdDrNDTfcwDPPPMOmTZswxrBq1Soee+wxbrvtts643floNvmS109MefQAD33zn5natw9dLtOzcoAu18F3XEpBCaohtfESjhWRdhXrh/qx0x6Vep10NoOxHYJYOPTsASYnpzDaMDVVIC5BFGo8N4XrpsjkMgwOZBkeSmHriEw6hdfVhQ4DrrlkHWsdnzVZBy+dpx5qSjVFbnALN77nxzk2dpSBVSvJ9Q5Sq/mkbE19Omm1hOUqQblM5PtUKhVK0yWma0WkXiGYqiQLT2iD6xlqgWAUOCkPTzxSSuEQUg1g79HjBLh4xiGlNbmczZp+mxWuwsQRjXodvxxRtxRRqotqrcaTT+1m5eAgtrUM4R/nyAutpHMRZWyx5vZsa2l2G7NNbqEZ9U0RRCGbtq/ld3/7M2xeuxFIvNjJ0iTKJIGJwihi/95n+Nd/+xcOjBxnYvQEfi0iLTaWgqgZaXxhBfl0z+WlXY/FaHgxXDCk63ken/zkJ2cId8eOHXz+85+fCWbj+35nGvApYJgddqPCgOPP7mLvYw8STVXQUUxPPsuO9avRe8fRYUQ67eGFDrbjYlma4sQUjYamv7+PTNcgXV09/OCxp3lu/3GmpyqsHBzCUTGxCSmXazTqPhBTLI4hU5BSWVzPxe+NWbuql9du3UbX+GG2Dq6kp7eb6VrASKlGeu0arn3ne0n3dlPbu5vHDx1i044raWiLyZGD2KbB8Lp1jJ8Yw9KGbCbDwIZBeo0hCCqY6THSRlOcnOLOO+6gL5/CdjKIgWq9TiNUdHV1k+vtJxVCOn2CRmRhWS65bA/TE8cYma4zsGU93dksdnEKXQ0o1nzsrI0fRzz33HO8/rrrsdPpZbmP8+WFczMNeC7hzvWB2ylX01qFQ0SwLAt0hB82eO31V/HpX/8frF21BoCx48f4yy/8Abff813SIsTGQmKFDhQRIToikRJEUdVJSM851Tjjc3lp1+OijDK2ZcsW3vve9/Lwww9j2zY7d+7kT//0TwmCYLmr9vKBgRiTdDzVq1RPHCcsl/AbDbBtJk4c54ot69i58VLGKmOMj43y3L5RKqIplafQaO5+/AnecP11lEfGGd+9Fz+2OHJ8hAOHDhMHDbJph1TGIpXOku3rSeIrqCRoeVCLEWzChs9brn011omnueqSPrpTOXy3i+nQUEx3c8Pb3kN2aJhj+/dQLU7Qne1m9fAaQuORsW1K40cYPT7B4aNjZDJpTKFMdGSUeqPB1VdcTqYrwImmmT5UpFSYwvJjcllDJuOQ9lJMRxGNWHDSGRxlyOfyTE6WibVh3eatHLUdjlSmOPbEASwMGYSsl6YqYHDpXdfF5OQkhUKBzOrVc0M/LkHHmjT/zFNSme+Jzk07XXt+ma2U2UDeyZTcNk1TFEJL9rAJggaZLof3/sQH+eD7PsxgzyAAk2PH+M3f/gUe+P5T5FyLqjHERM04okmcjGR1PJXEWphh+JaKu5gnu5g8cjYklsW2L3ydWnhZk+6KFSt45zvfyfDwMH/3d3/HNddcQ19fX4dwzxBJ1xboOMAvTxE16sksoDiiVpxExXVq0QCqO8u2LVu48tJXUSXkxOQEI+Nj+L7P9+/+HkZsvK4BCqUKx0+MI5bBStmI6zBRqhBMVlC2h+u6mDikK5cil8+wfe1abrj8MnqnRrl062oyElGatqhGLvHASq554zXk12ygETeYKpbIZnOsXL0GraFSq9O/YiVpV8h1deFl8liex9DwEEEUUZwsEgcB+d4BwumA5/cfwA8iKnEdx3Uggor2eeS5g4ibZ2VfD73pLFFssCybSrXKieIUOo7RKo3lOmQyWYwI2Z5urDikbEEYR0xMTHDo4EHWDA83NcelC3ae6PFnU144WRE+ybtt9bw2ISikuaKjoMA2GBPheIa3vuUmfuLHfopLNm/HsWyM1hw4sIff+exneOz7T5FNp6mHgpknGZiZVSHMTEUWHkuwNJLC/L3nty4uOHnBdV1e//rXc9VVV3H77bfz/PPP8653vYsHHniAv//7v+90jJ0hBAXGUK/VqTQaBGGEaBulNf09LpddcQkD29/CvmKVkWPHKB4dQcI6A7bL5m2Xkk2ncZWiVKtxpB6QdR2qxXHKpo5lKYhisrZNNpPC8tLEcYQJNLpWZcXKPFds20CGGpdvHSbvBZwoNHhmIiLKC69/x5twB1aCpTCNEC+dI3AzuNkubC/N1OgItXLIQE8GZbsMrd+Il83Q09vblJaEnKdIxVNUxkc5dHCUdCaPoyDQMWhB0hm87h4Oj0wxUSyzurcfN52nP9VNJdaMFo4QBBGe5+HX6oxPV0kpm4FVq9m2dSM65fLs/v00Gg0OHTrEa6+5Jpk1tbQ3cZ5xduQFA7MjD1TC6iqZ0td8oUhCtCQvboOgrQBLFEZCLtm+lV/4L7/KFZddjucmHWRB0ODf//Wf+JNb/pzSZJmsnaYWJq0tjMyuujsHs1Tbrqi+9NgSL3wNTk9eWDythZcN6Q4MDNDf38/VV1/NwMAA+/fv56tf/Srvf//72bVrF1/4wheoVCrLXc2XNVQz8PfxsXEmyxXqocYz4IhhZX+aoaEeMqtXsWJlhvVXvIZUBIqYY0cPc/TQfo4dP0ZYmUIR05/12LhpPVdsWku90aDuN5icmOTE2ATFSo1AFOXpaVYOreDKHdvY3JcnUBHXvvWH6U6VGHn6afY+N82RWhevv/EG0v0riO1EI0ylUqzbtJVxzwPLJYgiVq0eIu1ENKYLVKpVuvr7sFNeEjNXx8RxQD3QhNUqjz36DOXpiK6UjVYhJgITCZYRBvtWMD7uk/My1H04Nn4cUikaxHR35bBsh7ff9CYs1+Nr37uLIweP8eCTT7J737NsvfwybNehUfY5fvw4vu+TSqeXdLzu2Q9izmyMBGnplQpRYEyyyLmykihfSStA46QsrnjVK7n00svoTfeSy+W44YYb6O7pb5aneX7PE/zDv/4j//Qv38DTFp5yqCOgExI3CO1xQIR57vtp138Z5AXTSlsY5xXpep6H3YzUpJRix44ddHV1cfXVVzM+Ps7Bgwd57rnneOSRR9ixYwc333wzt9xyC4cOHVrmml8YiHUSgGb/3mcol6bQsRBg8NH09aTwHJtCNWCsWMLEmnJd8cpXvIr+DTl0qourb3gzcRhQLIwxeuB5Jo+NcOL4KLVqmZQYhvJdbNo+QDqXw/JcDBESVamXTtAdVNhyzRtZuWkbBx+9hyd3j3L4SI0gn6XqhxgNQd1n7OhxGsVpXCV4KRdLUkSNBrbnoQDXzaCMS63SwDOQz9v4tSrd+SxjJ46x+74HeOiu+3AwoAOCKAIE3/cYmZjgaLGGZVl4jlCYqhIA9UoZY1vYQBBG3PbNf+fGN97Ie9/9Lv727/+BTC7LZHESTUw+k6ZW8qlUytTrdbx0urke29Lcw7MjL8jJ1sxqvYKyrGTcdpeNncxygFjYvvNS3v3j72Vl7yCbtmyjp6trbslaMzE5yt/81S187Ru3USrVyeIRWIZAG5JOt2TB0FYUmzn0Ju1EtvC6aIvLAS9GXmgd+fTKOEleeDkEMU+n07z1rW9lw4YNHDt2DEg6HcbGxti3bx+FQoHh4WHWrFlDJpOhXC7zyCOP8LWvfa2zvtlZhCaZ+lspl3BtlyjQGAfEVmQyKRqB5sHHd7PjlVcwevAovQNrmC5P09edZaIwzsDKXkbGRvjhN7+Jp3YN8oaBlRTHJpkcH2Xy8AGef/gHHNi9G1sEx1Gk0xb5NKxd2cPG4QF6sg6HHnuY7/zrbZRPjBBKBi/fR2n8CMWxdfgGHr3/AY4+vZtqaZIojkBZZHp6yPf2smp4mNdccy0DQ6txxaVULpH2rGSdtSBi/PARHrrnPuJGg7QFGE1koN6IOHzsKMfLIdrNkMukqYYh2raIfJ/YaJSyibVOxhVXq9x59z1s3b6d7dsuYapaoThdQgRsW2EpoVqtUCqV6O7tw8jSBcA5u0HMmRkHK6JQYmFZyYvnzW+/kY9//BN4doqWAJvL5sl35efkr1VqlIoT+GGD++/7Jn/zD19lZGSMrPJwlU1Da4ym2fHWlCvmn9Ni53qSx3k653e69unvO/f6mgXS5mLZSbe3t5dPfOIT3HPPPRw6dIihoaGZHt/LLruMjRs3EkURe/bs4Tvf+U5n6Nc5hDSb4vV6gzjSaA2xn4zD7R5cg+pZiR1mqPsBaTdLX1c3W3ZcQjbnUa4XWblmFSplMV6c5Dv33sOrLnsl4+OTbNm6hfX9XbhWzGPHnsWTJIJYLqXI2YZcEKD9iIfv/i6jU2V2rOxn9c5Xk+rpw/QMkR7O49p1enpWcOVl61irxpg8WqNYLKIErFRArTLGw7c/xnMPfZ/tr7