lipnet ml.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "id": "494dbe47",
   "metadata": {},
   "source": [
    "\n",
    "# Install and Import Dependencies\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 99,
   "id": "06c697f9",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Requirement already satisfied: opencv-python in /home/tvrraviteja/anaconda3/lib/python3.9/site-packages (4.7.0.68)\n",
      "Requirement already satisfied: matplotlib in /home/tvrraviteja/anaconda3/lib/python3.9/site-packages (3.5.2)\n",
      "Requirement already satisfied: imageio in /home/tvrraviteja/anaconda3/lib/python3.9/site-packages (2.19.3)\n",
      "Requirement already satisfied: gdown in /home/tvrraviteja/anaconda3/lib/python3.9/site-packages (4.6.3)\n",
      "Requirement already satisfied: numpy>=1.17.3 in /home/tvrraviteja/anaconda3/lib/python3.9/site-packages (from opencv-python) (1.21.5)\n",
      "Requirement already satisfied: fonttools>=4.22.0 in /home/tvrraviteja/anaconda3/lib/python3.9/site-packages (from matplotlib) (4.25.0)\n",
      "Requirement already satisfied: pyparsing>=2.2.1 in /home/tvrraviteja/anaconda3/lib/python3.9/site-packages (from matplotlib) (3.0.9)\n",
      "Requirement already satisfied: kiwisolver>=1.0.1 in /home/tvrraviteja/anaconda3/lib/python3.9/site-packages (from matplotlib) (1.4.2)\n",
      "Requirement already satisfied: pillow>=6.2.0 in /home/tvrraviteja/anaconda3/lib/python3.9/site-packages (from matplotlib) (9.2.0)\n",
      "Requirement already satisfied: packaging>=20.0 in /home/tvrraviteja/anaconda3/lib/python3.9/site-packages (from matplotlib) (21.3)\n",
      "Requirement already satisfied: cycler>=0.10 in /home/tvrraviteja/anaconda3/lib/python3.9/site-packages (from matplotlib) (0.11.0)\n",
      "Requirement already satisfied: python-dateutil>=2.7 in /home/tvrraviteja/anaconda3/lib/python3.9/site-packages (from matplotlib) (2.8.2)\n",
      "Requirement already satisfied: tqdm in /home/tvrraviteja/anaconda3/lib/python3.9/site-packages (from gdown) (4.64.1)\n",
      "Requirement already satisfied: beautifulsoup4 in /home/tvrraviteja/anaconda3/lib/python3.9/site-packages (from gdown) (4.11.1)\n",
      "Requirement already satisfied: filelock in /home/tvrraviteja/anaconda3/lib/python3.9/site-packages (from gdown) (3.6.0)\n",
      "Requirement already satisfied: requests[socks] in /home/tvrraviteja/anaconda3/lib/python3.9/site-packages (from gdown) (2.28.1)\n",
      "Requirement already satisfied: six in /home/tvrraviteja/anaconda3/lib/python3.9/site-packages (from gdown) (1.16.0)\n",
      "Requirement already satisfied: soupsieve>1.2 in /home/tvrraviteja/anaconda3/lib/python3.9/site-packages (from beautifulsoup4->gdown) (2.3.1)\n",
      "Requirement already satisfied: charset-normalizer<3,>=2 in /home/tvrraviteja/anaconda3/lib/python3.9/site-packages (from requests[socks]->gdown) (2.0.4)\n",
      "Requirement already satisfied: urllib3<1.27,>=1.21.1 in /home/tvrraviteja/anaconda3/lib/python3.9/site-packages (from requests[socks]->gdown) (1.26.11)\n",
      "Requirement already satisfied: idna<4,>=2.5 in /home/tvrraviteja/anaconda3/lib/python3.9/site-packages (from requests[socks]->gdown) (3.3)\n",
      "Requirement already satisfied: certifi>=2017.4.17 in /home/tvrraviteja/anaconda3/lib/python3.9/site-packages (from requests[socks]->gdown) (2022.9.14)\n",
      "Requirement already satisfied: PySocks!=1.5.7,>=1.5.6 in /home/tvrraviteja/anaconda3/lib/python3.9/site-packages (from requests[socks]->gdown) (1.7.1)\n"
     ]
    }
   ],
   "source": [
    "!pip install opencv-python matplotlib imageio gdown "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 100,
   "id": "0ee911f0",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Requirement already satisfied: tensorflow==2.10.1 in /home/tvrraviteja/anaconda3/lib/python3.9/site-packages (2.10.1)\n",
      "Requirement already satisfied: gast<=0.4.0,>=0.2.1 in /home/tvrraviteja/anaconda3/lib/python3.9/site-packages (from tensorflow==2.10.1) (0.4.0)\n",
      "Requirement already satisfied: numpy>=1.20 in /home/tvrraviteja/anaconda3/lib/python3.9/site-packages (from tensorflow==2.10.1) (1.21.5)\n",
      "Requirement already satisfied: libclang>=13.0.0 in /home/tvrraviteja/anaconda3/lib/python3.9/site-packages (from tensorflow==2.10.1) (15.0.6.1)\n",
      "Requirement already satisfied: grpcio<2.0,>=1.24.3 in /home/tvrraviteja/anaconda3/lib/python3.9/site-packages (from tensorflow==2.10.1) (1.51.1)\n",
      "Requirement already satisfied: astunparse>=1.6.0 in /home/tvrraviteja/anaconda3/lib/python3.9/site-packages (from tensorflow==2.10.1) (1.6.3)\n",
      "Requirement already satisfied: setuptools in /home/tvrraviteja/anaconda3/lib/python3.9/site-packages (from tensorflow==2.10.1) (63.4.1)\n",
      "Requirement already satisfied: tensorflow-estimator<2.11,>=2.10.0 in /home/tvrraviteja/anaconda3/lib/python3.9/site-packages (from tensorflow==2.10.1) (2.10.0)\n",
      "Requirement already satisfied: h5py>=2.9.0 in /home/tvrraviteja/anaconda3/lib/python3.9/site-packages (from tensorflow==2.10.1) (3.7.0)\n",
      "Requirement already satisfied: keras-preprocessing>=1.1.1 in /home/tvrraviteja/anaconda3/lib/python3.9/site-packages (from tensorflow==2.10.1) (1.1.2)\n",
      "Requirement already satisfied: flatbuffers>=2.0 in /home/tvrraviteja/anaconda3/lib/python3.9/site-packages (from tensorflow==2.10.1) (23.1.21)\n",
      "Requirement already satisfied: tensorboard<2.11,>=2.10 in /home/tvrraviteja/anaconda3/lib/python3.9/site-packages (from tensorflow==2.10.1) (2.10.1)\n",
      "Requirement already satisfied: tensorflow-io-gcs-filesystem>=0.23.1 in /home/tvrraviteja/anaconda3/lib/python3.9/site-packages (from tensorflow==2.10.1) (0.30.0)\n",
      "Requirement already satisfied: google-pasta>=0.1.1 in /home/tvrraviteja/anaconda3/lib/python3.9/site-packages (from tensorflow==2.10.1) (0.2.0)\n",
      "Requirement already satisfied: termcolor>=1.1.0 in /home/tvrraviteja/anaconda3/lib/python3.9/site-packages (from tensorflow==2.10.1) (2.2.0)\n",
      "Requirement already satisfied: opt-einsum>=2.3.2 in /home/tvrraviteja/anaconda3/lib/python3.9/site-packages (from tensorflow==2.10.1) (3.3.0)\n",
      "Requirement already satisfied: protobuf<3.20,>=3.9.2 in /home/tvrraviteja/anaconda3/lib/python3.9/site-packages (from tensorflow==2.10.1) (3.19.6)\n",
      "Requirement already satisfied: wrapt>=1.11.0 in /home/tvrraviteja/anaconda3/lib/python3.9/site-packages (from tensorflow==2.10.1) (1.14.1)\n",
      "Requirement already satisfied: packaging in /home/tvrraviteja/anaconda3/lib/python3.9/site-packages (from tensorflow==2.10.1) (21.3)\n",
      "Requirement already satisfied: six>=1.12.0 in /home/tvrraviteja/anaconda3/lib/python3.9/site-packages (from tensorflow==2.10.1) (1.16.0)\n",
      "Requirement already satisfied: typing-extensions>=3.6.6 in /home/tvrraviteja/anaconda3/lib/python3.9/site-packages (from tensorflow==2.10.1) (4.3.0)\n",
      "Requirement already satisfied: keras<2.11,>=2.10.0 in /home/tvrraviteja/anaconda3/lib/python3.9/site-packages (from tensorflow==2.10.1) (2.10.0)\n",
      "Requirement already satisfied: absl-py>=1.0.0 in /home/tvrraviteja/anaconda3/lib/python3.9/site-packages (from tensorflow==2.10.1) (1.4.0)\n",
      "Requirement already satisfied: wheel<1.0,>=0.23.0 in /home/tvrraviteja/anaconda3/lib/python3.9/site-packages (from astunparse>=1.6.0->tensorflow==2.10.1) (0.37.1)\n",
      "Requirement already satisfied: google-auth<3,>=1.6.3 in /home/tvrraviteja/anaconda3/lib/python3.9/site-packages (from tensorboard<2.11,>=2.10->tensorflow==2.10.1) (2.16.0)\n",
      "Requirement already satisfied: tensorboard-data-server<0.7.0,>=0.6.0 in /home/tvrraviteja/anaconda3/lib/python3.9/site-packages (from tensorboard<2.11,>=2.10->tensorflow==2.10.1) (0.6.1)\n",
      "Requirement already satisfied: tensorboard-plugin-wit>=1.6.0 in /home/tvrraviteja/anaconda3/lib/python3.9/site-packages (from tensorboard<2.11,>=2.10->tensorflow==2.10.1) (1.8.1)\n",
      "Requirement already satisfied: requests<3,>=2.21.0 in /home/tvrraviteja/anaconda3/lib/python3.9/site-packages (from tensorboard<2.11,>=2.10->tensorflow==2.10.1) (2.28.1)\n",
      "Requirement already satisfied: markdown>=2.6.8 in /home/tvrraviteja/anaconda3/lib/python3.9/site-packages (from tensorboard<2.11,>=2.10->tensorflow==2.10.1) (3.3.4)\n",
      "Requirement already satisfied: google-auth-oauthlib<0.5,>=0.4.1 in /home/tvrraviteja/anaconda3/lib/python3.9/site-packages (from tensorboard<2.11,>=2.10->tensorflow==2.10.1) (0.4.6)\n",
      "Requirement already satisfied: werkzeug>=1.0.1 in /home/tvrraviteja/anaconda3/lib/python3.9/site-packages (from tensorboard<2.11,>=2.10->tensorflow==2.10.1) (2.0.3)\n",
      "Requirement already satisfied: pyparsing!=3.0.5,>=2.0.2 in /home/tvrraviteja/anaconda3/lib/python3.9/site-packages (from packaging->tensorflow==2.10.1) (3.0.9)\n",
      "Requirement already satisfied: pyasn1-modules>=0.2.1 in /home/tvrraviteja/anaconda3/lib/python3.9/site-packages (from google-auth<3,>=1.6.3->tensorboard<2.11,>=2.10->tensorflow==2.10.1) (0.2.8)\n",
      "Requirement already satisfied: cachetools<6.0,>=2.0.0 in /home/tvrraviteja/anaconda3/lib/python3.9/site-packages (from google-auth<3,>=1.6.3->tensorboard<2.11,>=2.10->tensorflow==2.10.1) (5.3.0)\n",
      "Requirement already satisfied: rsa<5,>=3.1.4 in /home/tvrraviteja/anaconda3/lib/python3.9/site-packages (from google-auth<3,>=1.6.3->tensorboard<2.11,>=2.10->tensorflow==2.10.1) (4.9)\n",
      "Requirement already satisfied: requests-oauthlib>=0.7.0 in /home/tvrraviteja/anaconda3/lib/python3.9/site-packages (from google-auth-oauthlib<0.5,>=0.4.1->tensorboard<2.11,>=2.10->tensorflow==2.10.1) (1.3.1)\n",
      "Requirement already satisfied: charset-normalizer<3,>=2 in /home/tvrraviteja/anaconda3/lib/python3.9/site-packages (from requests<3,>=2.21.0->tensorboard<2.11,>=2.10->tensorflow==2.10.1) (2.0.4)\n",
      "Requirement already satisfied: urllib3<1.27,>=1.21.1 in /home/tvrraviteja/anaconda3/lib/python3.9/site-packages (from requests<3,>=2.21.0->tensorboard<2.11,>=2.10->tensorflow==2.10.1) (1.26.11)\n",
      "Requirement already satisfied: certifi>=2017.4.17 in /home/tvrraviteja/anaconda3/lib/python3.9/site-packages (from requests<3,>=2.21.0->tensorboard<2.11,>=2.10->tensorflow==2.10.1) (2022.9.14)\n",
      "Requirement already satisfied: idna<4,>=2.5 in /home/tvrraviteja/anaconda3/lib/python3.9/site-packages (from requests<3,>=2.21.0->tensorboard<2.11,>=2.10->tensorflow==2.10.1) (3.3)\n",
      "Requirement already satisfied: pyasn1<0.5.0,>=0.4.6 in /home/tvrraviteja/anaconda3/lib/python3.9/site-packages (from pyasn1-modules>=0.2.1->google-auth<3,>=1.6.3->tensorboard<2.11,>=2.10->tensorflow==2.10.1) (0.4.8)\n",
      "Requirement already satisfied: oauthlib>=3.0.0 in /home/tvrraviteja/anaconda3/lib/python3.9/site-packages (from requests-oauthlib>=0.7.0->google-auth-oauthlib<0.5,>=0.4.1->tensorboard<2.11,>=2.10->tensorflow==2.10.1) (3.2.2)\n"
     ]
    }
   ],
   "source": [
    "!pip install tensorflow==2.10.1"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 101,
   "id": "8f144b66",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Requirement already satisfied: opencv-python in /home/tvrraviteja/anaconda3/lib/python3.9/site-packages (4.7.0.68)\r\n",
      "Requirement already satisfied: numpy>=1.19.3 in /home/tvrraviteja/anaconda3/lib/python3.9/site-packages (from opencv-python) (1.21.5)\r\n"
     ]
    }
   ],
   "source": [
    "!pip install opencv-python"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 102,
   "id": "5cf7dfde",
   "metadata": {},
   "outputs": [],
   "source": [
    "import os\n",
    "import cv2\n",
    "import tensorflow as tf\n",
    "import numpy as np\n",
    "from typing import List\n",
    "from matplotlib import pyplot as plt\n",
    "import imageio # to convert numpy array to gif"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "8107a817",
   "metadata": {},
   "source": [
    "# Building Data Loading Functions"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 103,
   "id": "7904234c",
   "metadata": {},
   "outputs": [],
   "source": [
    "import gdown #To get data straight away from gogole drive"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 104,
   "id": "d36a9144",
   "metadata": {},
   "outputs": [],
   "source": [
    "def load_video(path:str) -> List[float]: \n",
    "\n",
    "    cap = cv2.VideoCapture(path)\n",
    "    frames = []\n",
    "    for _ in range(int(cap.get(cv2.CAP_PROP_FRAME_COUNT))): \n",
    "        ret, frame = cap.read()\n",
    "        frame = tf.image.rgb_to_grayscale(frame) #for less pre-processing data\n",
    "        frames.append(frame[190:236,80:220,:])#static slicing function to isolate the mouth region in the videos\n",
    "    cap.release()\n",
    "    \n",
    "    mean = tf.math.reduce_mean(frames)\n",
    "    std = tf.math.reduce_std(tf.cast(frames, tf.float32))\n",
    "    return tf.cast((frames - mean), tf.float32) / std #standerdize or scale our particular image feature by using mean and std"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 105,
   "id": "ea37d738",
   "metadata": {},
   "outputs": [],
   "source": [
    "vocab = [x for x in \"abcdefghijklmnopqrstuvwxyz'?!123456789 \"] #Every character we might expect in the video"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 106,
   "id": "06ab4853",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "['a',\n",
       " 'b',\n",
       " 'c',\n",
       " 'd',\n",
       " 'e',\n",
       " 'f',\n",
       " 'g',\n",
       " 'h',\n",
       " 'i',\n",
       " 'j',\n",
       " 'k',\n",
       " 'l',\n",
       " 'm',\n",
       " 'n',\n",
       " 'o',\n",
       " 'p',\n",
       " 'q',\n",
       " 'r',\n",
       " 's',\n",
       " 't',\n",
       " 'u',\n",
       " 'v',\n",
       " 'w',\n",
       " 'x',\n",
       " 'y',\n",
       " 'z',\n",
       " \"'\",\n",
       " '?',\n",
       " '!',\n",
       " '1',\n",
       " '2',\n",
       " '3',\n",
       " '4',\n",
       " '5',\n",
       " '6',\n",
       " '7',\n",
       " '8',\n",
       " '9',\n",
       " ' ']"
      ]
     },
     "execution_count": 106,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "vocab"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 107,
   "id": "7f527314",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "The vocabulary is: ['', 'a', 'b', 'c', 'd', 'e', 'f', 'g', 'h', 'i', 'j', 'k', 'l', 'm', 'n', 'o', 'p', 'q', 'r', 's', 't', 'u', 'v', 'w', 'x', 'y', 'z', \"'\", '?', '!', '1', '2', '3', '4', '5', '6', '7', '8', '9', ' '] (size =40)\n"
     ]
    }
   ],
   "source": [
    "#To convert char to num and vice versa\n",
    "char_to_num = tf.keras.layers.StringLookup(vocabulary=vocab, oov_token=\"\")\n",
    "num_to_char = tf.keras.layers.StringLookup(\n",
    "    vocabulary=char_to_num.get_vocabulary(), oov_token=\"\", invert=True\n",
    ")\n",
    "\n",
    "print(\n",
    "    f\"The vocabulary is: {char_to_num.get_vocabulary()} \"\n",
    "    f\"(size ={char_to_num.vocabulary_size()})\"\n",