qcDZdswU6lqJdOUC5N8fhDD/PU/fejK2V60yliv0akNTGKY5M1jkxWibwclp2iHGq0ctEWuNkMOghwMmmCMAQdYwRGR0fJ5HMMDg3R39eHNgZbWVhiY9suQRByYvQE6zduWtqbaGbv5ZmPXmiWgZnxmEUUKIPfaPD299zMr33qv5PPzCVYAK1jntn7NEf376fRqHP3XXfy6K6HCLXBryrERGRwCATiWM2qFgLGLESi8+sOC5PfYhIAi9hnKim8UBmt3c5TeaGvr49f/MVf5B//8R+57rrrMMbwjW98Y2bkwQMPPEChUFjOKl5UsATCOKRRr2LiCMuyqNZ9PNfh6GSDnVdsY+fqlTz4g0foc3voMYbi2CRCL0rZ1OoB2WwP/b0ruHznTkQbNm/ZwomxMbKimZgsYCwBy6A8wfIEJVCpVBl7cg+hitm2eS1rex0Gex3sfIzJlYnMMai7OBmftT111m7NooaHqE9lmDhRIMZQKMf0r+2hWK0x8uj9HHjkPkKxcVJZgkbIdGma2lgR10Qox0Icj1pDU6w0OFKsEbk5YidNtruXarVCXUuyOnDKo4HBKIHm+mrKsmiEIdPlCpY7SaaRZXBgADvlIdgosbGUTaFQnJ0x9aKatmeOxan1xXQStdGLAWUJlm0RBAEf+OiH+OiH/9MM4T63Zw+3fet/EfoRacehVCly9/13UihWsW0XK7SwJCSMQLCJjCJEWrNwkvjG8/jp5Jq0L3/edpYyo3qcQkA5s86xF9PB9rIJYt7X18cv//Iv87d/+7dcf/31FAoFbr311rnLfXewpLAU+FHA4OAAR44eIIxjGkGEqWoeeGwfo/oh1l1+Da+7+kpEp1m3cQvjIxMQRKS9NEcPH2PL5ktQseCaJDTj5Ng4hYlxBtcMMz1RwEdhO4rQVgS2gw5DpioB2q+zcsAlGDtIjQamJ4+JQ/CnsWwhGPPxS4eRoIpdn0Q1SgQnDhCMjJFK9zKUWUVuMM3ohKZcD/DFphopIj8mnK5TGysxVahiwhp+vcrAqpVM+7D/eJEw1U26pw/xUni5HCEG4ohYYmzXIWt3oY1BOTFOykPHMcpxiI0mDEMa9TqlQhEvkybf20s2k8NxXIrFItoYLFm6EQxzpwG36OgM0OI1SeIjxIHPf/yxD/CJn/0EmVQajGHv7if5zd/8NZ4/fpAMFoG20KKwteCIS1RL5BsjNi2SnEuwp6E1G5oLoiZFzI5Om23Gy0mvmnaSfLFa7ovHQq+r8zKI+Xvf+15uvfVWrrnmGkZHR/nqV7/aIdxlROunW65WKU6VGZ8sEWKh3DSRiSlMRoz++wN8/zsPkevN0btqI8ObN7F+4xac9BaGBns5dHAvcbCG++6+i+Mjx+jpGaBUq9LT10fkB5SLU1i4KGWjxRCLS90P8asxQ115jF8kjupMjAqp7kF63EEcYxC/ju1pxCqj4pi4ETB29BilY0fxa3Umx6eJUnWkaw2lag0dxrjZbk4Uphkr1ilN+RSKFfxIsMRhfLpGYE9RqAYExsPNdqMtGxGDNhHZfBYdR4RhAxyLtONSLpdp+D5KKVzXZUUuh21ZOI6DMck6cYVCET/UDA+uIuO5VCoV4ihC2Wq+2HruMKcF/lLkBYMYg7EgJmbVmnX8zEd+MiFcDE898QN+5dO/ysjxIhnHpq6dZBFQSabmGq1ArNmymDMYYV7n3snE2B6mp7WnaY13bvUW6uaeRtG+KOVca6ELsxTywuJYFtJdtWoVmUyGgYEBqtVqh3DPA5jmv6lGzO6jJSb8DNUoRxwHuLFPRlm4WDhRTPn4KOOHR9nz6A9wMmnsbIahDeu5ZOeljPf309OdI9Ozk30HD9Hd28fze54hPbyG5/cdJePXcb1+LANRCFOlEptXD+P5RcJqQF3FBLrBc3v3szIwrBwaIpV2kCjGGI3SmrASMnFsinrJp14LmA6FQlygUbQpTPv45RrijFGsBpTrmoavqPkRsRj683ni0OHE2BRTkcHq6sHOeGRyWZJALDrxlTwXoxJN08tlKVUraEsRGU3YaGBE4XhesnKwZSOWDQRgNOVKGdfrJxahVq/T47o0V5lZgvvYTgnCLH29OHmhBYVFFAW86x0/xvCqVQA8//gT/Oqnf4UTxwqkXJda2CRRSUItGmOQOT345iS6mq3tycc2rb+GJE6DapKtGDQxgsLYGi92iIwmao7LFy3zuE9OouHTvwYvTV4474KYr169mnvuuYdrr72WL3zhC8tGuK7rkk6nqVarMzEaLmYYIFIudbeXRpdGmzxBtURUnyKIfPxqBRdIqSTqbmV6CjM9hWXbFI6P8uyDj/Lt1D/TNzjI8CVbuOTSS1nX38fG666jdHyMkRMFbL9MJQpYNdhDxoqIg5hapYIJfPyqgLHw4wrBiQb7RusMrqmw/pJLSOUsbKXAr2L7dQqTFYKpED+AaijUopiJyQkKVU2sLRrBNMp1iYIYHTcdTRMxPTWFjSIIDOJ5aMfG8mymyyUK4xPEcUS+p4eBVStJpTOEUYzteLip9IznppoPWRBp4loDpSyM0eSyGYyJwTLYaYdYGaq1Kn3d3Ut2D89qEHMRlAgSw/qVqwGo1cv88f/7hxydOEFaZWnErV1n6TRxsptUtEDrfuFRB22YGTEhWJZKVg+JI3rX5FnRv4KNA5vYt/85Ro+dILYE1xXihp5ZXqpZyLzjLEzwZwMLq+Hnmbywfft2du3aRblcplarLfnxBwYGuPnmm+np6WF0dJTBwUEOHz7MbbfddvFOH5ZW889gWYrYaII4olSuMDk+iiLGhCG5lEd3NouyIMh0EfgBphHgVipklSJWFeKJAsV9z/P0XXdi5/KsXruWsNZAh1UiNJOlacKwQX82hRsZjtQLdDmCLR6OkyEUTdmPKZcrPHHwGaL795Pv6aKnJ0faCulJC7rSwAocjC/EsSLGxYkdssqiahRKWYg4xLpBcwV2ojjCaKhFPgEgjkOmK0+5WmH/3r2EDR8wjE1MMlkqsXHrdjLZHHEIGS9HPmMTxxodRygg1hoR0FpTLddxbAcUzZCVAZatKEwVGB5alQRbXwo0n/WXFsScpgRriEnWkhtsrthw+OAhHn7wKTw3TRApQCMq8W5n6M20yLf9+PM7w05RsxZHWkJERFe+i4/8zId4x9vfgeek8WyXRqPBdGmKW/7i89xx17exHZuGhJiQ0xgU/VIkhXa75UfP06ZP7hucg2UhXdu2KRQKHD9+fMmPvWbNGj7+8Y9z6623UqlUOHToED09Pbzuda/j05/+NJ/73OcuzqDmJmkV+bU6pclximOjjB89jmMLfhRTLpdQCOW6z1QjxHVsUukcdjqDk9aYwCcwMcYY/DgkGzTQYQMzVaJxYhTPsliVSaGbupwQM12qkcIQYcA2eDa4KcB2EOVgWRrXtoki4fhImb2HCzSiBqJiXBeynkPOdsjYaTzPIzIusXKJxCIwTkK0oolMQKQjRBShiYiVIlKCl81gbIsDzx+i0WggurUMe0ShWMTs289lr7gc13YRz8J1XOJYY1mCUmC0IYoiJguTRGEyljeKNcVSCWUperryjI6NsGXzZnJeZqlu49ymLmcSxJyZfBaCbnthTE5N4kc+StzEq4fWrN2ZfAvL16caL9yOxF9NApwr/FqDX/iNX+NH3vGWOTP73LRHV283/8fv/Ca9f9HHX//Vl3GdFA0dJbPamh13s+d+diWFuVd0XvrFFsT8VLBtmw9+8IN85StfYXh4mFKpxIoVK7j++uvJ5XLceuutfOITn+Azn/nMRSk3KCCs1wmrFRrTJaqlAmEUEYVVJAxAFHEcEyiL8nQFkWks28FLeaTTKVJpF2Up7Ngh/t/tvXm0XVd95/nZe5/pTu/dN096mp8GW7Zl2cgyYAdkMxkzmIApApUYgtMklU6vrEVWVa9QVKUqHXqtSharq7tDQpGQpmsBAQfTVDAYA7bxEDBGEpYt25Il2dKT3vzene894+4/zr33jXqWbOl5ut+1pLvfvefss8+593zPb39/v/37EWFLSQKBGYbgehD4hGFIpCOiKMAQGiUFhhQEQsfT9kAghYzT+iExRYijFJ5h4AobXyQoeC6VSjXOwUCAJXwGBzuxU52kkm2YEdSKRdAhfgBBFBKGsbMrEhAICA2FmU4yPjtDqVRCiZhIIZ7W6iiiVCxQKuTpbO/AlCZRGBOMIQ2UIVCqUXtPgBYYZoJIRBD5lPIlnnnmGMqy2L59J+mBtSHdcycxfzFH2rKe4s+1Jop8Kn68xD4Ka7HMIjRNoVrU8y+8qEy49PMFuqgmLqdOLGlIoQj8Ch/48G/yzv03LllKPT9u07L4vTt/n+dPnOThnz2KaUKgBVoLZMPkXLSS7XwdYi/HqbbSuc7jDUW6l19+OUePHuXEiROYpsknPvEJTp06xbe//W1uvPFG+vr6OHnyJPv37+dHP/rRKz3cNYcGAt/DrZYxpMaUAivh4OJCAFpHRFrHpCnjcj74VWqupFAyMG0bISSObVFzHFKWwgwDbA1pJ0HGbKNbRkgZEUU+BC66VsHUEYFWRAhCbaADhRcJapEgjCTZrh6SkYFZcTF8H6MKWSeDmUpRKpYwDAfa+ujeshNhWPRkM4yPnmFufIJ8oYCMYnJHSyKtCQRgWyjHYfqF56lrBHFp9sZUOYrw/BqF4hyp1A7KpRqlUiHON1AVGKZBMpnEMAza2zpoy3QQovF0SOS7eLUyZ8+Ok25v4/kXTrNxYHgt/GhL5IXGW+ezDHi5PSyIH1JhGOGXq0A9YECq5nFirhUL/2BlIl9KVss/jx1wsYUbRT6dXYN88lN3kkqllmy7uP9kMsm/+b0/5MDhg3jFCkIJdBhb6lrMuxMXPohY1F46ppW2WWm7+riXjOjFnj1vGNLt6enh9ttv54tf/CJbt26lVCrxi1/8gq1bt/Lnf/7n3HPPPTz66KMYhsGf/MmfcODAAaanp1/pYa8ptNZk2jIopQjCOMayv7uX4fVXMnrqOONnz1CulPD9GlLGiaUNVf/ZBfG6ea0VhVKVklHANk0cw8AxTFLJBJbn04GPYyqyKYfe9jRDbetIRD4zuTxV1wNl4gWCWiWkFmhcH2ZPTxJJhdOWprc7Q8ZTuAGYdpLIyVCNJIlsB1deczWbt48wsmWIw08e4wd3/w8K+WkolDG1jGUMGSdnSaTbGFy3nrfd8l6OH3uaB3/6UybOniHS9TiOKERhEvgBSioCP4jTXEYBCEEYBnG8rlQYhkEqlcRyEhhSoOwEVaU4M36Was1jamaGSM/rmZcyAU69TiMLKffClgEv6AuJAURSopw4IbsScTpGLQWRaBY/n6dTscDy0wuZuXEI0TBrF2if86V8lDKIQpdsZzt//hd/wabhddDQi1e5blsv286nP/X7/Ne//C+YlsQVNJPnrEyW53LmrTYTWKm91GFXt9jf6PJCT08Pn/3sZ/nKV75CoVBg8+bNfPCDH+TBBx/knnvuAeC2225jbm6OJ554grvuuouPfOQj/M3f/M0bKpRNC8Hw0DpGNm9n/MwsWA75UoUtiQ627ryGdcObmBo/zcSZMxQKxfie0hGgQSq8wMWyHKIwgFBTDX1cJEUhKbkehhTMOhIrsDBLJdom5tjWleCKjX3sumYH3b29ZLK9jE/mOHToCDNzRYq5GoVyjUIQMDYzTX7sDG4YYRkmSSeJlWrHbs+yc9tm9uzbTV9vB73ZBGrvLg4eOEDu5HNkEwLhVzg1OYPQcdRKd+8g23fuYmT7Zey4bCdvfsuN/NO3v8XDDz2A9l0s08SybEQk49LuqLgcuA5QhkGkNWEY32C+H5DPuzi+h+048YKKVJKhwSGUMqjWaoRRhFTqklu7FyfLWN0y1JqIOGzONC2AuPIxGstQBD5IoQjrKR6XHabRV+OBoxvBa6LOxwsWNojY/6Xx6Wzr5HP/8T+w79pr5kcsIPBqPPizH3Dg4GEG+ru47UN3NK1gIQS333Yb37v7Hzl1arT+MJBxNbZlEkNjoAtfV/rsXJ+fa/sF04xXm7yglEJKyeDg4CU/ViaT4Y//+I/56le/2sy7Ozg4yBe/+MVFqSC/9rWvceedd2LbNpOTk1SrVXbt2sXhw4cv+RhfLRBoUskE73/fLTz33Any+RxerUapWiGZThMqm4GN2+lft5mZmRnGx0bJz03j1moEgY/SETJ0SRoCz4/qlk9EhKTmVzFNg0reBWFg2QnKtmJqYpZfjY/T+5hNT0cHw+vXrL8lSAAAIABJREFU0T/YR9eWTVwxNEhxaoYnHj9ALl8lVw5JuClmAyCdYM6tUisW6LSTjE3lGB+dpq+3D18aKALaDEXGVHS3d3P61PF6EQJBpqODts5Okuk0E5OTmIZBFGk+fPtH6ezM8qMf3oMhBdIwkAqkIUgkbTQZQBNpjef7RFHsuRcipo9SMU+1UqYtkybhWCRtG0sZWMpERxEoVZ9CX0JLF15ClrFzEERdOhBSEMnYabauZ5i+tixlz0OoCF/oes4OIGrowPPOK123bhuRus01DXUtmMaDWwi0Eugo4FOf/J+48c03IoAojJBKUikV+PL/9b/x9e98Hx35aF/i+ynu+OTvNK+nk0px++0f4i/+9/8Dy5D4UT2ITIj6LIx5kp+/Wqu0z/cBtVTRXf16vyJ1omdmZti+fTsdHR3s3bs3rpt0iXDTTTdx7733Mjs7y4YNG9i8efOKuXfn5uaoVqu4rsuf/umf8sgjj/COd7wDudZJqF9ByLr+ddVVu/it37qd/r5ugqBG1a0QAsVaQNGDwMrQ1r+eHVe+iT17b2Dnrt309w+RTqeROkSh0fFdiJAQap9A+4Q6QCtJxfeZLZc5kysz5RtMk+aUZ3J4ssx9B4/yrR8+wtf/v/v53n3/wpHT4zhdPWzeuJmdG4fZ0tfJcGeC9d0Zdm4YZNu6PvzCHMp3Ofjzf2FueoaaG3LymRPkR0fJWiadnZ3MFIt4fkAIOOk0VjJBqVJmZmaGqek5SqUaYSi46aZ38pa33ICQBqBRhsQ0FJlMimy2nd7ebnr7uhkcGqCvv5eOznaSqQRCgSYi8F0Kc7NUSnFtOUKNZZhEazRjaiSPaaiycet8PPRLSaL+XgQiEkwV5gBYt3EL//E//RlbN28im3Sw7QgjAYZjoAyJlAIlJbLxT8RRzULEy8Ln/67bvEKDlPG+AjLZAd5y0/56Tt1YYw98j//+d3/NV//xexCZKJFCGin+8e6vMTp6etGo99/0bjq70xBFSBFb2EKD0AuiKhZeo6VtFrbF/PYrbVv/N78sOV4UIoRefjkX4BWxdH/84x/zR3/0R3z961/n1ltv5eabb+bhhx/m5MmTuK7L1NTURZnWCyHo6enh2LFjmKbJBz7wAb71rW+dMzLhRz/6EXv27GFiYoKBgQGmpqbIZrNvqKQ7SkocW/KOm9/O6VMn+X/+4atMjI2STrdhGBblqo80k3GEAeALh87+DfQODONVixTmZpiZmYJcHs+t4Qc+AonUGs91MW0bQ0mCIEKG4HkRpmNTTSqqUYQRSsxAUPGgeOg02tRYIqBDCDodk7ZsmoE2E6ECpBZUTEhmbNyzL+D093DqyNPkJqd5+tBBJl94nm5TcejAgdjy1hohFXY6hZlIMDM7R6VWQyobyzQptaXZMDzA2/e/k+eOHadULmBbDkEQorUg4SQwLSNO0iIEURQRRVEc0eF5aK0pFguUCwVqrkugQ0rlMtWqG1u6xL/JS2rtvmx5YbEVFxGhUTz6wH28553vwrIs3nrDfvZdfyNPHHqcpw4+xdTcHA89+gBnxkbRhkY3koTW5YSEcvDxkZ5CWxIFhDq+HlorBBplSrzQZX3vAL3d8WISUc9Z8Yuf38/ffv3rWNogkgo/iIPYZk+M853/8V3+lz/4o+Z4Ozt7uPHNu/ne9x9ACgMd0Sy4Oq8CNGz/hqg8b6Gv3BZL9pvfZCE31zehWfXiHHhFSLdcLvNP//RP3HrrrQghuPvuu1m3bh07duzg8ssvp1QqLUvfWKlUGBsba/4dRRFHjhyhWq3i+/6KeXVt28ZxHJLJJMlkEq01o6Oj5xzXqVOneNe73sXhw4e54YYbeOyxx9i9ezc//elPL97Jv0qx6HepwbZNdu/exU+H+jn1wmmOPnmIDVt2EvohURgvoPC1IF/zcUKTZMIGR5EdytIxtInQq5Kbm2VuaopSfg63UiaMQvyaj6FUrLbVp31BFEEQYZsmQRAhdIirNZEO0ZHEl4JAQbHq4wQFpNDYSmIbNlKZpKwEthBMnXyeWq6K6WQolGYxJLxw6nlmZ+cIgxAhBFYigVCKmucRigqOnQBlI6WkWKpw6vQY6wb72Xf9DTz44E9w7CRuzadUrGDbNulMCjtpIQ1ZjzeOf4ulUkQynSKRdig4NlNT0xiWhZVIEUS6eYFfnfICLLZ6F9rIGqUcfvrTf2H4r/+OOz79CdKpNIZhsOfafey5dh9aa9777LsZHxuPowW0rlv2AiUkju0Q+iG+53P4sce458H78MIqwlYoYRIFITXXw/cDrtt7NY5tNUdVLOX5+y//PaEXIY0UUoT1ckEhmDY/f+gX/MHvBph2o/iBwdt+Yz/fvft+pBXWoxeiRefWsOdi6XWBNdu4aA3CXSTNzj/ENAsdgYsJWojFeSBWwivmSDt69CjHjx9nx44d7N+/nyiKKBQKVKtVDh8+TC6XW7R9f38/Y2Njzfpn7e3t3HLLLYRhyMjICMePH2/+oIvFIs888wy2bVMoFHj88ce5/vrrefjhh1cdk+u6aK3xPI/Ozk6OHz/Otm3bLtk1eDWiMWuSQtPVkUXJuFzP5NQYz0eCzp4hvLIBlokwTQzTpub7CFMjpYEfSaLQwxAWPf3rGRgYRgQ+lUKe/NwsM7OzlMpFKtUKoY4ItSYEgkqENkMStgnKJ9SxNGEKC4RBRRiUI4UVGpjCwBIK4YWYMkJVK1iVEDlVQDOKk2zDEx7V8gxefg6CCKkFoQ5JppNYyQTptgxdXT30dHbjhhFz+QK+W6NULjM1Pc2uK67iySefJJFIUam6TE3PoJRCTSksR2HaBolkkmQiQTKVwjBMnEQC07aQQjKXz9VL9Uhq1dold6A1v79FB6oTwYvKC8sxb/eGREJjYPAP/+/fcd8D3+Ezd/4+e/ZcR6AjDGXQ3d3D9h272L5j14uOb++Nb+ZDn/7temSBbCawcatVJsZPs3nbZYu2f/hfHuJXvz6MkXAIA5+4PogilCFCWkxPT1Eul8na80ut2zJ9EEYIQyH9erIhAURxpWJZP7OYY2W93aizFqfxbCRV11IgotgXIJjvo7l1nWNF/RiRhoAAQ77KHGkNhGHIU089xZEjR7Cs+OnW09PDxo0b6e7uXrTthg0bGBkZaZJuMplkdnaWXC7HoUOHqFarPPHEE0gp2bx5M93d3QwODtLV1dUs/fPNb35z1fFEUUQ+n49DpzIZuru7SdRrXL0SUQymaWJZFqZprtER45tQEecXyLZlSSUzKGWgpCY3O8rc3DiOk8R2kqTaOzATKZRhoX0DYdpoodDaIlepEkkH01AoZaI6knR3DtC/OcR3q1TKRYr5OcrlAsVCAbdWJYoigiiiLZvFVIooitBBiO8HcZxtpImCgECAIETJOGRJKpNAhxAEVNw8haAAOkIEAVrXK8uGAdqSmAmbVDbNrit30Z7KEFZrDKwfoFCu8szR48zMhkzn5ti6ZRPrN26kVq3FCz2SieYqozAS+GUft1aioKoIYvmpq+qRaU9iWRaWadKeTuGWyxhBQFPmu9TZxhZEaS33wp+vc2g+DKoxfQ51hBlJRs/m+Pd/9u9pb0+DFFjS4fbf+iib129C+xG+jnVshEBKGacL9UNMqbhs1xX0D65noG9lB/r6TZsWn4rWHD70SBwXLAEU6DjpkEQQGiEq8JAsnhX39A7S2Z0mV64hlB/XcEMgtSSSISKKZQstIoRWxKRbb4uoHrmh6u+B0BItw9gS1oCIEFrUCTuqM6/ESdps3rqNdcNDzJ6ZOedX9KoIGdNa47ouAKOjoytKAI888siiv5VSdHd3x1+sUlx11VVcf/31QGwVVyoVDMNgYmKCVCpFEATNY5wLtm2TTCYpFAoUi0Wefvppdu3axZ133kkURRw9epRisXiRzjrOAXHFFVcwM7P8CxJC0NfXx/DwMI7jXLRjnhvz08o4Y5Sgq7OToaEhJsbGqNaq1GoelapLPj+HPz2Nmpog2dZGMpkhnWrHdpJk2rKY9QdFtVpDJJNEUqCkwnVrCB2QTqTJJFK0d/fGZBgF+L6HW6tRrVQIPZ9ysYg0JE5S4VgWiUQSS1qUCkXcmk8UeISRT3wXRARhnCMgEhGhW0FEEUprFLq5jWE72AmHbSMj1Go1HrjvJ2xeN8z2y0fI5Yu0t7cRacHU1Dhzc3Ns27adY0ePIaXEti18L17mLIUEGWuOAoHrugghmJ3NMTM7RyJhY9sO4+NjPP7Lxxgc7MUw5iMXLrm8wEp0ulg+uKC2jrN7uUIgfIGhbHJzHoYWGMLjK//1r4kURHUnktACZCxN6CiO5/W9iJ3bL+e6N11DwkpQ8T38KCT0fbqz3bz3/e+jt3dw+fWpKrQBQiqkiPCiAC0NhGkQlCrse8/tpJckFFo3PMwX/uILvHB6AqkkprJQgnoEw4K0ODpCCBkrDFGEkHUdWutmuyHVRDpePNPQ8wUQRrGhpmRcLaQv28XIFZfT0dnOkz//9Tm/o1cF6b4UhGHIxMRE8++FRC0aT1mlePvb386HP/xhKpUKt912G3fdddc5rdatW7dSqVRIJpMcPHiQ/fv3Mz09zYEDB+js7GR4eJh0On3RzkEIwa9+9SuOHz/efG9kZIShoSFc1+X48eMcP36c66677qId80Jg2zY7duzgqaeeIpFMUyxNEoYBjm2gBIQEFHNT+NUS5fwMfhDFM4SefhKZLBqB50mUYRIpgZaKqusTeSG2ZaJQCKGIMBGmhWNlsNMhhBGRmmRudgYvCqiFEWU/xDZNbNsik07FFjTgex6B5+HXPAI/iDN+eVFc3r0RVC/jOGLp2FiJBGNjY/ziwCGOP3uMM+s38JGP/Sa+52GbFusGB5mdniSfzzPUP8CZ0TOxsyyK45HjSBvdDHsMggDLspBSxiFJ0qBUqpJOJxhat46fPfwAUzMTTTK51JruRUtivqCXxmquWMvU+HVLMdDxkmqtDYhoLruNte66kixio0oJxYljz/LMsWewDIkIIzBiaZZQ8P0ffIdsOsGua3bzqU/9Mdm2DoQQfPQjv8OjvzzE6clTSNtAaIUMJWHg0T7Qz+0f/02kWhxhJKTBdW+5mbW8a6JIU6lUiKKQWrWCG57bwHvNku5q0FoThiFhGPKTn/yEO++8k6997WvccccdfOpTn+IHP/gBY2Nji8g3m83ynve8h29+85u8+93v5uTJk+zZs4dHHnmEXbtiraohbVxMJBKJZv8Q/2Aff/xxJiYmmuP78Y9/zGc/+9mLfuzVoLVGSsnVV1/Nvff+iEKhhGFaBKUy1WoVQyl6+nqYmp2hVs5j2zZ7r93L+Pg4J48+QSITx8GmM1mEski3ZUmm05jCplytIpSBISVSWgShT+D5GHWrRChNR28/tUBTLEyBEd/8lmGgDUlkAKYCKbEcCzMMSUQadITUGu0FVMtl3GoV3/UIgxAvCGhLOGSyWY4dO0Yxnyfb3h7njXAcojAm9UQyRbatnbnZaQylaGvLUCyWcGyb0NBE4XxsrtYawzAwDCMOLZQSP9BxgH9nN5ddvo2f/PTH2E6iyYaXvBx74yf9MuWF5VIDTSKd1zAax4rz6EbQtALn/U/z8bl+aAPghyFCKHTdJyaQnDk1wajQ/PqpZyiXDP7Xf/enGIZiw84R/ts//D1//6X/m8cOHOLMmRcIdIhlau787d9lx5bFNeiq1RJPPnkYHUWYhorr/BEhGqvTBI0KrMsvC+duC1mPRW7417QGCUrEdd6On36ef/7+9ynmSvQPZZk86/PtG69f8St6XZLuQjQskaGhIaampjh+/Dgf//jHyeVyPPTQQ5w8eZKhoSE++clP8q1vfYtiscjp06d585vfzF/+5V9SrVZf6VN4RbFlyxbWb9jE1PQcpmVjWRZR4BP6HvnZWbTvI3VEe9IhNz1OaW4KM/Ko5ifxKjkKs1NI00GZNu3ZTrr7B7GUwKtVUXYSrWKPszRNfN9DSIkhFWiDnv4+osilVi1hWhIrkaCvp5Penm56evvJdnThey5Hn32aVCKB59aolMsENY900EEUhURhhA41pXIZLJPugT461w0xNj7Fxm0b2HXZZSgpMZTCSWbQQtHb001uZopCoUAmk6FareG6cXRMbLGClLJJoLI+FY1vxIiIiPaODsIQNqzfQldn3yKyvZTW7mJqbdqnL0FeaPTUsHZpOo/qikOz7/it2BnVWLK7lN8b24NAaNn8vKHGakwEYIUm9/3wh/zOJ+5g/cb1APR09fJvP/dnTIxNce/376FQmmX7thF+453vQS64joHv8uX/87/wlX/8OmlbICMLLRvjFSBj6SOKaObsiXRMqETx6ORCchV1S1w22gKURqCJIoEwdD1xemyBG74gFD6jpyQyss/5Hb3uSRfgiSeeYO/evbS1tfGZz3yGb3zjGxw4cICRkRFuvvlmpqen+cY3vsGpU6f4yEc+Qnt7O3/1V3/1hidcgEQiydv37+fwk0/R1tYWh375HgFxlIdlWRhKEgUBk2NjSCGwTQVBgFstoIWBNB0My2HOr1Eq5GjPdpLOtGMikYaJYViYpoUA/MBHmnHyalNY9A8MMjZ6mlKxFMf9uhUcy2TLyHZ27d6NaSg2bd1ErVomNzdLIZenUqpSq9XwPI9SsUipVCJwa9iOQ+/AAD39Axw58izpdJprrr2WarVG4PlMFMbRSLLtWaRSVKpV0uk0k5NTWJaFDjVhGIEAZcRLesMwDkUTUhLpCM93UUqQyWRwEkl6ewfo7uo7j5Cti4OLl8R8ZUW48dZyCheLV9rqJa8r7NzM+Svq5qcWhMoil5vj2ReONUm3gb6BHn7707+z4mhLhTH+25e/xNe+8R3SKgG+wF/