    ")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 108,
   "id": "b8d3c97b",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<tf.Tensor: shape=(4,), dtype=int64, numpy=array([18,  1, 22,  9])>"
      ]
     },
     "execution_count": 108,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "char_to_num(['r','a','v','i']) #converts char to integers"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 109,
   "id": "35b4da81",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<tf.Tensor: shape=(4,), dtype=string, numpy=array([b'r', b'a', b'v', b'i'], dtype=object)>"
      ]
     },
     "execution_count": 109,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "num_to_char([18,  1,  22, 9]) #converts int to resp char"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "6a303a94",
   "metadata": {},
   "source": [
    "**We use these functions to convert and reconvert our text to encodings**"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 110,
   "id": "8dc25dfe",
   "metadata": {},
   "outputs": [],
   "source": [
    "def load_alignments(path:str) -> List[str]: \n",
    "    with open(path, 'r') as f: \n",
    "        lines = f.readlines() \n",
    "    tokens = []\n",
    "    for line in lines:\n",
    "        line = line.split()\n",
    "        if line[2] != 'sil': #To ignore silence\n",
    "            tokens = [*tokens,' ',line[2]]\n",
    "    return char_to_num(tf.reshape(tf.strings.unicode_split(tokens, input_encoding='UTF-8'), (-1)))[1:]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 111,
   "id": "bf993d20",
   "metadata": {},
   "outputs": [],
   "source": [
    "def load_data(path:str):\n",
    "    path = bytes.decode(path.numpy())\n",
    "    file_name = path.split('/')[-1].split('.')[0]\n",
    "    #file_name = path.split('\\\\')[-1].split('.')[0]\n",
    "    video_path = os.path.join('data','s1',f'{file_name}.mpg')\n",
    "    alignment_path = os.path.join('data','alignments','s1',f'{file_name}.align')\n",
    "    frames = load_video(video_path)\n",
    "    alignments = load_alignments(alignment_path)\n",
    "    \n",
    "    return frames,alignments"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 112,
   "id": "faae78cd",
   "metadata": {},
   "outputs": [],
   "source": [
    "test_path = '.data/s1/bbal6n.mpg'"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 113,
   "id": "da1e1bf1",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "'bbal6n'"
      ]
     },
     "execution_count": 113,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "#The function load_data does the exact split\n",
    "tf.convert_to_tensor(test_path).numpy().decode('utf-8').split('/')[-1].split('.')[0]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 114,
   "id": "6a137092",
   "metadata": {},
   "outputs": [],
   "source": [
    "frames, alignments = load_data(tf.convert_to_tensor(test_path))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 115,
   "id": "b74b5664",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.image.AxesImage at 0x7ff52c0bc6d0>"
      ]
     },
     "execution_count": 115,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "#lets look at a frame\n",
    "plt.imshow(frames[0])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 116,
   "id": "9ab0d9e6",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.image.AxesImage at 0x7ff52c096e50>"
      ]
     },
     "execution_count": 116,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.imshow(frames[37])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 117,
   "id": "a5b168a1",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<tf.Tensor: shape=(21,), dtype=int64, numpy=\n",
       "array([ 2,  9, 14, 39,  2, 12, 21,  5, 39,  1, 20, 39, 12, 39, 19,  9, 24,\n",
       "       39, 14, 15, 23])>"
      ]
     },
     "execution_count": 117,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "alignments #word rep of what is being said"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 118,
   "id": "dd3589bc",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "['b', 'i', 'n', ' ', 'b', 'l', 'u', 'e', ' ', 'a', 't', ' ', 'l', ' ', 's', 'i', 'x', ' ', 'n', 'o', 'w']\n"
     ]
    }
   ],
   "source": [
    "print([bytes.decode(x) for x in num_to_char(alignments.numpy()).numpy()])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 119,
   "id": "087e0fbb",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<tf.Tensor: shape=(), dtype=string, numpy=b'bin blue at l six now'>"
      ]
     },
     "execution_count": 119,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "tf.strings.reduce_join([bytes.decode(x) for x in num_to_char(alignments.numpy()).numpy()])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 120,
   "id": "2941b8dd",
   "metadata": {},
   "outputs": [],
   "source": [
    "#wrapping inside a mappable function\n",
    "def mappable_function(path:str) ->List[str]:\n",
    "    result = tf.py_function(load_data, [path], (tf.float32, tf.int64))\n",
    "    return result"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "014bbd21",
   "metadata": {},
   "source": [
    "# Create Data Pipeline"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 121,
   "id": "ce49177c",
   "metadata": {},
   "outputs": [],
   "source": [
    "data = tf.data.Dataset.list_files('./data/s1/*.mpg')\n",
    "data = data.shuffle(500, reshuffle_each_iteration=False)\n",
    "data = data.map(mappable_function)\n",
    "data = data.padded_batch(2, padded_shapes=([75,None,None,None],[40]))\n",
    "data = data.prefetch(tf.data.AUTOTUNE)\n",
    "# Added for split \n",
    "train = data.take(450)\n",
    "test = data.skip(450)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 122,
   "id": "2ec91d18",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "50"
      ]
     },
     "execution_count": 122,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "len(test)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 123,
   "id": "c78dd1bf",
   "metadata": {},
   "outputs": [],
   "source": [
    "frames, alignments = data.as_numpy_iterator().next()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 124,
   "id": "73d19cc6",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([[[[[ 1.4830974 ],\n",
       "          [ 1.4469243 ],\n",
       "          [ 1.5192705 ],\n",
       "          ...,\n",
       "          [ 0.32555798],\n",
       "          [ 0.36173108],\n",
       "          [ 0.36173108]],\n",
       "\n",
       "         [[ 1.5916167 ],\n",
       "          [ 1.3745781 ],\n",
       "          [ 1.5192705 ],\n",
       "          ...,\n",
       "          [ 0.32555798],\n",
       "          [ 0.36173108],\n",
       "          [ 0.36173108]],\n",
       "\n",
       "         [[ 1.4469243 ],\n",
       "          [ 1.4469243 ],\n",
       "          [ 1.4469243 ],\n",
       "          ...,\n",
       "          [ 0.28938487],\n",
       "          [ 0.28938487],\n",
       "          [ 0.28938487]],\n",
       "\n",
       "         ...,\n",
       "\n",
       "         [[ 1.0490202 ],\n",
       "          [ 1.0490202 ],\n",
       "          [ 1.0490202 ],\n",
       "          ...,\n",
       "          [ 0.03617311],\n",
       "          [ 0.03617311],\n",
       "          [ 0.        ]],\n",
       "\n",
       "         [[ 1.0128471 ],\n",
       "          [ 1.0128471 ],\n",
       "          [ 1.0128471 ],\n",
       "          ...,\n",
       "          [ 0.        ],\n",
       "          [ 9.224143  ],\n",
       "          [ 9.224143  ]],\n",
       "\n",
       "         [[ 1.0128471 ],\n",
       "          [ 1.0128471 ],\n",
       "          [ 1.0128471 ],\n",
       "          ...,\n",
       "          [ 9.224143  ],\n",
       "          [ 9.224143  ],\n",
       "          [ 9.151796  ]]],\n",
       "\n",
       "\n",
       "        [[[ 1.5192705 ],\n",
       "          [ 1.5192705 ],\n",
       "          [ 1.5192705 ],\n",
       "          ...,\n",
       "          [ 0.28938487],\n",
       "          [ 0.28938487],\n",
       "          [ 0.28938487]],\n",
       "\n",
       "         [[ 1.5916167 ],\n",
       "          [ 1.5192705 ],\n",
       "          [ 1.4469243 ],\n",
       "          ...,\n",
       "          [ 0.28938487],\n",
       "          [ 0.28938487],\n",
       "          [ 0.28938487]],\n",
       "\n",
       "         [[ 1.4469243 ],\n",
       "          [ 1.4469243 ],\n",
       "          [ 1.4469243 ],\n",
       "          ...,\n",
       "          [ 0.25321177],\n",
       "          [ 0.25321177],\n",
       "          [ 0.25321177]],\n",
       "\n",
       "         ...,\n",
       "\n",
       "         [[ 1.0490202 ],\n",
       "          [ 1.0490202 ],\n",
       "          [ 1.0490202 ],\n",
       "          ...,\n",
       "          [ 0.07234622],\n",
       "          [ 0.07234622],\n",
       "          [ 0.03617311]],\n",
       "\n",
       "         [[ 1.0128471 ],\n",
       "          [ 1.0128471 ],\n",
       "          [ 1.0128471 ],\n",
       "          ...,\n",
       "          [ 0.10851932],\n",
       "          [ 0.07234622],\n",
       "          [ 0.07234622]],\n",
       "\n",
       "         [[ 1.0128471 ],\n",
       "          [ 1.0128471 ],\n",
       "          [ 1.0128471 ],\n",
       "          ...,\n",
       "          [ 0.07234622],\n",
       "          [ 0.07234622],\n",
       "          [ 0.03617311]]],\n",
       "\n",
       "\n",
       "        [[[ 1.5192705 ],\n",
       "          [ 1.5554436 ],\n",
       "          [ 1.4469243 ],\n",
       "          ...,\n",
       "          [ 0.28938487],\n",
       "          [ 0.28938487],\n",
       "          [ 0.28938487]],\n",
       "\n",
       "         [[ 1.5554436 ],\n",
       "          [ 1.5192705 ],\n",
       "          [ 1.3745781 ],\n",
       "          ...,\n",
       "          [ 0.28938487],\n",
       "          [ 0.28938487],\n",
       "          [ 0.28938487]],\n",
       "\n",
       "         [[ 1.4830974 ],\n",
       "          [ 1.4830974 ],\n",
       "          [ 1.4469243 ],\n",
       "          ...,\n",
       "          [ 0.25321177],\n",
       "          [ 0.25321177],\n",
       "          [ 0.25321177]],\n",
       "\n",
       "         ...,\n",
       "\n",
       "         [[ 1.0128471 ],\n",
       "          [ 1.0128471 ],\n",
       "          [ 1.0128471 ],\n",
       "          ...,\n",
       "          [ 0.03617311],\n",
       "          [ 0.03617311],\n",
       "          [ 0.        ]],\n",
       "\n",
       "         [[ 0.9766739 ],\n",
       "          [ 0.9766739 ],\n",
       "          [ 0.9766739 ],\n",
       "          ...,\n",
       "          [ 0.07234622],\n",
       "          [ 0.07234622],\n",
       "          [ 0.07234622]],\n",
       "\n",
       "         [[ 0.9766739 ],\n",
       "          [ 0.9766739 ],\n",
       "          [ 0.9766739 ],\n",
       "          ...,\n",
       "          [ 0.        ],\n",
       "          [ 0.03617311],\n",
       "          [ 9.224143  ]]],\n",
       "\n",
       "\n",
       "        ...,\n",
       "\n",
       "\n",
       "        [[[ 1.3745781 ],\n",
       "          [ 1.3745781 ],\n",
       "          [ 1.338405  ],\n",
       "          ...,\n",
       "          [ 0.32555798],\n",
       "          [ 0.32555798],\n",
       "          [ 0.32555798]],\n",
       "\n",
       "         [[ 1.338405  ],\n",
       "          [ 1.4107512 ],\n",
       "          [ 1.2660588 ],\n",
       "          ...,\n",
       "          [ 0.32555798],\n",
       "          [ 0.32555798],\n",
       "          [ 0.32555798]],\n",
       "\n",
       "         [[ 1.4107512 ],\n",
       "          [ 1.3745781 ],\n",
       "          [ 1.338405  ],\n",
       "          ...,\n",
       "          [ 0.32555798],\n",
       "          [ 0.36173108],\n",
       "          [ 0.36173108]],\n",
       "\n",
       "         ...,\n",
       "\n",
       "         [[ 1.0851933 ],\n",
       "          [ 1.0490202 ],\n",
       "          [ 1.0128471 ],\n",
       "          ...,\n",
       "          [ 9.224143  ],\n",
       "          [ 9.224143  ],\n",
       "          [ 9.224143  ]],\n",
       "\n",
       "         [[ 1.0490202 ],\n",
       "          [ 1.0490202 ],\n",
       "          [ 1.0128471 ],\n",
       "          ...,\n",
       "          [ 0.        ],\n",
       "          [ 0.        ],\n",
       "          [ 0.        ]],\n",
       "\n",
       "         [[ 1.0490202 ],\n",
       "          [ 1.0490202 ],\n",
       "          [ 1.0128471 ],\n",
       "          ...,\n",
       "          [ 0.        ],\n",
       "          [ 0.        ],\n",
       "          [ 9.187969  ]]],\n",
       "\n",
       "\n",
       "        [[[ 1.3745781 ],\n",
       "          [ 1.4830974 ],\n",
       "          [ 1.4830974 ],\n",
       "          ...,\n",
       "          [ 0.28938487],\n",
       "          [ 0.32555798],\n",
       "          [ 0.32555798]],\n",
       "\n",
       "         [[ 1.4107512 ],\n",
       "          [ 1.4830974 ],\n",
       "          [ 1.3745781 ],\n",
       "          ...,\n",
       "          [ 0.28938487],\n",
       "          [ 0.32555798],\n",
       "          [ 0.32555798]],\n",
       "\n",
       "         [[ 1.4830974 ],\n",
       "          [ 1.4107512 ],\n",
       "          [ 1.4107512 ],\n",
       "          ...,\n",
       "          [ 0.32555798],\n",
       "          [ 0.32555798],\n",
       "          [ 0.32555798]],\n",
       "\n",
       "         ...,\n",
       "\n",
       "         [[ 1.0851933 ],\n",
       "          [ 1.0490202 ],\n",
       "          [ 1.0128471 ],\n",
       "          ...,\n",
       "          [ 0.03617311],\n",
       "          [ 0.        ],\n",
       "          [ 0.        ]],\n",
       "\n",
       "         [[ 1.0490202 ],\n",
       "          [ 1.0490202 ],\n",
       "          [ 1.0128471 ],\n",
       "          ...,\n",
       "          [ 0.        ],\n",
       "          [ 0.        ],\n",
       "          [ 0.        ]],\n",
       "\n",
       "         [[ 1.0490202 ],\n",
       "          [ 1.0490202 ],\n",
       "          [ 1.0128471 ],\n",
       "          ...,\n",
       "          [ 0.        ],\n",
       "          [ 0.        ],\n",
       "          [ 0.        ]]],\n",
       "\n",
       "\n",
       "        [[[ 1.4830974 ],\n",
       "          [ 1.5192705 ],\n",
       "          [ 1.5554436 ],\n",
       "          ...,\n",
       "          [ 0.28938487],\n",
       "          [ 0.32555798],\n",
       "          [ 0.32555798]],\n",
       "\n",
       "         [[ 1.5554436 ],\n",
       "          [ 1.5192705 ],\n",
       "          [ 1.4830974 ],\n",
       "          ...,\n",
       "          [ 0.28938487],\n",
       "          [ 0.32555798],\n",
       "          [ 0.32555798]],\n",
       "\n",
       "         [[ 1.5192705 ],\n",
       "          [ 1.4830974 ],\n",
       "          [ 1.3745781 ],\n",
       "          ...,\n",
       "          [ 0.28938487],\n",
       "          [ 0.25321177],\n",
       "          [ 0.25321177]],\n",
       "\n",
       "         ...,\n",
       "\n",
       "         [[ 1.0851933 ],\n",
       "          [ 1.0490202 ],\n",
       "          [ 1.0128471 ],\n",
       "          ...,\n",
       "          [ 0.03617311],\n",
       "          [ 0.        ],\n",
       "          [ 0.        ]],\n",
       "\n",
       "         [[ 1.0490202 ],\n",
       "          [ 1.0490202 ],\n",
       "          [ 1.0128471 ],\n",
       "          ...,\n",
       "          [ 0.        ],\n",
       "          [ 0.        ],\n",
       "          [ 0.        ]],\n",
       "\n",
       "         [[ 1.0490202 ],\n",
       "          [ 1.0490202 ],\n",
       "          [ 1.0128471 ],\n",
       "          ...,\n",
       "          [ 0.        ],\n",
       "          [ 0.        ],\n",
       "          [ 0.        ]]]],\n",
       "\n",
       "\n",
       "\n",
       "       [[[[ 1.4206222 ],\n",
       "          [ 1.4206222 ],\n",
       "          [ 1.4612114 ],\n",
       "          ...,\n",
       "          [ 0.04058921],\n",
       "          [ 0.        ],\n",
       "          [ 0.        ]],\n",
       "\n",
       "         [[ 1.4206222 ],\n",
       "          [ 1.4206222 ],\n",
       "          [ 1.4612114 ],\n",
       "          ...,\n",
       "          [ 0.04058921],\n",
       "          [ 0.        ],\n",
       "          [ 0.        ]],\n",
       "\n",
       "         [[ 1.3394439 ],\n",
       "          [ 1.3394439 ],\n",
       "          [ 1.3394439 ],\n",
       "          ...,\n",
       "          [ 0.        ],\n",
       "          [ 0.        ],\n",
       "          [ 0.        ]],\n",
       "\n",
       "         ...,\n",
       "\n",
       "         [[ 0.8929626 ],\n",
       "          [ 0.8929626 ],\n",
       "          [ 0.8929626 ],\n",
       "          ...,\n",
       "          [10.106712  ],\n",
       "          [10.106712  ],\n",
       "          [10.106712  ]],\n",
       "\n",
       "         [[ 0.8929626 ],\n",
       "          [ 0.8929626 ],\n",
       "          [ 0.8929626 ],\n",
       "          ...,\n",
       "          [10.066124  ],\n",
       "          [10.066124  ],\n",
       "          [10.066124  ]],\n",
       "\n",
       "         [[ 0.8929626 ],\n",
       "          [ 0.8929626 ],\n",
       "          [ 0.8929626 ],\n",
       "          ...,\n",
       "          [10.066124  ],\n",
       "          [10.066124  ],\n",
       "          [10.066124  ]]],\n",
       "\n",
       "\n",
       "        [[[ 1.380033  ],\n",
       "          [ 1.380033  ],\n",
       "          [ 1.4206222 ],\n",
       "          ...,\n",
       "          [ 0.        ],\n",
       "          [ 0.        ],\n",
       "          [ 0.        ]],\n",
       "\n",
       "         [[ 1.380033  ],\n",
       "          [ 1.380033  ],\n",
       "          [ 1.4206222 ],\n",
       "          ...,\n",
       "          [ 0.        ],\n",
       "          [ 0.        ],\n",
       "          [ 0.        ]],\n",
       "\n",
       "         [[ 1.3394439 ],\n",
       "          [ 1.3394439 ],\n",
       "          [ 1.2988547 ],\n",
       "          ...,\n",
       "          [ 0.        ],\n",
       "          [ 0.        ],\n",
       "          [ 0.        ]],\n",
       "\n",
       "         ...,\n",
       "\n",
       "         [[ 0.8929626 ],\n",
       "          [ 0.8929626 ],\n",
       "          [ 0.8929626 ],\n",
       "          ...,\n",
       "          [10.066124  ],\n",
       "          [10.066124  ],\n",
       "          [10.066124  ]],\n",
       "\n",
       "         [[ 0.8929626 ],\n",
       "          [ 0.8929626 ],\n",
       "          [ 0.8929626 ],\n",
       "          ...,\n",
       "          [10.066124  ],\n",
       "          [10.066124  ],\n",
       "          [10.066124  ]],\n",
       "\n",
       "         [[ 0.8929626 ],\n",
       "          [ 0.8929626 ],\n",
       "          [ 0.8929626 ],\n",
       "          ...,\n",
       "          [10.066124  ],\n",
       "          [10.066124  ],\n",
       "          [10.066124  ]]],\n",
       "\n",
       "\n",
       "        [[[ 1.380033  ],\n",
       "          [ 1.380033  ],\n",
       "          [ 1.380033  ],\n",
       "          ...,\n",
       "          [ 0.        ],\n",
       "          [ 0.        ],\n",
       "          [ 0.        ]],\n",
       "\n",
       "         [[ 1.4206222 ],\n",
       "          [ 1.4206222 ],\n",
       "          [ 1.4206222 ],\n",
       "          ...,\n",
       "          [ 0.        ],\n",
       "          [ 0.        ],\n",
       "          [ 0.        ]],\n",
       "\n",
       "         [[ 1.380033  ],\n",
       "          [ 1.380033  ],\n",
       "          [ 1.3394439 ],\n",
       "          ...,\n",
       "          [10.350248  ],\n",
       "          [10.350248  ],\n",
       "          [10.350248  ]],\n",
       "\n",
       "         ...,\n",
       "\n",
       "         [[ 0.9335518 ],\n",
       "          [ 0.8929626 ],\n",
       "          [ 0.8929626 ],\n",
       "          ...,\n",
       "          [10.066124  ],\n",
       "          [10.066124  ],\n",
       "          [10.066124  ]],\n",
       "\n",
       "         [[ 0.9335518 ],\n",
       "          [ 0.8929626 ],\n",
       "          [ 0.8929626 ],\n",
       "          ...,\n",
       "          [10.066124  ],\n",
       "          [10.066124  ],\n",
       "          [10.066124  ]],\n",
       "\n",
       "         [[ 0.9335518 ],\n",
       "          [ 0.8929626 ],\n",
       "          [ 0.8929626 ],\n",
       "          ...,\n",
       "          [10.066124  ],\n",
       "          [10.066124  ],\n",
       "          [10.066124  ]]],\n",
       "\n",
       "\n",
       "        ...,\n",
       "\n",
       "\n",
       "        [[[ 1.4206222 ],\n",
       "          [ 1.4206222 ],\n",
       "          [ 1.4612114 ],\n",
       "          ...,\n",
       "          [10.350248  ],\n",
       "          [ 0.08117842],\n",
       "          [ 0.08117842]],\n",
       "\n",
       "         [[ 1.4206222 ],\n",
       "          [ 1.4206222 ],\n",
       "          [ 1.4612114 ],\n",
       "          ...,\n",
       "          [10.350248  ],\n",
       "          [ 0.08117842],\n",
       "          [ 0.08117842]],\n",
       "\n",
       "         [[ 1.380033  ],\n",
       "          [ 1.380033  ],\n",
       "          [ 1.380033  ],\n",
       "          ...,\n",
       "          [ 0.        ],\n",
       "          [ 0.        ],\n",
       "          [ 0.        ]],\n",
       "\n",
       "         ...,\n",
       "\n",
       "         [[ 0.974141  ],\n",
       "          [ 0.974141  ],\n",
       "          [ 0.974141  ],\n",
       "          ...,\n",
       "          [10.106712  ],\n",
       "          [10.106712  ],\n",
       "          [10.106712  ]],\n",
       "\n",
       "         [[ 0.974141  ],\n",
       "          [ 0.974141  ],\n",
       "          [ 0.974141  ],\n",
       "          ...,\n",
       "          [10.106712  ],\n",
       "          [10.106712  ],\n",
       "          [10.106712  ]],\n",
       "\n",
       "         [[ 0.974141  ],\n",
       "          [ 0.974141  ],\n",
       "          [ 0.9335518 ],\n",
       "          ...,\n",
       "          [10.066124  ],\n",
       "          [10.066124  ],\n",
       "          [10.066124  ]]],\n",
       "\n",
       "\n",
       "        [[[ 1.4612114 ],\n",
       "          [ 1.4612114 ],\n",
       "          [ 1.4206222 ],\n",
       "          ...,\n",
       "          [10.309659  ],\n",
       "          [ 0.        ],\n",
       "          [ 0.        ]],\n",
       "\n",
       "         [[ 1.4612114 ],\n",
       "          [ 1.4612114 ],\n",
       "          [ 1.4206222 ],\n",
       "          ...,\n",
       "          [10.309659  ],\n",
       "          [ 0.        ],\n",
       "          [ 0.        ]],\n",
       "\n",
       "         [[ 1.380033  ],\n",
       "          [ 1.380033  ],\n",
       "          [ 1.380033  ],\n",
       "          ...,\n",
       "          [10.350248  ],\n",
       "          [10.350248  ],\n",
       "          [10.350248  ]],\n",
       "\n",
       "         ...,\n",
       "\n",
       "         [[ 0.974141  ],\n",
       "          [ 0.974141  ],\n",
       "          [ 0.974141  ],\n",
       "          ...,\n",
       "          [10.106712  ],\n",
       "          [10.106712  ],\n",
       "          [10.106712  ]],\n",
       "\n",
       "         [[ 0.974141  ],\n",
       "          [ 0.974141  ],\n",
       "          [ 0.974141  ],\n",
       "          ...,\n",
       "          [10.106712  ],\n",
       "          [10.106712  ],\n",
       "          [10.106712  ]],\n",
       "\n",
       "         [[ 0.974141  ],\n",
       "          [ 0.974141  ],\n",
       "          [ 0.9335518 ],\n",
       "          ...,\n",
       "          [10.106712  ],\n",
       "          [10.106712  ],\n",
       "          [10.106712  ]]],\n",
       "\n",
       "\n",
       "        [[[ 1.4612114 ],\n",
       "          [ 1.4612114 ],\n",
       "          [ 1.4206222 ],\n",
       "          ...,\n",
       "          [10.350248  ],\n",
       "          [ 0.        ],\n",
       "          [ 0.        ]],\n",
       "\n",
       "         [[ 1.4612114 ],\n",
       "          [ 1.4612114 ],\n",
       "          [ 1.4206222 ],\n",
       "          ...,\n",
       "          [10.350248  ],\n",
       "          [ 0.        ],\n",
       "          [ 0.        ]],\n",
       "\n",
       "         [[ 1.3394439 ],\n",
       "          [ 1.3394439 ],\n",
       "          [ 1.3394439 ],\n",
       "          ...,\n",
       "          [10.350248  ],\n",
       "          [10.350248  ],\n",
       "          [10.350248  ]],\n",
       "\n",
       "         ...,\n",
       "\n",
       "         [[ 0.9335518 ],\n",
       "          [ 0.9335518 ],\n",
       "          [ 0.9335518 ],\n",
       "          ...,\n",
       "          [10.106712  ],\n",
       "          [10.106712  ],\n",
       "          [10.106712  ]],\n",
       "\n",
       "         [[ 0.9335518 ],\n",
       "          [ 0.9335518 ],\n",
       "          [ 0.9335518 ],\n",
       "          ...,\n",
       "          [10.106712  ],\n",
       "          [10.106712  ],\n",
       "          [10.106712  ]],\n",
       "\n",
       "         [[ 0.9335518 ],\n",
       "          [ 0.9335518 ],\n",
       "          [ 0.9335518 ],\n",
       "          ...,\n",
       "          [10.106712  ],\n",
       "          [10.106712  ],\n",
       "          [10.106712  ]]]]], dtype=float32)"
      ]
     },
     "execution_count": 124,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "frames"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 125,
   "id": "12e55fbd",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "2"
      ]
     },
     "execution_count": 125,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "len(frames)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 126,
   "id": "7bcd5b83",
   "metadata": {},
   "outputs": [],
   "source": [
    "test = data.as_numpy_iterator()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 127,
   "id": "efa2c2e6",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([[[[[1.5577686 ],\n",
       "          [1.5215415 ],\n",
       "          [1.4853142 ],\n",
       "          ...,\n",
       "          [0.36227176],\n",
       "          [0.36227176],\n",
       "          [0.36227176]],\n",
       "\n",
       "         [[1.6302229 ],\n",
       "          [1.4128599 ],\n",
       "          [1.4853142 ],\n",
       "          ...,\n",
       "          [0.36227176],\n",
       "          [0.36227176],\n",
       "          [0.36227176]],\n",
       "\n",
       "         [[1.4853142 ],\n",
       "          [1.4853142 ],\n",
       "          [1.5215415 ],\n",
       "          ...,\n",
       "          [0.39849895],\n",
       "          [0.3260446 ],\n",
       "          [0.3260446 ]],\n",
       "\n",
       "         ...,\n",
       "\n",
       "         [[1.0868154 ],\n",
       "          [1.0505881 ],\n",
       "          [1.0505881 ],\n",
       "          ...,\n",
       "          [0.14490871],\n",
       "          [0.14490871],\n",
       "          [0.14490871]],\n",
       "\n",
       "         [[1.0505881 ],\n",
       "          [1.0505881 ],\n",
       "          [1.0505881 ],\n",
       "          ...,\n",
       "          [0.10868153],\n",
       "          [0.10868153],\n",
       "          [0.10868153]],\n",
       "\n",
       "         [[1.0505881 ],\n",
       "          [1.0505881 ],\n",
       "          [1.0505881 ],\n",
       "          ...,\n",
       "          [0.10868153],\n",
       "          [0.10868153],\n",
       "          [0.10868153]]],\n",
       "\n",
       "\n",
       "        [[[1.449087  ],\n",
       "          [1.3766327 ],\n",
       "          [1.4853142 ],\n",
       "          ...,\n",
       "          [0.36227176],\n",
       "          [0.36227176],\n",
       "          [0.36227176]],\n",
       "\n",
       "         [[1.449087  ],\n",
       "          [1.3766327 ],\n",
       "          [1.4128599 ],\n",
       "          ...,\n",
       "          [0.36227176],\n",
       "          [0.36227176],\n",
       "          [0.36227176]],\n",
       "\n",
       "         [[1.4853142 ],\n",
       "          [1.4853142 ],\n",
       "          [1.449087  ],\n",
       "          ...,\n",
       "          [0.36227176],\n",
       "          [0.36227176],\n",
       "          [0.36227176]],\n",
       "\n",
       "         ...,\n",
       "\n",
       "         [[1.0505881 ],\n",
       "          [1.0143609 ],\n",
       "          [1.0143609 ],\n",
       "          ...,\n",
       "          [0.14490871],\n",
       "          [0.10868153],\n",
       "          [0.10868153]],\n",
       "\n",
       "         [[1.0143609 ],\n",
       "          [1.0143609 ],\n",
       "          [1.0143609 ],\n",
       "          ...,\n",
       "          [0.10868153],\n",
       "          [0.07245436],\n",
       "          [0.07245436]],\n",
       "\n",
       "         [[1.0143609 ],\n",
       "          [1.0143609 ],\n",
       "          [1.0143609 ],\n",
       "          ...,\n",
       "          [0.07245436],\n",
       "          [0.07245436],\n",
       "          [0.03622718]]],\n",
       "\n",
       "\n",
       "        [[[1.4128599 ],\n",
       "          [1.3404056 ],\n",
       "          [1.449087  ],\n",
       "          ...,\n",
       "          [0.3260446 ],\n",
       "          [0.36227176],\n",
       "          [0.36227176]],\n",
       "\n",
       "         [[1.4128599 ],\n",
       "          [1.2679513 ],\n",
       "          [1.449087  ],\n",
       "          ...,\n",
       "          [0.3260446 ],\n",
       "          [0.36227176],\n",
       "          [0.36227176]],\n",
       "\n",
       "         [[1.4853142 ],\n",
       "          [1.4853142 ],\n",