oX4vN0wUuwoWPpfr40fV6mQ15ZMGuzXMV87KJG78by1hBTOhCNp1t58IbgnQfffRRPvvZz/LII4/w+c9/nn379jEyMsK9995LtVpleHiY973vfdi2zUMPPcSBAwcuiZTwWkLDEjMMxXVv2sOWTet5opjDTthU3SpEIWEUEXke2Ha8cglNGAWEocbz6ukU8cH3ueLyHdzynvfy858/xtHnTjBbKVKwU0RC0d7RRTKVwTYtokDHU3QlkdIAM0lH/zq8s6epVctMjM8wNzXLoSeepv27/0xnRzsDfd10dWQxjTgZS+D5KCExDYOE7VDI5QmjAMNx6OsfoD2bZc+eq8jn82zdtoWZXJ7R8QlOv3CKbdu2kR4aQJomXhAy1NnLTidFEIS4rovneURhhOu5+J4XJyz3fbTWBDUfXfVJpJK0p9tIpVIk29L0DfWvWYm0BlO8vCTmS+fcCzpeIDLMt5ee3NJtlx9j/n2xKC420BFSWNx7z/fYd+1eMqnlZd4XopTP8exzB/nyP3yZnz98kIxO4EpJGIoGz8aSh14ww9DzzuKGBb68retDWtDWorlfbBnPX6f4MLpxQFZ42jTxhiBd13X54he/yC233MK+ffs4c+YMYRjywQ9+kI6ODvL5PPfddx/Hjx9/w5MtLNcdOzo6+Ff/6qOcOHGcYrFAIpGIn+6eRxiGVGu15nS7sUpLa91c4LJu3Tr+w+c/z/XXX88nfmuG+x/8Gfc/+DCHn3qGfKHIZKVIJpMl255FGSZeEGGk07F3GImVSNE7MMTM+BhepYQXRPjlGlV3nMnxMU4cU5iGxLFNUskUjuVg2/H0LgxDypUyWkN3Tw99/f14nsf27duxLAsn4ZA/dZrDTx2GIOLaa/YQEmvt+VyB7u5uBgauaJ5LEASEYUgQBPi+T6VSiQtQhiHjY2P8+uBB0uk0nZ1ZTMemrS3N0GD/itf1UmBxWP5KN/+FRC+s/LlY1j4XwSwmoOYqsBW2aPYjwngxxo8eAPmfufXmd/OmfXtJpeYd2DXP5djRIzz77GHuv/+HPPbLwxAYpGQKV0IYNo63RPKAxU70c7zfbC+QUESdqBsqiW5a7PPb6qapr5fn11mANwTpQryi7a677sI0Tdra4tIjWmvy+XyLaF8EQgj27t3LHXfcwZe+9CWq1SqmaSKEaJJQw3m5EIZhsH79ej73uc9x3XXXIYSgu6+LD33o/Vxz7R6+971/5sEHHuLs2TGC0gz5oIyQJsi4KkEylUHIWJRznDh1oufWiKIQoSO0H4AS+MSlc4JQU6n5GFI1o1cgzilhJxJUqzUKhQLpdJpMJkOtViOfy5Gfm6VWqdCV7UTVq/0mHJuiFExOTTI6Oopt281/Sikcx0FKiWmapNPpuNBqVxdah7huDakkiaRDV2cH/f19a/hd1V+Bi+VIe3GB4tz7nZuq41C+RVZwg91kiFJJHrj3Xn5y7z1cu+9atm4YpC2VoliscHZummeOHGFqqoCJgRWY+BLciGbSu+XHXPlczucxuPzRoxfxbbPfRUT7Brd0F8L3/RXjYltYHaZp8v73vx/TNPnbv/1bTp+OE42kUqnmIo6GJ18p1XRefuxjH+O6665rEqAmQFoGmzas44PvuwXHkDz04EOcOn0KIpcw9AlwmR4vYTlJ7GSSbLYdQ0oKhQJutUbScVBSAgEgCMLYIxNqjZSaQNbjYf1gfoGJ4zA+Mc4vf/lLrrvuOqSUjI6OYlkWrlsj295Ge1tc/bizrYNkwkEKQW4ux/PPPx/LCnVrN47btZtx3bZt09HRge1YJFJJkpkk2Y4smXSS/t4eUsm1qRpRv8Dxi1hIf+ezDHhJB+dsr0bc59vHUmt3Qa5hTRwLG/pIYWEJwVOP/5oDv3gM05BE2kBKMAKBJSRepPGRcca3BWQ7b40uTGK+VNpYrc0F7Dev9zYeHqsYum880m3hwtGYbtm2za233srOnTt57rnnME2Trq4u2trayGQyTes3ruIaE69t203pIe4nXlCgTMWmLVv40Ec+TE9fL/fe8wOOHT1G5PuYop7boZpjJj/F7ISkr7ePTMLELfrkZkskEzZ2wsQw4/VzWsfB6lrMB7iresYrlOL6t74V13W5//772bBhAwMDA82CpMl0kn/9rz8OQUgy4VCtVJAitvBN0ySTyeD7Pp7nNcm3XC6Tz+fjhO9BEC8RNhV20mb3lVcglaRcLjE40I9trd1t9vKTmF/Iti+nj4XvLiCyuhAbCU2kY+1UhBKEg+fH2/po3PqJNlaFxbvVA9rqkQcLrf2X1j6/bWm26xp1q0ZaCy8XC7VIwzAYGRlh+/btTTJuhEw1tlt1ybSWzdWUWip6+gd47wc+wMiWEb733e/yy8cfp5DP4XkutgDlSKo1l9Mnj2IaBrYh0IEmn59B5yGRTJFItWFZNtIw0GhkXV5okPzg0DBve9tN3HXXt5idnWVqaopkIkGxVCKdTnPZ5TtJWDZetUqxkKdYqRIEPoZh4Hku5UoFpSSWbcfpK+uyQkNSKZfLMSEHHlIpDMtEqrgKQntbD8YlTF26FBcnifn5ONhe6n6co10PSKuXbl/o1GpuJ+rZcJtPlsVjaRSJePGcuedjtZ9/H8uPt/B1OVqk28IFobHab+lyzXO1l+y9QHNsxHMKkk6CK6/ezfCG9Tz22GM8/PBDPHfsGMViCdd1ySQThEGKSqWKW3OR+FiGAqEIfZ9ibibOIGZacRKark5M06ZSrVLzPPoHh9BCMD4+xZaNWxjq6SE/O4OWkv7BzbhVl1K+AEJT9Vz8MEJrA8NQuH6NueIcYaiJAo2MYmtMyNgKbpCvUgqEIJPOYCiz7iWPSCadJTflpUXT6XPR5IXz1XZfTFJoYKnlu5ik5u3F+feb9KaXjkgvo9D57ld66KyG8xn/Sn2sVG1Zr3q0Fum28KJYSqIv1Qu/NANW8+apk1Z3Tw83v+MdXF6v2HHixAkmz44xMz1NrVaLdTshKJVKzM0VGBufoLrg/SjwKOQ93CCkLZsFIUinMwz095NKprj8sh2MnT7FZTu2UywVeOros0xPT+Cn2kmnUwQ6jNfcIwCJbTmUa9V44FLQSIaioziSftGiCAAl6OzoaD6QksnkGiYrWnqN11peuBj7LSWvc+21kgSwvK/FuHSyykpHXe0h1yLdFl410FpjWRbr16+nv7+fYrHI9OQUszPTTE9NMz0zTalYJJ/P4/kh4+MTHH/uBGNjY3hBgFQGKScJhsX6DZvoH+gnnUrhOA6h75KwFVddtYvu7k5KxRzbt2yiWClTLOQ5dvQo6fY0yXQGy0jg+3F/o2fGODV6BtOyMYREB2FcOViHSCFRhor1a9Oio7uT9vYs2WyW/v5+SoU5lKEWrZq69BeRSyQvXNwp+YX3sfA8FvaxtH2+x34p4zjPPvTCcS5Hi3RbeMWxNBGMEKIZHdCRzeJ7G3Bdl2KxyNTUFKOjo+TzBYbXr2Pbtm089PCjlCs1+gYGyRXKbNy6jevffD1zs9OMnz2L1hFnzp7BdkxufsdNzOXyFEslHMdkx9bN/Pevf4f7H/wZ77/tfdxw2eWEgeCZZ46jTJPOrm56yhWuuno3hoAjh5/k2DNHqFTKgKhnR3Pp7Opi285tJBIJ0uk0bs0l8ANMw7zkSW4W4sLkhbWwZC+kj5X00HOR1yJRYZVtzmd8Lzam1ftYfHUFiJa80MJrAOciJVUPQbMdh1QmQ7azk8F163BrNXzPxfcjJqdnefjRn1MoldmybTsbNm8jDIm93/V+Hdvmbfv3U6n5jE/NYdppNm9ZT6Wcx7Ft3nrDW/nABz5A/8AgR44cww98sp1dXLvverwowLQMDCXZt/davnf33Tz99BFyczmmpkpk2tLceOMNdHV3NcPJarUqpXIZw1jbW+zCkpi/2vByxnYxHhTn2+/i9uIe6tJHK3qhhdc06hqpISXpdLpedjtCRxFRJNiyZTOPH3qCofXr6erpQygTy0niOBUgXpU2ODSIUHDi5CnCACbGzrJ1ZDNaC6699lqUadLT0wMIJiYn0VqTTKUxTRMlDcx6OstMyuF3P/0pBHDw4EG+8IUvsG54HU7CbmqWEIzBAAAH2ElEQVTLQRCQy+UJfRfTXONbTDcvGStPuV+L8sKlkAYu1fgbm7XkhRZed5AIKVECevp66OjI0tnRhdYRhtQoBUiBMiyqVY/cXJGJ6UkINbNzs8zlZihXa6BsUpk2uro7cCyTqlvjzNkzBDrCsQWOGZFKJ0mmkpiGihOY6xClFDPD6+jt7WN4w2aCSFHzNF4Is7kCqYRJwjZAynopmLXBuan15coEayEvvBqP/eJ9rCwvtCzdFl6jWEl2EEKgo3qIkdAMDg6QzbahdYTveXi1CrVqBcM0MUyLmutz9sxkPZlKSE9XFu1XKFWqKMNEKElfby8SzcTYGIV8jlRbmuF1fWzbNISq58ttDCXScX2ttvY0A4MDKGVSrfpYToQfRLiuz/C6ARQBaBBaL533X8LrVX9t/v9ykpi3cD5YWV5oWbotvM6gob7qSDA0NIRhmni+D2Gc4EgpRUQcDVEtlfB9H9OxkUKwb9+bmJmeZGJqlnS6Dd8PEFLheSHPPPsctZpLX2cPg/19sXWr5OIbS8RLjtOpJJ7rUizkMUwH03bikvMySRj4tLUlqadzXTviW+Rfernywhp6/F8V0sDFkxdWc6TJVT5roYVXL8T8D7unp5u+vj7m5mapVquUy2VyuRyGYaAB27KpVqtYtkU6nSaXy5FIJJEyrm9WKpYpFiuMjk3y/AujGIaiv6eT9vZMXBFiidUiiIt3ppMJLNNgYmKMYiFPLjeD77lMToxxZnSUtkw6Lqi5hqbmYv+/WNa+8On0udqvZB/nu9/a9KGXbisWyjnL0SLdFl6TaPz8hYhzQuzZczWFQoF8PkehUIgXU9TX4lu2RT6fJ+EkSKVSnDp1muPHn0NKVS9Y6jM1nePXTzxNrlAimUjS191Fop43YuEKvMaafiEgnUyyc/s2SsUCZ0ZPMzU+xtzsFFEYkEo5tGXSxMtaV7N7LvJ1qV+Yc9HHuXGuMK0LGftL3e+13cdK8sJq+7RIt4XXBa668kpM06RQyDM7O8vs7CxhEOIkEhiGQa1WY3x8HCljOWJkZIRNmzbFuX8jzcnnX2D0zDh+oOns6GD98NAqMqxGaI1tW3zsYx/lvbe8h40bhjEMSTE3x5W7LmPvtddgGBIhm1lY1waaBdPblY57IYS0dNvzjaO90P1e230s+35fRF5oabotvOYhhGBwoI8rL9/Bgw8/yqwXkEi30d0/gGma+F6NVMrm9OgLQMDWzRsxTZOEEZuFIZrJ8XFyxSKJZJLtO7bT1dG+2hHrhTsFw0MD/OHv/x5zuTzVmotlm/R0d2KZRt0CksTetDW4ECxVFxfqkAvfO9/2xfXyv177WBYH3coy1sLrEYsS7KCxDYN33fQbPPvM04xNFTk7PkF79zgjmzYSGRVSSZu5YpHnjp/Acz36enuxTYtyuUy5UiFXzFOulti85XI2bdqAMupkuYwt6xKDUoDAUAojoUgle5Y4sRY2107UvbRJzF8tfSw9p1dbHwtfl6MlL7Tw+oAQXHnlFfzPf/hv2Dg8hFstcfLECXKFEsJIUvElg0PrSCSSnDo9ypNPHeGZo0c5c/YsZ86cpVAoMNDXx9W7rySdThKtlp5ygUNFCBbpvM32/BaL/rrU0Ivkhfp7KxLFOXtYpf1q6ePl7ndx+1juZm1FL7TwBoDWEYZhcPXuK/iDz/wuO7Zuopif5dlnnyXQiooXgVKs27CetmyWqusyl8sxMTmJ53n0dHfz5n3XsnnDEFKKeY/Uaw1iYaNB+y/HQ/9SpuqvZB+r7Xdp+lgpUmG16AWxWsJpIVYrr9bCWkLrVUSil9DdRezrFcfiYoMQBpozE5N88667efQXBxgcXM+WLVvQ2qW9vY22dAbPdQnr1SCEgJ07Rtiz5woStlnvRsc1Li79ooaLeoDdu/fouKLC0kO83Gn9Qu2k1cfqnwNacfDQ4yt+ty3SfY2gRbrnBx3p+B+aiuty8NBhHn7kl1RqHqm2DKYZJ9BxTAspBFEYsmXrZq6/7mqSjomsZ4jSSOTaLN+9qIe4avceLeqOu8WUcCEktLD9UonsYvTx2iDlpVncNBq04lCLdF/baJHu+SGujwaaCFG3VytVj1OjZ3nu1GnmZudwXReiCMe26e/r5+qrryTbnsIQGoiAOJG5eA2S7u6r9zQVXbGMNFpYG2jQBgcP/XLFC9+KXmjhdQWBoFGwsEE3qaTNzm2b2D6yiSAM8VyPKAoxTBPLNFFKNi0Vgarnon2NElX9pEXzjyUfvKgVt5rFeq52q49F+2kWfL4cLdJt4fUF0fxvEW3q+sowU0nMpLM4aTrUQ62W7/dag4YVkpgvfIgsPLuV2i/2OSzLqvUS+rgY43ipfbz4+C+sj1YS8xZaOAcWViuW8vUZuCOW/XXxs4y9lh9KcHHGv7yPxTLOaosjXp+/vBZaWILVqhWvVSmdtYLgXPJC4/V82kv3eyn9Xeo+Lsa5vPzxi3P2uzJalm4Lbxi83sh1JegFDSGoJ0h7iUTYlC4vZL/XUx8XuF8j7mDVhTUt0m2hhdcVhFYgQjQarUU9gmNeq17JVbSwvcg9pFm036ulj/Pe72L0cSHj1/H2QsBqqu6qIWMttNBCCy1cXLQ03RZaaKGFNUSLdFtooYUW1hAt0m2hhRZaWEO0SLeFFlpoYQ3RIt0WWmihhTVEi3RbaKGFFtYQ/z/gVfJF4JqJ+AAAAABJRU5ErkJggg==\n"
          },
          "metadata": {
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "Abstract"
      ],
      "metadata": {
        "id": "FKKY8V97FHcF"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "modelha = load_model('/content/drive/MyDrive/WarpedDreams/CycleGAN/TrainedModels/C_Highboost_Abstract_Model/g_model_AtoB_001600.h5', cust, compile=False)\n",
        "modella=load_model('/content/drive/MyDrive/WarpedDreams/CycleGAN/TrainedModels/C_Laplacian_Abstract_Model/g_model_AtoB_002000.h5', cust, compile=False)\n",
        "modelua=load_model('/content/drive/MyDrive/WarpedDreams/CycleGAN/TrainedModels/C_Unsharp_Abstract_Model/g_model_AtoB_002000.h5', cust, compile=False)"
      ],
      "metadata": {
        "id": "Mk0jkinM5DBH"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "def show_plota(imagesX, imagesY1):\n",
        "\timages = vstack((imagesX, imagesY1))\n",
        "\ttitles = ['Real', 'Generated']\n",
        "\t# scale from [-1,1] to [0,1]\n",
        "\timages = (images + 1) / 2.0\n",
        "\n",
        "\t# plot images row by row\n",
        "\tfor i in range(len(images)):\n",
        "\n",
        "\t\t# define subplot\n",
        "\t\tpyplot.subplot(1, len(images), 1 + i)\n",
        "\t\n",
        "\t\t# turn off axis\n",
        "\t\tpyplot.axis('off')\n",
        "\t\n",
        "\t\t# plot raw pixel data\n",
        "\t\tpyplot.imshow(images[i])\n",
        "\t\n",
        "\t\t# title\n",
        "\t\tpyplot.title(titles[i])\n",
        "\tpyplot.show()"
      ],
      "metadata": {
        "id": "8-PZ7nBb5-mE"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "#unsharp\n",
        "test_image=cv2.imread('//content/drive/MyDrive/realistic_car/n02958343_10092-5.jpg')\n",
        "test_image=unsharp(test_image)\n",
        "test_image = img_to_array(test_image)\n",
        "test_image_input = np.array([test_image])  # Convert single image to a batch.\n",