       "          [1.4128599 ],\n",
       "          ...,\n",
       "          [0.39849895],\n",
       "          [0.36227176],\n",
       "          [0.36227176]],\n",
       "\n",
       "         ...,\n",
       "\n",
       "         [[1.0505881 ],\n",
       "          [1.0143609 ],\n",
       "          [1.0143609 ],\n",
       "          ...,\n",
       "          [0.18113588],\n",
       "          [0.14490871],\n",
       "          [0.14490871]],\n",
       "\n",
       "         [[1.0143609 ],\n",
       "          [1.0143609 ],\n",
       "          [1.0143609 ],\n",
       "          ...,\n",
       "          [0.14490871],\n",
       "          [0.14490871],\n",
       "          [0.14490871]],\n",
       "\n",
       "         [[1.0143609 ],\n",
       "          [1.0143609 ],\n",
       "          [1.0143609 ],\n",
       "          ...,\n",
       "          [0.10868153],\n",
       "          [0.10868153],\n",
       "          [0.10868153]]],\n",
       "\n",
       "\n",
       "        ...,\n",
       "\n",
       "\n",
       "        [[[1.5577686 ],\n",
       "          [1.5215415 ],\n",
       "          [1.3404056 ],\n",
       "          ...,\n",
       "          [0.36227176],\n",
       "          [0.36227176],\n",
       "          [0.36227176]],\n",
       "\n",
       "         [[1.6302229 ],\n",
       "          [1.449087  ],\n",
       "          [1.3404056 ],\n",
       "          ...,\n",
       "          [0.36227176],\n",
       "          [0.36227176],\n",
       "          [0.36227176]],\n",
       "\n",
       "         [[1.449087  ],\n",
       "          [1.449087  ],\n",
       "          [1.3766327 ],\n",
       "          ...,\n",
       "          [0.36227176],\n",
       "          [0.36227176],\n",
       "          [0.36227176]],\n",
       "\n",
       "         ...,\n",
       "\n",
       "         [[1.0143609 ],\n",
       "          [1.0143609 ],\n",
       "          [1.0143609 ],\n",
       "          ...,\n",
       "          [0.18113588],\n",
       "          [0.14490871],\n",
       "          [0.10868153]],\n",
       "\n",
       "         [[1.0143609 ],\n",
       "          [1.0143609 ],\n",
       "          [1.0143609 ],\n",
       "          ...,\n",
       "          [0.14490871],\n",
       "          [0.14490871],\n",
       "          [0.10868153]],\n",
       "\n",
       "         [[1.0143609 ],\n",
       "          [1.0143609 ],\n",
       "          [0.9781338 ],\n",
       "          ...,\n",
       "          [0.10868153],\n",
       "          [0.10868153],\n",
       "          [0.07245436]]],\n",
       "\n",
       "\n",
       "        [[[1.5215415 ],\n",
       "          [1.4853142 ],\n",
       "          [1.3766327 ],\n",
       "          ...,\n",
       "          [0.36227176],\n",
       "          [0.36227176],\n",
       "          [0.36227176]],\n",
       "\n",
       "         [[1.5577686 ],\n",
       "          [1.4853142 ],\n",
       "          [1.4128599 ],\n",
       "          ...,\n",
       "          [0.36227176],\n",
       "          [0.36227176],\n",
       "          [0.36227176]],\n",
       "\n",
       "         [[1.4128599 ],\n",
       "          [1.4128599 ],\n",
       "          [1.4128599 ],\n",
       "          ...,\n",
       "          [0.36227176],\n",
       "          [0.36227176],\n",
       "          [0.36227176]],\n",
       "\n",
       "         ...,\n",
       "\n",
       "         [[1.0143609 ],\n",
       "          [1.0143609 ],\n",
       "          [1.0143609 ],\n",
       "          ...,\n",
       "          [0.14490871],\n",
       "          [0.14490871],\n",
       "          [0.10868153]],\n",
       "\n",
       "         [[1.0143609 ],\n",
       "          [1.0143609 ],\n",
       "          [1.0143609 ],\n",
       "          ...,\n",
       "          [0.14490871],\n",
       "          [0.10868153],\n",
       "          [0.07245436]],\n",
       "\n",
       "         [[1.0143609 ],\n",
       "          [1.0143609 ],\n",
       "          [0.9781338 ],\n",
       "          ...,\n",
       "          [0.10868153],\n",
       "          [0.07245436],\n",
       "          [0.07245436]]],\n",
       "\n",
       "\n",
       "        [[[1.4853142 ],\n",
       "          [1.4853142 ],\n",
       "          [1.4128599 ],\n",
       "          ...,\n",
       "          [0.36227176],\n",
       "          [0.36227176],\n",
       "          [0.36227176]],\n",
       "\n",
       "         [[1.4128599 ],\n",
       "          [1.4853142 ],\n",
       "          [1.449087  ],\n",
       "          ...,\n",
       "          [0.36227176],\n",
       "          [0.36227176],\n",
       "          [0.36227176]],\n",
       "\n",
       "         [[1.4128599 ],\n",
       "          [1.4128599 ],\n",
       "          [1.4128599 ],\n",
       "          ...,\n",
       "          [0.39849895],\n",
       "          [0.36227176],\n",
       "          [0.36227176]],\n",
       "\n",
       "         ...,\n",
       "\n",
       "         [[1.0143609 ],\n",
       "          [1.0143609 ],\n",
       "          [1.0143609 ],\n",
       "          ...,\n",
       "          [0.14490871],\n",
       "          [0.14490871],\n",
       "          [0.10868153]],\n",
       "\n",
       "         [[1.0143609 ],\n",
       "          [1.0143609 ],\n",
       "          [1.0143609 ],\n",
       "          ...,\n",
       "          [0.14490871],\n",
       "          [0.10868153],\n",
       "          [0.07245436]],\n",
       "\n",
       "         [[1.0143609 ],\n",
       "          [1.0143609 ],\n",
       "          [1.0143609 ],\n",
       "          ...,\n",
       "          [0.10868153],\n",
       "          [0.07245436],\n",
       "          [0.07245436]]]],\n",
       "\n",
       "\n",
       "\n",
       "       [[[[1.3891008 ],\n",
       "          [1.3891008 ],\n",
       "          [1.3891008 ],\n",
       "          ...,\n",
       "          [8.627047  ],\n",
       "          [9.138821  ],\n",
       "          [9.285042  ]],\n",
       "\n",
       "         [[1.3891008 ],\n",
       "          [1.3891008 ],\n",
       "          [1.3891008 ],\n",
       "          ...,\n",
       "          [8.517381  ],\n",
       "          [9.321597  ],\n",
       "          [9.211931  ]],\n",
       "\n",
       "         [[1.2794349 ],\n",
       "          [1.2794349 ],\n",
       "          [1.2794349 ],\n",
       "          ...,\n",
       "          [0.10966585],\n",
       "          [0.10966585],\n",
       "          [0.10966585]],\n",
       "\n",
       "         ...,\n",
       "\n",
       "         [[0.95043737],\n",
       "          [0.95043737],\n",
       "          [0.95043737],\n",
       "          ...,\n",
       "          [9.175376  ],\n",
       "          [9.138821  ],\n",
       "          [9.138821  ]],\n",
       "\n",
       "         [[0.9138821 ],\n",
       "          [0.9138821 ],\n",
       "          [0.9138821 ],\n",
       "          ...,\n",
       "          [9.138821  ],\n",
       "          [9.175376  ],\n",
       "          [9.175376  ]],\n",
       "\n",
       "         [[0.9138821 ],\n",
       "          [0.9138821 ],\n",
       "          [0.9138821 ],\n",
       "          ...,\n",
       "          [9.138821  ],\n",
       "          [9.175376  ],\n",
       "          [9.175376  ]]],\n",
       "\n",
       "\n",
       "        [[[1.3159902 ],\n",
       "          [1.3159902 ],\n",
       "          [1.3159902 ],\n",
       "          ...,\n",
       "          [8.700157  ],\n",
       "          [9.06571   ],\n",
       "          [9.2484865 ]],\n",
       "\n",
       "         [[1.3159902 ],\n",
       "          [1.3159902 ],\n",
       "          [1.3159902 ],\n",
       "          ...,\n",
       "          [8.590491  ],\n",
       "          [9.2484865 ],\n",
       "          [9.211931  ]],\n",
       "\n",
       "         [[1.2794349 ],\n",
       "          [1.2794349 ],\n",
       "          [1.2794349 ],\n",
       "          ...,\n",
       "          [9.321597  ],\n",
       "          [0.07311057],\n",
       "          [0.07311057]],\n",
       "\n",
       "         ...,\n",
       "\n",
       "         [[0.9138821 ],\n",
       "          [0.9138821 ],\n",
       "          [0.9138821 ],\n",
       "          ...,\n",
       "          [9.211931  ],\n",
       "          [9.211931  ],\n",
       "          [9.211931  ]],\n",
       "\n",
       "         [[0.8773268 ],\n",
       "          [0.8773268 ],\n",
       "          [0.8773268 ],\n",
       "          ...,\n",
       "          [9.138821  ],\n",
       "          [9.138821  ],\n",
       "          [9.138821  ]],\n",
       "\n",
       "         [[0.8773268 ],\n",
       "          [0.8773268 ],\n",
       "          [0.8773268 ],\n",
       "          ...,\n",
       "          [9.138821  ],\n",
       "          [9.138821  ],\n",
       "          [9.138821  ]]],\n",
       "\n",
       "\n",
       "        [[[1.2794349 ],\n",
       "          [1.2794349 ],\n",
       "          [1.2794349 ],\n",
       "          ...,\n",
       "          [8.736712  ],\n",
       "          [9.102265  ],\n",
       "          [9.2484865 ]],\n",
       "\n",
       "         [[1.2794349 ],\n",
       "          [1.2794349 ],\n",
       "          [1.2794349 ],\n",
       "          ...,\n",
       "          [8.517381  ],\n",
       "          [9.175376  ],\n",
       "          [9.2484865 ]],\n",
       "\n",
       "         [[1.2428796 ],\n",
       "          [1.2428796 ],\n",
       "          [1.2428796 ],\n",
       "          ...,\n",
       "          [0.03655528],\n",
       "          [0.03655528],\n",
       "          [0.03655528]],\n",
       "\n",
       "         ...,\n",
       "\n",
       "         [[0.95043737],\n",
       "          [0.95043737],\n",
       "          [0.9138821 ],\n",
       "          ...,\n",
       "          [9.2484865 ],\n",
       "          [9.211931  ],\n",
       "          [9.211931  ]],\n",
       "\n",
       "         [[0.9138821 ],\n",
       "          [0.9138821 ],\n",
       "          [0.8773268 ],\n",
       "          ...,\n",
       "          [9.175376  ],\n",
       "          [9.138821  ],\n",
       "          [9.138821  ]],\n",
       "\n",
       "         [[0.9138821 ],\n",
       "          [0.9138821 ],\n",
       "          [0.8773268 ],\n",
       "          ...,\n",
       "          [9.175376  ],\n",
       "          [9.138821  ],\n",
       "          [9.138821  ]]],\n",
       "\n",
       "\n",
       "        ...,\n",
       "\n",
       "\n",
       "        [[[1.2794349 ],\n",
       "          [1.2794349 ],\n",
       "          [1.2794349 ],\n",
       "          ...,\n",
       "          [8.956044  ],\n",
       "          [0.        ],\n",
       "          [0.        ]],\n",
       "\n",
       "         [[1.2794349 ],\n",
       "          [1.2794349 ],\n",
       "          [1.2794349 ],\n",
       "          ...,\n",
       "          [9.029155  ],\n",
       "          [0.        ],\n",
       "          [0.07311057]],\n",
       "\n",
       "         [[1.2428796 ],\n",
       "          [1.2428796 ],\n",
       "          [1.2428796 ],\n",
       "          ...,\n",
       "          [0.03655528],\n",
       "          [0.07311057],\n",
       "          [0.07311057]],\n",
       "\n",
       "         ...,\n",
       "\n",
       "         [[0.8773268 ],\n",
       "          [0.8773268 ],\n",
       "          [0.8407715 ],\n",
       "          ...,\n",
       "          [9.138821  ],\n",
       "          [9.102265  ],\n",
       "          [9.102265  ]],\n",
       "\n",
       "         [[0.8773268 ],\n",
       "          [0.8773268 ],\n",
       "          [0.8407715 ],\n",
       "          ...,\n",
       "          [9.138821  ],\n",
       "          [9.102265  ],\n",
       "          [9.102265  ]],\n",
       "\n",
       "         [[0.8773268 ],\n",
       "          [0.8773268 ],\n",
       "          [0.8407715 ],\n",
       "          ...,\n",
       "          [9.138821  ],\n",
       "          [9.102265  ],\n",
       "          [9.102265  ]]],\n",
       "\n",
       "\n",
       "        [[[1.2794349 ],\n",
       "          [1.2794349 ],\n",
       "          [1.2794349 ],\n",
       "          ...,\n",
       "          [8.956044  ],\n",
       "          [0.        ],\n",
       "          [0.07311057]],\n",
       "\n",
       "         [[1.2794349 ],\n",
       "          [1.3525455 ],\n",
       "          [1.2794349 ],\n",
       "          ...,\n",
       "          [8.919489  ],\n",
       "          [0.        ],\n",
       "          [0.07311057]],\n",
       "\n",
       "         [[1.2794349 ],\n",
       "          [1.2428796 ],\n",
       "          [1.2428796 ],\n",
       "          ...,\n",
       "          [0.03655528],\n",
       "          [0.07311057],\n",
       "          [0.07311057]],\n",
       "\n",
       "         ...,\n",
       "\n",
       "         [[0.8773268 ],\n",
       "          [0.8773268 ],\n",
       "          [0.8773268 ],\n",
       "          ...,\n",
       "          [9.102265  ],\n",
       "          [9.06571   ],\n",
       "          [9.06571   ]],\n",
       "\n",
       "         [[0.8773268 ],\n",
       "          [0.8773268 ],\n",
       "          [0.8773268 ],\n",
       "          ...,\n",
       "          [9.102265  ],\n",
       "          [9.06571   ],\n",
       "          [9.06571   ]],\n",
       "\n",
       "         [[0.8773268 ],\n",
       "          [0.8773268 ],\n",
       "          [0.8773268 ],\n",
       "          ...,\n",
       "          [9.102265  ],\n",
       "          [9.06571   ],\n",
       "          [9.06571   ]]],\n",
       "\n",
       "\n",
       "        [[[1.2794349 ],\n",
       "          [1.2794349 ],\n",
       "          [1.2794349 ],\n",
       "          ...,\n",
       "          [8.9925995 ],\n",
       "          [9.321597  ],\n",
       "          [0.10966585]],\n",
       "\n",
       "         [[1.2794349 ],\n",
       "          [1.3525455 ],\n",
       "          [1.2794349 ],\n",
       "          ...,\n",
       "          [8.919489  ],\n",
       "          [0.07311057],\n",
       "          [0.        ]],\n",
       "\n",
       "         [[1.3525455 ],\n",
       "          [1.2794349 ],\n",
       "          [1.2794349 ],\n",
       "          ...,\n",
       "          [0.        ],\n",
       "          [0.10966585],\n",
       "          [0.1827764 ]],\n",
       "\n",
       "         ...,\n",
       "\n",
       "         [[0.95043737],\n",
       "          [0.9138821 ],\n",
       "          [0.9138821 ],\n",
       "          ...,\n",
       "          [9.138821  ],\n",
       "          [9.102265  ],\n",
       "          [9.102265  ]],\n",
       "\n",
       "         [[0.95043737],\n",
       "          [0.9138821 ],\n",
       "          [0.9138821 ],\n",
       "          ...,\n",
       "          [9.138821  ],\n",
       "          [9.102265  ],\n",
       "          [9.102265  ]],\n",
       "\n",
       "         [[0.95043737],\n",
       "          [0.9138821 ],\n",
       "          [0.9138821 ],\n",
       "          ...,\n",
       "          [9.138821  ],\n",
       "          [9.102265  ],\n",
       "          [9.102265  ]]]]], dtype=float32)"
      ]
     },
     "execution_count": 127,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "val = test.next(); val[0]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 128,
   "id": "0efe7107",
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Lossy conversion from float32 to uint8. Range [0.0, 9.237930297851562]. Convert image to uint8 prior to saving to suppress this warning.\n",
      "Lossy conversion from float32 to uint8. Range [0.0, 9.237930297851562]. Convert image to uint8 prior to saving to suppress this warning.\n",
      "Lossy conversion from float32 to uint8. Range [0.0, 9.237930297851562]. Convert image to uint8 prior to saving to suppress this warning.\n",
      "Lossy conversion from float32 to uint8. Range [0.0, 9.237930297851562]. Convert image to uint8 prior to saving to suppress this warning.\n",
      "Lossy conversion from float32 to uint8. Range [0.0, 9.237930297851562]. Convert image to uint8 prior to saving to suppress this warning.\n",
      "Lossy conversion from float32 to uint8. Range [0.0, 9.237930297851562]. Convert image to uint8 prior to saving to suppress this warning.\n",
      "Lossy conversion from float32 to uint8. Range [0.0, 9.237930297851562]. Convert image to uint8 prior to saving to suppress this warning.\n",
      "Lossy conversion from float32 to uint8. Range [0.0, 9.237930297851562]. Convert image to uint8 prior to saving to suppress this warning.\n",
      "Lossy conversion from float32 to uint8. Range [0.0, 9.237930297851562]. Convert image to uint8 prior to saving to suppress this warning.\n",
      "Lossy conversion from float32 to uint8. Range [0.0, 9.237930297851562]. Convert image to uint8 prior to saving to suppress this warning.\n",
      "Lossy conversion from float32 to uint8. Range [0.0, 9.237930297851562]. Convert image to uint8 prior to saving to suppress this warning.\n",
      "Lossy conversion from float32 to uint8. Range [0.0, 9.237930297851562]. Convert image to uint8 prior to saving to suppress this warning.\n",
      "Lossy conversion from float32 to uint8. Range [0.0, 9.237930297851562]. Convert image to uint8 prior to saving to suppress this warning.\n",
      "Lossy conversion from float32 to uint8. Range [0.0, 9.237930297851562]. Convert image to uint8 prior to saving to suppress this warning.\n",
      "Lossy conversion from float32 to uint8. Range [0.0, 9.237930297851562]. Convert image to uint8 prior to saving to suppress this warning.\n",
      "Lossy conversion from float32 to uint8. Range [0.0, 9.237930297851562]. Convert image to uint8 prior to saving to suppress this warning.\n",
      "Lossy conversion from float32 to uint8. Range [0.0, 9.237930297851562]. Convert image to uint8 prior to saving to suppress this warning.\n",
      "Lossy conversion from float32 to uint8. Range [0.0, 9.237930297851562]. Convert image to uint8 prior to saving to suppress this warning.\n",
      "Lossy conversion from float32 to uint8. Range [0.0, 9.237930297851562]. Convert image to uint8 prior to saving to suppress this warning.\n",
      "Lossy conversion from float32 to uint8. Range [0.0, 9.237930297851562]. Convert image to uint8 prior to saving to suppress this warning.\n",
      "Lossy conversion from float32 to uint8. Range [0.0, 9.237930297851562]. Convert image to uint8 prior to saving to suppress this warning.\n",
      "Lossy conversion from float32 to uint8. Range [0.0, 9.237930297851562]. Convert image to uint8 prior to saving to suppress this warning.\n",
      "Lossy conversion from float32 to uint8. Range [0.0, 9.237930297851562]. Convert image to uint8 prior to saving to suppress this warning.\n",
      "Lossy conversion from float32 to uint8. Range [0.0, 9.237930297851562]. Convert image to uint8 prior to saving to suppress this warning.\n",
      "Lossy conversion from float32 to uint8. Range [0.0, 9.237930297851562]. Convert image to uint8 prior to saving to suppress this warning.\n",
      "Lossy conversion from float32 to uint8. Range [0.0, 9.237930297851562]. Convert image to uint8 prior to saving to suppress this warning.\n",
      "Lossy conversion from float32 to uint8. Range [0.0, 9.237930297851562]. Convert image to uint8 prior to saving to suppress this warning.\n",
      "Lossy conversion from float32 to uint8. Range [0.0, 9.237930297851562]. Convert image to uint8 prior to saving to suppress this warning.\n",
      "Lossy conversion from float32 to uint8. Range [0.0, 9.237930297851562]. Convert image to uint8 prior to saving to suppress this warning.\n",
      "Lossy conversion from float32 to uint8. Range [0.0, 9.237930297851562]. Convert image to uint8 prior to saving to suppress this warning.\n",
      "Lossy conversion from float32 to uint8. Range [0.0, 9.237930297851562]. Convert image to uint8 prior to saving to suppress this warning.\n",
      "Lossy conversion from float32 to uint8. Range [0.0, 9.237930297851562]. Convert image to uint8 prior to saving to suppress this warning.\n",
      "Lossy conversion from float32 to uint8. Range [0.0, 9.237930297851562]. Convert image to uint8 prior to saving to suppress this warning.\n",
      "Lossy conversion from float32 to uint8. Range [0.0, 9.237930297851562]. Convert image to uint8 prior to saving to suppress this warning.\n",
      "Lossy conversion from float32 to uint8. Range [0.0, 9.237930297851562]. Convert image to uint8 prior to saving to suppress this warning.\n",
      "Lossy conversion from float32 to uint8. Range [0.0, 9.237930297851562]. Convert image to uint8 prior to saving to suppress this warning.\n",
      "Lossy conversion from float32 to uint8. Range [0.0, 9.237930297851562]. Convert image to uint8 prior to saving to suppress this warning.\n",
      "Lossy conversion from float32 to uint8. Range [0.0, 9.237930297851562]. Convert image to uint8 prior to saving to suppress this warning.\n",
      "Lossy conversion from float32 to uint8. Range [0.0, 9.237930297851562]. Convert image to uint8 prior to saving to suppress this warning.\n",
      "Lossy conversion from float32 to uint8. Range [0.0, 9.237930297851562]. Convert image to uint8 prior to saving to suppress this warning.\n",
      "Lossy conversion from float32 to uint8. Range [0.0, 9.237930297851562]. Convert image to uint8 prior to saving to suppress this warning.\n",
      "Lossy conversion from float32 to uint8. Range [0.0, 9.237930297851562]. Convert image to uint8 prior to saving to suppress this warning.\n",
      "Lossy conversion from float32 to uint8. Range [0.0, 9.237930297851562]. Convert image to uint8 prior to saving to suppress this warning.\n",
      "Lossy conversion from float32 to uint8. Range [0.0, 9.237930297851562]. Convert image to uint8 prior to saving to suppress this warning.\n",
      "Lossy conversion from float32 to uint8. Range [0.0, 9.237930297851562]. Convert image to uint8 prior to saving to suppress this warning.\n",
      "Lossy conversion from float32 to uint8. Range [0.0, 9.237930297851562]. Convert image to uint8 prior to saving to suppress this warning.\n",
      "Lossy conversion from float32 to uint8. Range [0.0, 9.237930297851562]. Convert image to uint8 prior to saving to suppress this warning.\n",
      "Lossy conversion from float32 to uint8. Range [0.0, 9.237930297851562]. Convert image to uint8 prior to saving to suppress this warning.\n",
      "Lossy conversion from float32 to uint8. Range [0.0, 9.237930297851562]. Convert image to uint8 prior to saving to suppress this warning.\n",
      "Lossy conversion from float32 to uint8. Range [0.0, 9.237930297851562]. Convert image to uint8 prior to saving to suppress this warning.\n",
      "Lossy conversion from float32 to uint8. Range [0.0, 9.237930297851562]. Convert image to uint8 prior to saving to suppress this warning.\n",
      "Lossy conversion from float32 to uint8. Range [0.0, 9.237930297851562]. Convert image to uint8 prior to saving to suppress this warning.\n",
      "Lossy conversion from float32 to uint8. Range [0.0, 9.237930297851562]. Convert image to uint8 prior to saving to suppress this warning.\n",
      "Lossy conversion from float32 to uint8. Range [0.0, 9.237930297851562]. Convert image to uint8 prior to saving to suppress this warning.\n",
      "Lossy conversion from float32 to uint8. Range [0.0, 9.237930297851562]. Convert image to uint8 prior to saving to suppress this warning.\n",
      "Lossy conversion from float32 to uint8. Range [0.0, 9.237930297851562]. Convert image to uint8 prior to saving to suppress this warning.\n",
      "Lossy conversion from float32 to uint8. Range [0.0, 9.237930297851562]. Convert image to uint8 prior to saving to suppress this warning.\n",
      "Lossy conversion from float32 to uint8. Range [0.0, 9.237930297851562]. Convert image to uint8 prior to saving to suppress this warning.\n",
      "Lossy conversion from float32 to uint8. Range [0.0, 9.237930297851562]. Convert image to uint8 prior to saving to suppress this warning.\n",
      "Lossy conversion from float32 to uint8. Range [0.0, 9.237930297851562]. Convert image to uint8 prior to saving to suppress this warning.\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Lossy conversion from float32 to uint8. Range [0.0, 9.237930297851562]. Convert image to uint8 prior to saving to suppress this warning.\n",
      "Lossy conversion from float32 to uint8. Range [0.0, 9.237930297851562]. Convert image to uint8 prior to saving to suppress this warning.\n",
      "Lossy conversion from float32 to uint8. Range [0.0, 9.237930297851562]. Convert image to uint8 prior to saving to suppress this warning.\n",
      "Lossy conversion from float32 to uint8. Range [0.0, 9.237930297851562]. Convert image to uint8 prior to saving to suppress this warning.\n",
      "Lossy conversion from float32 to uint8. Range [0.0, 9.237930297851562]. Convert image to uint8 prior to saving to suppress this warning.\n",
      "Lossy conversion from float32 to uint8. Range [0.0, 9.237930297851562]. Convert image to uint8 prior to saving to suppress this warning.\n",
      "Lossy conversion from float32 to uint8. Range [0.0, 9.237930297851562]. Convert image to uint8 prior to saving to suppress this warning.\n",
      "Lossy conversion from float32 to uint8. Range [0.0, 9.237930297851562]. Convert image to uint8 prior to saving to suppress this warning.\n",
      "Lossy conversion from float32 to uint8. Range [0.0, 9.237930297851562]. Convert image to uint8 prior to saving to suppress this warning.\n",
      "Lossy conversion from float32 to uint8. Range [0.0, 9.237930297851562]. Convert image to uint8 prior to saving to suppress this warning.\n",
      "Lossy conversion from float32 to uint8. Range [0.0, 9.237930297851562]. Convert image to uint8 prior to saving to suppress this warning.\n",
      "Lossy conversion from float32 to uint8. Range [0.0, 9.237930297851562]. Convert image to uint8 prior to saving to suppress this warning.\n",
      "Lossy conversion from float32 to uint8. Range [0.0, 9.237930297851562]. Convert image to uint8 prior to saving to suppress this warning.\n",
      "Lossy conversion from float32 to uint8. Range [0.0, 9.237930297851562]. Convert image to uint8 prior to saving to suppress this warning.\n",
      "Lossy conversion from float32 to uint8. Range [0.0, 9.237930297851562]. Convert image to uint8 prior to saving to suppress this warning.\n"
     ]
    }
   ],
   "source": [
    "#converting array to gif\n",
    "imageio.mimsave('./animation.gif', val[0][0], fps=10)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 129,
   "id": "5337943b",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.image.AxesImage at 0x7ff5513dc460>"
      ]
     },
     "execution_count": 129,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# 0:videos, 0: 1st video out of the batch,  0: return the 1st frame in the video \n",
    "plt.imshow(val[0][0][35])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 130,
   "id": "69aa9afa",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<tf.Tensor: shape=(), dtype=string, numpy=b'set white with p five again'>"
      ]
     },
     "execution_count": 130,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "#annotation for the video we selected\n",
    "tf.strings.reduce_join([num_to_char(word) for word in val[1][0]])"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "31190804",
   "metadata": {},
   "source": [
    "# Designing the Deep Neural Network"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 131,
   "id": "12942e6d",
   "metadata": {},
   "outputs": [],
   "source": [
    "from tensorflow.keras.models import Sequential \n",
    "from tensorflow.keras.layers import Conv3D, LSTM, Dense, Dropout, Bidirectional, MaxPool3D, Activation, Reshape, SpatialDropout3D, BatchNormalization, TimeDistributed, Flatten\n",
    "from tensorflow.keras.optimizers import Adam\n",
    "from tensorflow.keras.callbacks import ModelCheckpoint, LearningRateScheduler"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 132,
   "id": "2654dda4",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(75, 46, 140, 1)"
      ]
     },
     "execution_count": 132,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "data.as_numpy_iterator().next()[0][0].shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 133,
   "id": "f7c03043",
   "metadata": {},
   "outputs": [],
   "source": [
    "model = Sequential()\n",
    "model.add(Conv3D(128, 3, input_shape=(75,46,140,1), padding='same'))\n",
    "model.add(Activation('relu'))\n",
    "model.add(MaxPool3D((1,2,2)))\n",
    "\n",
    "model.add(Conv3D(256, 3, padding='same'))\n",
    "model.add(Activation('relu'))\n",
    "model.add(MaxPool3D((1,2,2)))\n",
    "\n",
    "model.add(Conv3D(75, 3, padding='same'))\n",
    "model.add(Activation('relu'))\n",
    "model.add(MaxPool3D((1,2,2)))\n",
    "\n",
    "model.add(TimeDistributed(Flatten()))\n",
    "\n",
    "model.add(Bidirectional(LSTM(128, kernel_initializer='Orthogonal', return_sequences=True)))\n",
    "model.add(Dropout(.5))\n",
    "\n",
    "model.add(Bidirectional(LSTM(128, kernel_initializer='Orthogonal', return_sequences=True)))\n",
    "model.add(Dropout(.5))\n",
    "\n",
    "model.add(Dense(char_to_num.vocabulary_size()+1, kernel_initializer='he_normal', activation='softmax'))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 134,
   "id": "20b92144",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Model: \"sequential_2\"\n",
      "_________________________________________________________________\n",
      " Layer (type)                Output Shape              Param #   \n",
      "=================================================================\n",
      " conv3d_6 (Conv3D)           (None, 75, 46, 140, 128)  3584      \n",