        "test_image_input = (test_image_input - 127.5) / 127.5\n",
        "gen_image = modelua.predict(test_image_input)\n",
        "\n",
        "show_plota(test_image_input,gen_image)"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 199
        },
        "outputId": "95e84be3-9f4d-48ad-c49b-7f4c28c1f8b8",
        "id": "U_p--MLF5DBI"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 432x288 with 2 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "#lap\n",
        "test_image=cv2.imread('//content/drive/MyDrive/realistic_car/n02958343_10092-5.jpg')\n",
        "test_image=lapfil(test_image)\n",
        "test_image = img_to_array(test_image)\n",
        "test_image_input = np.array([test_image])  # Convert single image to a batch.\n",
        "test_image_input = (test_image_input - 127.5) / 127.5\n",
        "gen_image = modell.predict(test_image_input)\n",
        "show_plota(test_image_input,gen_image)"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 199
        },
        "outputId": "cc5a26fb-cd84-4fe9-ece0-85a3c9638bd1",
        "id": "Qwpb6z4J5DBI"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 432x288 with 2 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "#Highboost\n",
        "test_image=cv2.imread('/content/drive/MyDrive/realistic_car/n02958343_10092-5.jpg')\n",
        "test_image=highboost(test_image)\n",
        "test_image = img_to_array(test_image)\n",
        "test_image_input = np.array([test_image])  # Convert single image to a batch.\n",
        "test_image_input = (test_image_input - 127.5) / 127.5\n",
        "gen_image = modelh.predict(test_image_input)\n",
        "\n",
        "show_plota(test_image_input,gen_image)"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 199
        },
        "outputId": "5f718358-daf0-44a4-cf83-ff8c6ace96e9",
        "id": "Zejve_Dt5DBJ"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 432x288 with 2 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "performance"
      ],
      "metadata": {
        "id": "CMKeDjc9Mg3P"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "from skimage.metrics import structural_similarity as ssim\n",
        "import numpy as np\n",
        "import cv2"
      ],
      "metadata": {
        "id": "YG0Lef73W1SB"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "def mse(imageA, imageB):\n",
        " # the 'Mean Squared Error' between the two images is the sum of the squared difference between the two images\n",
        " mse_error = np.sum((imageA.astype(\"float\") - imageB.astype(\"float\")) ** 2)\n",
        " mse_error /= float(imageA.shape[0] * imageA.shape[1])\n",
        "\t\n",
        " # return the MSE. The lower the error, the more \"similar\" the two images are.\n",
        " return mse_error\n",
        "\n",
        "def compare(imageA, imageB):\n",
        " # Calculate the MSE and SSIM\n",
        " m = mse(imageA, imageB)\n",
        " s = ssim(imageA, imageB)\n",
        "\n",
        " # Return the SSIM. The higher the value, the more \"similar\" the two images are.\n",
        " return s"
      ],
      "metadata": {
        "id": "pi43dXg5W44B"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "def calculate_metrics(image1, image2):\n",
        "  # Convert the images to grayscale\n",
        "  gray1 = cv2.cvtColor(image1, cv2.COLOR_BGR2GRAY)\n",
        "  gray2 = cv2.cvtColor(image2, cv2.COLOR_BGR2GRAY)\n",
        "\n",
        "  # Check for same size and ratio and report accordingly\n",
        "  ho, wo, _ = image1.shape\n",
        "  hc, wc, _ = image2.shape\n",
        "  ratio_orig = ho/wo\n",
        "  ratio_comp = hc/wc\n",
        "  dim = (wc, hc)\n",
        "\n",
        "  if round(ratio_orig, 2) != round(ratio_comp, 2):\n",
        "    print(\"\\nImages not of the same dimension. Check input.\")\n",
        "    exit()\n",
        "\n",
        "  # Resize first image if the second image is smaller\n",
        "  elif ho > hc and wo > wc:\n",
        "    print(\"\\nResizing original image for analysis...\")\n",
        "    gray1 = cv2.resize(gray1, dim)\n",
        "\n",
        "  elif ho < hc and wo < wc:\n",
        "    print(\"\\nCompressed image has a larger dimension than the original. Check input.\")\n",
        "    exit()\n",
        "\n",
        "  if round(ratio_orig, 2) == round(ratio_comp, 2):\n",