      "                                                                 \n",
      " activation_6 (Activation)   (None, 75, 46, 140, 128)  0         \n",
      "                                                                 \n",
      " max_pooling3d_6 (MaxPooling  (None, 75, 23, 70, 128)  0         \n",
      " 3D)                                                             \n",
      "                                                                 \n",
      " conv3d_7 (Conv3D)           (None, 75, 23, 70, 256)   884992    \n",
      "                                                                 \n",
      " activation_7 (Activation)   (None, 75, 23, 70, 256)   0         \n",
      "                                                                 \n",
      " max_pooling3d_7 (MaxPooling  (None, 75, 11, 35, 256)  0         \n",
      " 3D)                                                             \n",
      "                                                                 \n",
      " conv3d_8 (Conv3D)           (None, 75, 11, 35, 75)    518475    \n",
      "                                                                 \n",
      " activation_8 (Activation)   (None, 75, 11, 35, 75)    0         \n",
      "                                                                 \n",
      " max_pooling3d_8 (MaxPooling  (None, 75, 5, 17, 75)    0         \n",
      " 3D)                                                             \n",
      "                                                                 \n",
      " time_distributed_2 (TimeDis  (None, 75, 6375)         0         \n",
      " tributed)                                                       \n",
      "                                                                 \n",
      " bidirectional_4 (Bidirectio  (None, 75, 256)          6660096   \n",
      " nal)                                                            \n",
      "                                                                 \n",
      " dropout_4 (Dropout)         (None, 75, 256)           0         \n",
      "                                                                 \n",
      " bidirectional_5 (Bidirectio  (None, 75, 256)          394240    \n",
      " nal)                                                            \n",
      "                                                                 \n",
      " dropout_5 (Dropout)         (None, 75, 256)           0         \n",
      "                                                                 \n",
      " dense_2 (Dense)             (None, 75, 41)            10537     \n",
      "                                                                 \n",
      "=================================================================\n",
      "Total params: 8,471,924\n",
      "Trainable params: 8,471,924\n",
      "Non-trainable params: 0\n",
      "_________________________________________________________________\n"
     ]
    }
   ],
   "source": [
    "model.summary()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 135,
   "id": "75d12af8",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "1/1 [==============================] - 4s 4s/step\n"
     ]
    }
   ],
   "source": [
    "yhat = model.predict(val[0])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 136,
   "id": "dc0f2fd0",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<tf.Tensor: shape=(), dtype=string, numpy=b'????????????????????????????????????????????????????????????????????????ss6'>"
      ]
     },
     "execution_count": 136,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "tf.strings.reduce_join([num_to_char(x) for x in tf.argmax(yhat[0],axis=1)])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 137,
   "id": "e9f2924f",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<tf.Tensor: shape=(), dtype=string, numpy=b'????????????????????????????????????????????????????????????????????????ss6'>"
      ]
     },
     "execution_count": 137,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "tf.strings.reduce_join([num_to_char(tf.argmax(x)) for x in yhat[0]])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 138,
   "id": "a510500a",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(None, 75, 46, 140, 1)"
      ]
     },
     "execution_count": 138,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "model.input_shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 139,
   "id": "cc5ac696",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(None, 75, 41)"
      ]
     },
     "execution_count": 139,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "model.output_shape"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "5fb8ca50",
   "metadata": {},
   "source": [
    "# Setup Training Options and Train"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 140,
   "id": "eb4df79b",
   "metadata": {},
   "outputs": [],
   "source": [
    "def scheduler(epoch, lr):\n",
    "    if epoch < 30:\n",
    "        return lr\n",
    "    else:\n",
    "        return lr * tf.math.exp(-0.1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 141,
   "id": "85874341",
   "metadata": {},
   "outputs": [],
   "source": [
    "def CTCLoss(y_true, y_pred):\n",
    "    batch_len = tf.cast(tf.shape(y_true)[0], dtype=\"int64\")\n",
    "    input_length = tf.cast(tf.shape(y_pred)[1], dtype=\"int64\")\n",
    "    label_length = tf.cast(tf.shape(y_true)[1], dtype=\"int64\")\n",
    "\n",
    "    input_length = input_length * tf.ones(shape=(batch_len, 1), dtype=\"int64\")\n",
    "    label_length = label_length * tf.ones(shape=(batch_len, 1), dtype=\"int64\")\n",
    "\n",
    "    loss = tf.keras.backend.ctc_batch_cost(y_true, y_pred, input_length, label_length)\n",
    "    return loss"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 142,
   "id": "dfaf3e47",
   "metadata": {},
   "outputs": [],
   "source": [
    "class ProduceExample(tf.keras.callbacks.Callback): \n",
    "    def __init__(self, dataset) -> None: \n",
    "        self.dataset = dataset.as_numpy_iterator()\n",
    "    \n",
    "    def on_epoch_end(self, epoch, logs=None) -> None:\n",
    "        data = self.dataset.next()\n",
    "        yhat = self.model.predict(data[0])\n",
    "        decoded = tf.keras.backend.ctc_decode(yhat, [75,75], greedy=False)[0][0].numpy()\n",
    "        for x in range(len(yhat)):           \n",
    "            print('Original:', tf.strings.reduce_join(num_to_char(data[1][x])).numpy().decode('utf-8'))\n",
    "            print('Prediction:', tf.strings.reduce_join(num_to_char(decoded[x])).numpy().decode('utf-8'))\n",
    "            print('~'*100)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 143,
   "id": "dafb5e2c",
   "metadata": {},
   "outputs": [],
   "source": [
    "model.compile(optimizer=Adam(learning_rate=0.0001), loss=CTCLoss)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 144,
   "id": "7260cd1f",
   "metadata": {},
   "outputs": [],
   "source": [
    "checkpoint_callback = ModelCheckpoint(os.path.join('models','checkpoint'), monitor='loss', save_weights_only=True) "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 145,
   "id": "f05cf6ce",
   "metadata": {},
   "outputs": [],
   "source": [
    "schedule_callback = LearningRateScheduler(scheduler)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 146,
   "id": "9889ca17",
   "metadata": {},
   "outputs": [],
   "source": [
    "example_callback = ProduceExample(data)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 147,
   "id": "eb97fe0c",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 1/100\n",
      "  5/450 [..............................] - ETA: 4:03:15 - loss: 172.1925"
     ]
    },
    {
     "ename": "KeyboardInterrupt",
     "evalue": "",
     "output_type": "error",
     "traceback": [
      "\u001b[0;31m---------------------------------------------------------------------------\u001b[0m",
      "\u001b[0;31mKeyboardInterrupt\u001b[0m                         Traceback (most recent call last)",
      "\u001b[0;32m/tmp/ipykernel_3514/3536403359.py\u001b[0m in \u001b[0;36m<module>\u001b[0;34m\u001b[0m\n\u001b[0;32m----> 1\u001b[0;31m \u001b[0mmodel\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mfit\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mtrain\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mvalidation_data\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0mtest\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mepochs\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0;36m100\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mcallbacks\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0;34m[\u001b[0m\u001b[0mcheckpoint_callback\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mschedule_callback\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mexample_callback\u001b[0m\u001b[0;34m]\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m",
      "\u001b[0;32m~/anaconda3/lib/python3.9/site-packages/keras/utils/traceback_utils.py\u001b[0m in \u001b[0;36merror_handler\u001b[0;34m(*args, **kwargs)\u001b[0m\n\u001b[1;32m     63\u001b[0m         \u001b[0mfiltered_tb\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0;32mNone\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m     64\u001b[0m         \u001b[0;32mtry\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m---> 65\u001b[0;31m             \u001b[0;32mreturn\u001b[0m \u001b[0mfn\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m*\u001b[0m\u001b[0margs\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0;34m**\u001b[0m\u001b[0mkwargs\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m     66\u001b[0m         \u001b[0;32mexcept\u001b[0m \u001b[0mException\u001b[0m \u001b[0;32mas\u001b[0m \u001b[0me\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m     67\u001b[0m             \u001b[0mfiltered_tb\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0m_process_traceback_frames\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0me\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0m__traceback__\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
      "\u001b[0;32m~/anaconda3/lib/python3.9/site-packages/keras/engine/training.py\u001b[0m in \u001b[0;36mfit\u001b[0;34m(self, x, y, batch_size, epochs, verbose, callbacks, validation_split, validation_data, shuffle, class_weight, sample_weight, initial_epoch, steps_per_epoch, validation_steps, validation_batch_size, validation_freq, max_queue_size, workers, use_multiprocessing)\u001b[0m\n\u001b[1;32m   1562\u001b[0m                         ):\n\u001b[1;32m   1563\u001b[0m                             \u001b[0mcallbacks\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mon_train_batch_begin\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mstep\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m-> 1564\u001b[0;31m                             \u001b[0mtmp_logs\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mtrain_function\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0miterator\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m   1565\u001b[0m                             \u001b[0;32mif\u001b[0m \u001b[0mdata_handler\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mshould_sync\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m   1566\u001b[0m                                 \u001b[0mcontext\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0masync_wait\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
      "\u001b[0;32m~/anaconda3/lib/python3.9/site-packages/tensorflow/python/util/traceback_utils.py\u001b[0m in \u001b[0;36merror_handler\u001b[0;34m(*args, **kwargs)\u001b[0m\n\u001b[1;32m    148\u001b[0m     \u001b[0mfiltered_tb\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0;32mNone\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    149\u001b[0m     \u001b[0;32mtry\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m--> 150\u001b[0;31m       \u001b[0;32mreturn\u001b[0m \u001b[0mfn\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m*\u001b[0m\u001b[0margs\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0;34m**\u001b[0m\u001b[0mkwargs\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m    151\u001b[0m     \u001b[0;32mexcept\u001b[0m \u001b[0mException\u001b[0m \u001b[0;32mas\u001b[0m \u001b[0me\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    152\u001b[0m       \u001b[0mfiltered_tb\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0m_process_traceback_frames\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0me\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0m__traceback__\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
      "\u001b[0;32m~/anaconda3/lib/python3.9/site-packages/tensorflow/python/eager/def_function.py\u001b[0m in \u001b[0;36m__call__\u001b[0;34m(self, *args, **kwds)\u001b[0m\n\u001b[1;32m    913\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    914\u001b[0m       \u001b[0;32mwith\u001b[0m \u001b[0mOptionalXlaContext\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0m_jit_compile\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m--> 915\u001b[0;31m         \u001b[0mresult\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0m_call\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m*\u001b[0m\u001b[0margs\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0;34m**\u001b[0m\u001b[0mkwds\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m    916\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    917\u001b[0m       \u001b[0mnew_tracing_count\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mexperimental_get_tracing_count\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
      "\u001b[0;32m~/anaconda3/lib/python3.9/site-packages/tensorflow/python/eager/def_function.py\u001b[0m in \u001b[0;36m_call\u001b[0;34m(self, *args, **kwds)\u001b[0m\n\u001b[1;32m    945\u001b[0m       \u001b[0;31m# In this case we have created variables on the first call, so we run the\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    946\u001b[0m       \u001b[0;31m# defunned version which is guaranteed to never create variables.\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m--> 947\u001b[0;31m       \u001b[0;32mreturn\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0m_stateless_fn\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m*\u001b[0m\u001b[0margs\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0;34m**\u001b[0m\u001b[0mkwds\u001b[0m\u001b[0;34m)\u001b[0m  \u001b[0;31m# pylint: disable=not-callable\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m    948\u001b[0m     \u001b[0;32melif\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0m_stateful_fn\u001b[0m \u001b[0;32mis\u001b[0m \u001b[0;32mnot\u001b[0m \u001b[0;32mNone\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    949\u001b[0m       \u001b[0;31m# Release the lock early so that multiple threads can perform the call\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
      "\u001b[0;32m~/anaconda3/lib/python3.9/site-packages/tensorflow/python/eager/function.py\u001b[0m in \u001b[0;36m__call__\u001b[0;34m(self, *args, **kwargs)\u001b[0m\n\u001b[1;32m   2494\u001b[0m       (graph_function,\n\u001b[1;32m   2495\u001b[0m        filtered_flat_args) = self._maybe_define_function(args, kwargs)\n\u001b[0;32m-> 2496\u001b[0;31m     return graph_function._call_flat(\n\u001b[0m\u001b[1;32m   2497\u001b[0m         filtered_flat_args, captured_inputs=graph_function.captured_inputs)  # pylint: disable=protected-access\n\u001b[1;32m   2498\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n",
      "\u001b[0;32m~/anaconda3/lib/python3.9/site-packages/tensorflow/python/eager/function.py\u001b[0m in \u001b[0;36m_call_flat\u001b[0;34m(self, args, captured_inputs, cancellation_manager)\u001b[0m\n\u001b[1;32m   1860\u001b[0m         and executing_eagerly):\n\u001b[1;32m   1861\u001b[0m       \u001b[0;31m# No tape is watching; skip to running the function.\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m-> 1862\u001b[0;31m       return self._build_call_outputs(self._inference_function.call(\n\u001b[0m\u001b[1;32m   1863\u001b[0m           ctx, args, cancellation_manager=cancellation_manager))\n\u001b[1;32m   1864\u001b[0m     forward_backward = self._select_forward_and_backward_functions(\n",
      "\u001b[0;32m~/anaconda3/lib/python3.9/site-packages/tensorflow/python/eager/function.py\u001b[0m in \u001b[0;36mcall\u001b[0;34m(self, ctx, args, cancellation_manager)\u001b[0m\n\u001b[1;32m    497\u001b[0m       \u001b[0;32mwith\u001b[0m \u001b[0m_InterpolateFunctionError\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mself\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    498\u001b[0m         \u001b[0;32mif\u001b[0m \u001b[0mcancellation_manager\u001b[0m \u001b[0;32mis\u001b[0m \u001b[0;32mNone\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m--> 499\u001b[0;31m           outputs = execute.execute(\n\u001b[0m\u001b[1;32m    500\u001b[0m               \u001b[0mstr\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0msignature\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mname\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    501\u001b[0m               \u001b[0mnum_outputs\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0m_num_outputs\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
      "\u001b[0;32m~/anaconda3/lib/python3.9/site-packages/tensorflow/python/eager/execute.py\u001b[0m in \u001b[0;36mquick_execute\u001b[0;34m(op_name, num_outputs, inputs, attrs, ctx, name)\u001b[0m\n\u001b[1;32m     52\u001b[0m   \u001b[0;32mtry\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m     53\u001b[0m     \u001b[0mctx\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mensure_initialized\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m---> 54\u001b[0;31m     tensors = pywrap_tfe.TFE_Py_Execute(ctx._handle, device_name, op_name,\n\u001b[0m\u001b[1;32m     55\u001b[0m                                         inputs, attrs, num_outputs)\n\u001b[1;32m     56\u001b[0m   \u001b[0;32mexcept\u001b[0m \u001b[0mcore\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0m_NotOkStatusException\u001b[0m \u001b[0;32mas\u001b[0m \u001b[0me\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
      "\u001b[0;31mKeyboardInterrupt\u001b[0m: "
     ]
    }
   ],
   "source": [
    "model.fit(train, validation_data=test, epochs=100, callbacks=[checkpoint_callback, schedule_callback, example_callback])"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "523b6d33",
   "metadata": {},
   "source": [
    "**since fitting the model requires a GPU for faster epochs, we take the model checkpoint created to make it easier**"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "ad1cfab0",
   "metadata": {},
   "source": [
    "# Make Predictions"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 148,
   "id": "35b58422",
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Downloading...\n",
      "From: https://drive.google.com/uc?id=1vWscXs4Vt0a_1IH1-ct2TCgXAZT-N3_Y\n",
      "To: /home/tvrraviteja/Documents/LipNet/checkpoints.zip\n",
      "100%|██████████████████████████████████████| 94.5M/94.5M [00:27<00:00, 3.45MB/s]\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "['models/checkpoint.index',\n",
       " 'models/__MACOSX/._checkpoint.index',\n",
       " 'models/checkpoint.data-00000-of-00001',\n",
       " 'models/__MACOSX/._checkpoint.data-00000-of-00001',\n",
       " 'models/checkpoint',\n",
       " 'models/__MACOSX/._checkpoint']"
      ]
     },
     "execution_count": 148,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "url = 'https://drive.google.com/uc?id=1vWscXs4Vt0a_1IH1-ct2TCgXAZT-N3_Y'\n",
    "output = 'checkpoints.zip'\n",
    "gdown.download(url, output, quiet=False)\n",
    "gdown.extractall('checkpoints.zip', 'models')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 149,
   "id": "4e7881c2",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<tensorflow.python.checkpoint.checkpoint.CheckpointLoadStatus at 0x7ff52c04ed90>"
      ]
     },
     "execution_count": 149,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "model.load_weights('models/checkpoint')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 151,
   "id": "fcc0ac03",
   "metadata": {},
   "outputs": [],
   "source": [
    "test_data = data.as_numpy_iterator()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 153,
   "id": "03a9eb97",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(array([[[[[ 1.1169125 ],\n",
       "           [ 1.1169125 ],\n",
       "           [ 0.67812544],\n",
       "           ...,\n",
       "           [ 0.31911787],\n",
       "           [ 0.27922812],\n",
       "           [ 0.27922812]],\n",
       " \n",
       "          [[ 1.0371331 ],\n",
       "           [ 1.1568023 ],\n",
       "           [ 0.63823575],\n",
       "           ...,\n",
       "           [ 0.31911787],\n",
       "           [ 0.27922812],\n",
       "           [ 0.27922812]],\n",
       " \n",
       "          [[ 1.5556996 ],\n",
       "           [ 1.4360304 ],\n",
       "           [ 1.0770228 ],\n",
       "           ...,\n",
       "           [ 0.19944867],\n",
       "           [ 0.19944867],\n",
       "           [ 0.15955894]],\n",
       " \n",
       "          ...,\n",
       " \n",
       "          [[ 0.99724334],\n",
       "           [ 0.99724334],\n",
       "           [ 0.99724334],\n",
       "           ...,\n",
       "           [ 0.        ],\n",
       "           [ 0.        ],\n",
       "           [ 0.        ]],\n",
       " \n",
       "          [[ 0.99724334],\n",
       "           [ 0.99724334],\n",
       "           [ 0.99724334],\n",
       "           ...,\n",
       "           [ 0.        ],\n",
       "           [ 0.        ],\n",
       "           [ 0.        ]],\n",
       " \n",
       "          [[ 0.99724334],\n",
       "           [ 0.99724334],\n",
       "           [ 0.99724334],\n",
       "           ...,\n",
       "           [ 0.        ],\n",
       "           [ 0.        ],\n",
       "           [10.171882  ]]],\n",
       " \n",
       " \n",
       "         [[[ 1.1568023 ],\n",
       "           [ 1.1568023 ],\n",
       "           [ 0.67812544],\n",
       "           ...,\n",
       "           [ 0.19944867],\n",
       "           [ 0.15955894],\n",
       "           [ 0.15955894]],\n",
       " \n",
       "          [[ 1.196692  ],\n",
       "           [ 1.0770228 ],\n",
       "           [ 0.67812544],\n",
       "           ...,\n",
       "           [ 0.19944867],\n",
       "           [ 0.15955894],\n",
       "           [ 0.15955894]],\n",
       " \n",
       "          [[ 1.4759201 ],\n",
       "           [ 1.3562509 ],\n",
       "           [ 1.0371331 ],\n",
       "           ...,\n",
       "           [ 0.19944867],\n",
       "           [ 0.19944867],\n",
       "           [ 0.19944867]],\n",
       " \n",
       "          ...,\n",
       " \n",
       "          [[ 1.0770228 ],\n",
       "           [ 1.0770228 ],\n",
       "           [ 1.0371331 ],\n",
       "           ...,\n",
       "           [10.171882  ],\n",
       "           [10.171882  ],\n",
       "           [10.171882  ]],\n",
       " \n",
       "          [[ 1.0770228 ],\n",
       "           [ 1.0770228 ],\n",
       "           [ 1.0371331 ],\n",
       "           ...,\n",
       "           [10.171882  ],\n",
       "           [10.171882  ],\n",
       "           [10.131992  ]],\n",
       " \n",
       "          [[ 1.0770228 ],\n",
       "           [ 1.0770228 ],\n",
       "           [ 1.0371331 ],\n",
       "           ...,\n",
       "           [10.171882  ],\n",
       "           [10.131992  ],\n",
       "           [10.131992  ]]],\n",
       " \n",
       " \n",
       "         [[[ 1.2365817 ],\n",
       "           [ 1.2764715 ],\n",
       "           [ 0.7180152 ],\n",
       "           ...,\n",
       "           [ 0.27922812],\n",
       "           [ 0.27922812],\n",
       "           [ 0.27922812]],\n",
       " \n",
       "          [[ 1.1169125 ],\n",
       "           [ 1.196692  ],\n",
       "           [ 0.7180152 ],\n",
       "           ...,\n",
       "           [ 0.19944867],\n",
       "           [ 0.19944867],\n",
       "           [ 0.19944867]],\n",
       " \n",
       "          [[ 1.3961407 ],\n",
       "           [ 1.3163612 ],\n",
       "           [ 1.0770228 ],\n",
       "           ...,\n",
       "           [ 0.27922812],\n",
       "           [ 0.19944867],\n",
       "           [ 0.19944867]],\n",
       " \n",
       "          ...,\n",
       " \n",
       "          [[ 1.0770228 ],\n",
       "           [ 1.0770228 ],\n",
       "           [ 1.0371331 ],\n",
       "           ...,\n",
       "           [ 0.        ],\n",
       "           [ 0.        ],\n",
       "           [ 0.        ]],\n",
       " \n",
       "          [[ 1.0770228 ],\n",
       "           [ 1.0770228 ],\n",
       "           [ 1.0371331 ],\n",
       "           ...,\n",
       "           [ 0.        ],\n",
       "           [ 0.        ],\n",
       "           [10.171882  ]],\n",
       " \n",
       "          [[ 1.0371331 ],\n",
       "           [ 1.0371331 ],\n",
       "           [ 0.99724334],\n",
       "           ...,\n",
       "           [ 0.        ],\n",
       "           [10.171882  ],\n",
       "           [10.171882  ]]],\n",
       " \n",
       " \n",
       "         ...,\n",
       " \n",
       " \n",
       "         [[[ 1.4759201 ],\n",
       "           [ 1.3961407 ],\n",
       "           [ 1.1568023 ],\n",
       "           ...,\n",
       "           [ 0.2393384 ],\n",
       "           [ 0.27922812],\n",
       "           [ 0.27922812]],\n",
       " \n",
       "          [[ 1.4759201 ],\n",
       "           [ 1.3163612 ],\n",
       "           [ 1.0770228 ],\n",
       "           ...,\n",
       "           [ 0.2393384 ],\n",
       "           [ 0.27922812],\n",
       "           [ 0.27922812]],\n",
       " \n",
       "          [[ 1.3961407 ],\n",
       "           [ 1.3961407 ],\n",
       "           [ 1.2764715 ],\n",
       "           ...,\n",
       "           [ 0.27922812],\n",
       "           [ 0.27922812],\n",
       "           [ 0.27922812]],\n",
       " \n",
       "          ...,\n",
       " \n",
       "          [[ 1.0371331 ],\n",
       "           [ 1.0371331 ],\n",
       "           [ 1.0371331 ],\n",
       "           ...,\n",
       "           [10.131992  ],\n",
       "           [10.131992  ],\n",
       "           [10.131992  ]],\n",
       " \n",
       "          [[ 1.0371331 ],\n",
       "           [ 1.0371331 ],\n",
       "           [ 1.0371331 ],\n",
       "           ...,\n",
       "           [10.131992  ],\n",
       "           [10.131992  ],\n",
       "           [10.131992  ]],\n",
       " \n",
       "          [[ 1.0371331 ],\n",
       "           [ 1.0371331 ],\n",
       "           [ 1.0371331 ],\n",
       "           ...,\n",
       "           [10.131992  ],\n",
       "           [10.131992  ],\n",
       "           [10.131992  ]]],\n",
       " \n",
       " \n",
       "         [[[ 1.4360304 ],\n",
       "           [ 1.3961407 ],\n",
       "           [ 1.2764715 ],\n",
       "           ...,\n",
       "           [ 0.2393384 ],\n",
       "           [ 0.27922812],\n",
       "           [ 0.27922812]],\n",
       " \n",
       "          [[ 1.4759201 ],\n",
       "           [ 1.3163612 ],\n",
       "           [ 0.99724334],\n",
       "           ...,\n",
       "           [ 0.2393384 ],\n",
       "           [ 0.27922812],\n",
       "           [ 0.27922812]],\n",
       " \n",
       "          [[ 1.3961407 ],\n",
       "           [ 1.3961407 ],\n",
       "           [ 1.3562509 ],\n",
       "           ...,\n",
       "           [ 0.31911787],\n",
       "           [ 0.27922812],\n",
       "           [ 0.27922812]],\n",
       " \n",
       "          ...,\n",
       " \n",
       "          [[ 1.0770228 ],\n",
       "           [ 1.0770228 ],\n",
       "           [ 1.0770228 ],\n",
       "           ...,\n",
       "           [10.131992  ],\n",
       "           [10.131992  ],\n",
       "           [10.131992  ]],\n",
       " \n",
       "          [[ 1.0371331 ],\n",
       "           [ 1.0371331 ],\n",
       "           [ 1.0371331 ],\n",
       "           ...,\n",
       "           [10.131992  ],\n",
       "           [10.131992  ],\n",
       "           [10.131992  ]],\n",
       " \n",
       "          [[ 1.0371331 ],\n",
       "           [ 1.0371331 ],\n",
       "           [ 1.0371331 ],\n",
       "           ...,\n",
       "           [10.131992  ],\n",
       "           [10.131992  ],\n",
       "           [10.131992  ]]],\n",
       " \n",
       " \n",
       "         [[[ 1.3562509 ],\n",
       "           [ 1.4759201 ],\n",
       "           [ 1.1568023 ],\n",
       "           ...,\n",
       "           [ 0.2393384 ],\n",
       "           [ 0.27922812],\n",
       "           [ 0.27922812]],\n",
       " \n",
       "          [[ 1.4360304 ],\n",
       "           [ 1.3163612 ],\n",
       "           [ 0.99724334],\n",
       "           ...,\n",
       "           [ 0.2393384 ],\n",
       "           [ 0.27922812],\n",
       "           [ 0.27922812]],\n",
       " \n",
       "          [[ 1.3961407 ],\n",
       "           [ 1.3961407 ],\n",
       "           [ 1.3562509 ],\n",
       "           ...,\n",
       "           [ 0.27922812],\n",
       "           [ 0.19944867],\n",
       "           [ 0.19944867]],\n",
       " \n",
       "          ...,\n",
       " \n",
       "          [[ 1.1169125 ],\n",
       "           [ 1.1169125 ],\n",
       "           [ 1.0770228 ],\n",
       "           ...,\n",
       "           [10.171882  ],\n",
       "           [10.131992  ],\n",
       "           [10.131992  ]],\n",
       " \n",
       "          [[ 1.0770228 ],\n",
       "           [ 1.0770228 ],\n",
       "           [ 1.0371331 ],\n",
       "           ...,\n",