        "    mse_value = mse(gray1, gray2)\n",
        "    ssim_value = compare(gray1, gray2)\n",
        "    print(\"MSE:\", mse_value)\n",
        "    print(\"SSIM:\", ssim_value)\n",
        "    return mse_value, ssim_value\n",
        "\n",
        "  return -1, -1"
      ],
      "metadata": {
        "id": "uO-xNU71XBlp"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "def calc_avg_metrics(SSIM):\n",
        "  #SSIM  are dict to SSIM and MSE values of all test imgs\n",
        "  val = SSIM.values()\n",
        "  ssim = sum(val) / len(val)\n",
        "  return ssim"
      ],
      "metadata": {
        "id": "QMbE7fRfXIYG"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "from keras.models import load_model\n",
        "from numpy.random import randint\n",
        "from numpy import vstack\n",
        "from keras.preprocessing.image import img_to_array\n",
        "from keras.preprocessing.image import load_img"
      ],
      "metadata": {
        "id": "AXTng4m6QeOB"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "cust = {'InstanceNormalization': InstanceNormalization}\n",
        "model_AtoB = load_model('/content/drive/MyDrive/CycleGAN/TrainedModels/C_Highboost_Real_Model/g_model_AtoB_002000.h5', cust, compile=False)\n",
        "model_BtoA = load_model('/content/drive/MyDrive/CycleGAN/TrainedModels/C_Highboost_Real_Model/g_model_BtoA_002000.h5', cust, compile=False)\n",
        "\n",
        "test_image=cv2.imread('/content/drive/MyDrive/CycleGAN/Dataset/Test_Train_Split/C_Highboost_Real_Dataset/testB/n02958343_10092-1.png')\n",
        "test_image = img_to_array(test_image)\n",
        "test_image_input = np.array([test_image])  # Convert single image to a batch.\n",
        "test_image_input = (test_image_input - 127.5) / 127.5\n",
        "B_generated  = model_AtoB.predict(test_image_input)\n",
        "nnorm_image = cv2.normalize(B_generated, None, alpha = 0, beta = 255, norm_type = cv2.NORM_MINMAX, dtype = cv2.CV_32F)\n",
        "ngen_image = np.array(nnorm_image.astype(np.uint8))\n",
        "ng=np.reshape(ngen_image,(256,256,3))\n",
        "\n",
        "image1 = cv2.imread('/content/drive/MyDrive/CycleGAN/Dataset/Test_Train_Split/C_Highboost_Real_Dataset/testA/n02958343_10092-1.png',1)\n",
        "\n",
        "calculate_metrics(image1)\n"
      ],
      "metadata": {
        "id": "3gB6EUxeMkQK",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "b7709795-4f40-4c38-c5fb-42863293e861"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "SSIM: 0.56836475494675105\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "cust = {'InstanceNormalization': InstanceNormalization}\n",
        "model_AtoB = load_model('/content/drive/MyDrive/CycleGAN/TrainedModels/C_Highboost_Real_Model/g_model_AtoB_002000.h5', cust, compile=False)\n",
        "model_BtoA = load_model('/content/drive/MyDrive/CycleGAN/TrainedModels/C_Highboost_Real_Model/g_model_BtoA_002000.h5', cust, compile=False)\n",
        "\n",
        "test_image=cv2.imread('/content/drive/MyDrive/CycleGAN/Dataset/Test_Train_Split/C_Highboost_Real_Dataset/testB/n02958343_10092-1.png')\n",
        "test_image = img_to_array(test_image)\n",
        "test_image_input = np.array([test_image])  # Convert single image to a batch.\n",
        "test_image_input = (test_image_input - 127.5) / 127.5\n",
        "B_generated  = model_AtoB.predict(test_image_input)\n",
        "nnorm_image = cv2.normalize(B_generated, None, alpha = 0, beta = 255, norm_type = cv2.NORM_MINMAX, dtype = cv2.CV_32F)\n",
        "ngen_image = np.array(nnorm_image.astype(np.uint8))\n",
        "ng=np.reshape(ngen_image,(256,256,3))\n",
        "\n",
        "image1 = cv2.imread('/content/drive/MyDrive/CycleGAN/Dataset/Test_Train_Split/C_Highboost_Real_Dataset/testA/n02958343_10092-1.png',1)\n",
        "\n",
        "calculate_metrics(image1)\n"
      ],
      "metadata": {
        "id": "wkzPhnwyfHsT"
      },
      "execution_count": null,
      "outputs": []
    }
  ],
  "metadata": {
    "accelerator": "GPU",
    "colab": {
      "collapsed_sections": [],
      "provenance": [],
      "include_colab_link": true
    },
    "kernelspec": {
      "display_name": "Python 3",
      "name": "python3"
    },
    "language_info": {
      "name": "python"
    },
    "gpuClass": "standard"
  },
  "nbformat": 4,
  "nbformat_minor": 0
}