       "           [10.171882  ],\n",
       "           [10.131992  ],\n",
       "           [10.131992  ]],\n",
       " \n",
       "          [[ 1.0770228 ],\n",
       "           [ 1.0770228 ],\n",
       "           [ 1.0371331 ],\n",
       "           ...,\n",
       "           [10.171882  ],\n",
       "           [10.131992  ],\n",
       "           [10.131992  ]]]],\n",
       " \n",
       " \n",
       " \n",
       "        [[[[ 1.4830974 ],\n",
       "           [ 1.4469243 ],\n",
       "           [ 1.5192705 ],\n",
       "           ...,\n",
       "           [ 0.32555798],\n",
       "           [ 0.36173108],\n",
       "           [ 0.36173108]],\n",
       " \n",
       "          [[ 1.5916167 ],\n",
       "           [ 1.3745781 ],\n",
       "           [ 1.5192705 ],\n",
       "           ...,\n",
       "           [ 0.32555798],\n",
       "           [ 0.36173108],\n",
       "           [ 0.36173108]],\n",
       " \n",
       "          [[ 1.4469243 ],\n",
       "           [ 1.4469243 ],\n",
       "           [ 1.4469243 ],\n",
       "           ...,\n",
       "           [ 0.28938487],\n",
       "           [ 0.28938487],\n",
       "           [ 0.28938487]],\n",
       " \n",
       "          ...,\n",
       " \n",
       "          [[ 1.0490202 ],\n",
       "           [ 1.0490202 ],\n",
       "           [ 1.0490202 ],\n",
       "           ...,\n",
       "           [ 0.03617311],\n",
       "           [ 0.03617311],\n",
       "           [ 0.        ]],\n",
       " \n",
       "          [[ 1.0128471 ],\n",
       "           [ 1.0128471 ],\n",
       "           [ 1.0128471 ],\n",
       "           ...,\n",
       "           [ 0.        ],\n",
       "           [ 9.224143  ],\n",
       "           [ 9.224143  ]],\n",
       " \n",
       "          [[ 1.0128471 ],\n",
       "           [ 1.0128471 ],\n",
       "           [ 1.0128471 ],\n",
       "           ...,\n",
       "           [ 9.224143  ],\n",
       "           [ 9.224143  ],\n",
       "           [ 9.151796  ]]],\n",
       " \n",
       " \n",
       "         [[[ 1.5192705 ],\n",
       "           [ 1.5192705 ],\n",
       "           [ 1.5192705 ],\n",
       "           ...,\n",
       "           [ 0.28938487],\n",
       "           [ 0.28938487],\n",
       "           [ 0.28938487]],\n",
       " \n",
       "          [[ 1.5916167 ],\n",
       "           [ 1.5192705 ],\n",
       "           [ 1.4469243 ],\n",
       "           ...,\n",
       "           [ 0.28938487],\n",
       "           [ 0.28938487],\n",
       "           [ 0.28938487]],\n",
       " \n",
       "          [[ 1.4469243 ],\n",
       "           [ 1.4469243 ],\n",
       "           [ 1.4469243 ],\n",
       "           ...,\n",
       "           [ 0.25321177],\n",
       "           [ 0.25321177],\n",
       "           [ 0.25321177]],\n",
       " \n",
       "          ...,\n",
       " \n",
       "          [[ 1.0490202 ],\n",
       "           [ 1.0490202 ],\n",
       "           [ 1.0490202 ],\n",
       "           ...,\n",
       "           [ 0.07234622],\n",
       "           [ 0.07234622],\n",
       "           [ 0.03617311]],\n",
       " \n",
       "          [[ 1.0128471 ],\n",
       "           [ 1.0128471 ],\n",
       "           [ 1.0128471 ],\n",
       "           ...,\n",
       "           [ 0.10851932],\n",
       "           [ 0.07234622],\n",
       "           [ 0.07234622]],\n",
       " \n",
       "          [[ 1.0128471 ],\n",
       "           [ 1.0128471 ],\n",
       "           [ 1.0128471 ],\n",
       "           ...,\n",
       "           [ 0.07234622],\n",
       "           [ 0.07234622],\n",
       "           [ 0.03617311]]],\n",
       " \n",
       " \n",
       "         [[[ 1.5192705 ],\n",
       "           [ 1.5554436 ],\n",
       "           [ 1.4469243 ],\n",
       "           ...,\n",
       "           [ 0.28938487],\n",
       "           [ 0.28938487],\n",
       "           [ 0.28938487]],\n",
       " \n",
       "          [[ 1.5554436 ],\n",
       "           [ 1.5192705 ],\n",
       "           [ 1.3745781 ],\n",
       "           ...,\n",
       "           [ 0.28938487],\n",
       "           [ 0.28938487],\n",
       "           [ 0.28938487]],\n",
       " \n",
       "          [[ 1.4830974 ],\n",
       "           [ 1.4830974 ],\n",
       "           [ 1.4469243 ],\n",
       "           ...,\n",
       "           [ 0.25321177],\n",
       "           [ 0.25321177],\n",
       "           [ 0.25321177]],\n",
       " \n",
       "          ...,\n",
       " \n",
       "          [[ 1.0128471 ],\n",
       "           [ 1.0128471 ],\n",
       "           [ 1.0128471 ],\n",
       "           ...,\n",
       "           [ 0.03617311],\n",
       "           [ 0.03617311],\n",
       "           [ 0.        ]],\n",
       " \n",
       "          [[ 0.9766739 ],\n",
       "           [ 0.9766739 ],\n",
       "           [ 0.9766739 ],\n",
       "           ...,\n",
       "           [ 0.07234622],\n",
       "           [ 0.07234622],\n",
       "           [ 0.07234622]],\n",
       " \n",
       "          [[ 0.9766739 ],\n",
       "           [ 0.9766739 ],\n",
       "           [ 0.9766739 ],\n",
       "           ...,\n",
       "           [ 0.        ],\n",
       "           [ 0.03617311],\n",
       "           [ 9.224143  ]]],\n",
       " \n",
       " \n",
       "         ...,\n",
       " \n",
       " \n",
       "         [[[ 1.3745781 ],\n",
       "           [ 1.3745781 ],\n",
       "           [ 1.338405  ],\n",
       "           ...,\n",
       "           [ 0.32555798],\n",
       "           [ 0.32555798],\n",
       "           [ 0.32555798]],\n",
       " \n",
       "          [[ 1.338405  ],\n",
       "           [ 1.4107512 ],\n",
       "           [ 1.2660588 ],\n",
       "           ...,\n",
       "           [ 0.32555798],\n",
       "           [ 0.32555798],\n",
       "           [ 0.32555798]],\n",
       " \n",
       "          [[ 1.4107512 ],\n",
       "           [ 1.3745781 ],\n",
       "           [ 1.338405  ],\n",
       "           ...,\n",
       "           [ 0.32555798],\n",
       "           [ 0.36173108],\n",
       "           [ 0.36173108]],\n",
       " \n",
       "          ...,\n",
       " \n",
       "          [[ 1.0851933 ],\n",
       "           [ 1.0490202 ],\n",
       "           [ 1.0128471 ],\n",
       "           ...,\n",
       "           [ 9.224143  ],\n",
       "           [ 9.224143  ],\n",
       "           [ 9.224143  ]],\n",
       " \n",
       "          [[ 1.0490202 ],\n",
       "           [ 1.0490202 ],\n",
       "           [ 1.0128471 ],\n",
       "           ...,\n",
       "           [ 0.        ],\n",
       "           [ 0.        ],\n",
       "           [ 0.        ]],\n",
       " \n",
       "          [[ 1.0490202 ],\n",
       "           [ 1.0490202 ],\n",
       "           [ 1.0128471 ],\n",
       "           ...,\n",
       "           [ 0.        ],\n",
       "           [ 0.        ],\n",
       "           [ 9.187969  ]]],\n",
       " \n",
       " \n",
       "         [[[ 1.3745781 ],\n",
       "           [ 1.4830974 ],\n",
       "           [ 1.4830974 ],\n",
       "           ...,\n",
       "           [ 0.28938487],\n",
       "           [ 0.32555798],\n",
       "           [ 0.32555798]],\n",
       " \n",
       "          [[ 1.4107512 ],\n",
       "           [ 1.4830974 ],\n",
       "           [ 1.3745781 ],\n",
       "           ...,\n",
       "           [ 0.28938487],\n",
       "           [ 0.32555798],\n",
       "           [ 0.32555798]],\n",
       " \n",
       "          [[ 1.4830974 ],\n",
       "           [ 1.4107512 ],\n",
       "           [ 1.4107512 ],\n",
       "           ...,\n",
       "           [ 0.32555798],\n",
       "           [ 0.32555798],\n",
       "           [ 0.32555798]],\n",
       " \n",
       "          ...,\n",
       " \n",
       "          [[ 1.0851933 ],\n",
       "           [ 1.0490202 ],\n",
       "           [ 1.0128471 ],\n",
       "           ...,\n",
       "           [ 0.03617311],\n",
       "           [ 0.        ],\n",
       "           [ 0.        ]],\n",
       " \n",
       "          [[ 1.0490202 ],\n",
       "           [ 1.0490202 ],\n",
       "           [ 1.0128471 ],\n",
       "           ...,\n",
       "           [ 0.        ],\n",
       "           [ 0.        ],\n",
       "           [ 0.        ]],\n",
       " \n",
       "          [[ 1.0490202 ],\n",
       "           [ 1.0490202 ],\n",
       "           [ 1.0128471 ],\n",
       "           ...,\n",
       "           [ 0.        ],\n",
       "           [ 0.        ],\n",
       "           [ 0.        ]]],\n",
       " \n",
       " \n",
       "         [[[ 1.4830974 ],\n",
       "           [ 1.5192705 ],\n",
       "           [ 1.5554436 ],\n",
       "           ...,\n",
       "           [ 0.28938487],\n",
       "           [ 0.32555798],\n",
       "           [ 0.32555798]],\n",
       " \n",
       "          [[ 1.5554436 ],\n",
       "           [ 1.5192705 ],\n",
       "           [ 1.4830974 ],\n",
       "           ...,\n",
       "           [ 0.28938487],\n",
       "           [ 0.32555798],\n",
       "           [ 0.32555798]],\n",
       " \n",
       "          [[ 1.5192705 ],\n",
       "           [ 1.4830974 ],\n",
       "           [ 1.3745781 ],\n",
       "           ...,\n",
       "           [ 0.28938487],\n",
       "           [ 0.25321177],\n",
       "           [ 0.25321177]],\n",
       " \n",
       "          ...,\n",
       " \n",
       "          [[ 1.0851933 ],\n",
       "           [ 1.0490202 ],\n",
       "           [ 1.0128471 ],\n",
       "           ...,\n",
       "           [ 0.03617311],\n",
       "           [ 0.        ],\n",
       "           [ 0.        ]],\n",
       " \n",
       "          [[ 1.0490202 ],\n",
       "           [ 1.0490202 ],\n",
       "           [ 1.0128471 ],\n",
       "           ...,\n",
       "           [ 0.        ],\n",
       "           [ 0.        ],\n",
       "           [ 0.        ]],\n",
       " \n",
       "          [[ 1.0490202 ],\n",
       "           [ 1.0490202 ],\n",
       "           [ 1.0128471 ],\n",
       "           ...,\n",
       "           [ 0.        ],\n",
       "           [ 0.        ],\n",
       "           [ 0.        ]]]]], dtype=float32),\n",
       " array([[19,  5, 20, 39,  7, 18,  5,  5, 14, 39, 23,  9, 20,  8, 39, 10,\n",
       "         39,  6,  9, 22,  5, 39, 19, 15, 15, 14,  0,  0,  0,  0,  0,  0,\n",
       "          0,  0,  0,  0,  0,  0,  0,  0],\n",
       "        [ 2,  9, 14, 39,  7, 18,  5,  5, 14, 39,  2, 25, 39,  2, 39, 15,\n",
       "         14,  5, 39, 19, 15, 15, 14,  0,  0,  0,  0,  0,  0,  0,  0,  0,\n",
       "          0,  0,  0,  0,  0,  0,  0,  0]]))"
      ]
     },
     "execution_count": 153,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "test_data.next()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 152,
   "id": "137d8a00",
   "metadata": {},
   "outputs": [],
   "source": [
    "sample = test_data.next()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 154,
   "id": "de0ba034",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "1/1 [==============================] - 4s 4s/step\n"
     ]
    }
   ],
   "source": [
    "yhat = model.predict(sample[0])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 155,
   "id": "ffaaaf36",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ REAL TEXT\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "[<tf.Tensor: shape=(), dtype=string, numpy=b'bin green at u zero please'>,\n",
       " <tf.Tensor: shape=(), dtype=string, numpy=b'lay blue by y four please'>]"
      ]
     },
     "execution_count": 155,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "print('~'*100, 'REAL TEXT')\n",
    "[tf.strings.reduce_join([num_to_char(word) for word in sentence]) for sentence in sample[1]]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 156,
   "id": "377731e7",
   "metadata": {},
   "outputs": [],
   "source": [
    "decoded = tf.keras.backend.ctc_decode(yhat, input_length=[75,75], greedy=True)[0][0].numpy()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 157,
   "id": "352768e0",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ PREDICTIONS\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "[<tf.Tensor: shape=(), dtype=string, numpy=b'bin green at u zero please'>,\n",
       " <tf.Tensor: shape=(), dtype=string, numpy=b'lay blue by y four please'>]"
      ]
     },
     "execution_count": 157,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "print('~'*100, 'PREDICTIONS')\n",
    "[tf.strings.reduce_join([num_to_char(word) for word in sentence]) for sentence in decoded]"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "ec9fd76a",
   "metadata": {},
   "source": [
    "# Test on a Video"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 174,
   "id": "7d420d57",
   "metadata": {},
   "outputs": [],
   "source": [
    "sample = load_data(tf.convert_to_tensor('./data/s1/pgwz1a.mpg'))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 175,
   "id": "47cda32f",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ REAL TEXT\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "[<tf.Tensor: shape=(), dtype=string, numpy=b'place green with z one again'>]"
      ]
     },
     "execution_count": 175,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "print('~'*100, 'REAL TEXT')\n",
    "[tf.strings.reduce_join([num_to_char(word) for word in sentence]) for sentence in [sample[1]]]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 176,
   "id": "e00d7417",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "1/1 [==============================] - 1s 1s/step\n"
     ]
    }
   ],
   "source": [
    "yhat = model.predict(tf.expand_dims(sample[0], axis=0))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 177,
   "id": "d068af3f",
   "metadata": {},
   "outputs": [],
   "source": [
    "decoded = tf.keras.backend.ctc_decode(yhat, input_length=[75], greedy=True)[0][0].numpy()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 178,
   "id": "c57cfcb5",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ PREDICTIONS\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "[<tf.Tensor: shape=(), dtype=string, numpy=b'place green with z one again'>]"
      ]
     },
     "execution_count": 178,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "print('~'*100, 'PREDICTIONS')\n",
    "[tf.strings.reduce_join([num_to_char(word) for word in sentence]) for sentence in decoded]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "7496b03a",
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3 (ipykernel)",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.9.13"
  },
  "vscode": {
   "interpreter": {
    "hash": "716a3d6d7affda053562378f20e81717d1c68608f0888b6f571d3c32ad14bffd"
   }
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}