tensorflow-image-captioning.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/opt/conda/lib/python3.7/site-packages/tqdm/std.py:666: FutureWarning: The Panel class is removed from pandas. Accessing it from the top-level namespace will also be removed in the next version\n",
      "  from pandas import Panel\n"
     ]
    }
   ],
   "source": [
    "import tensorflow as tf\n",
    "import cv2\n",
    "import pandas as pd\n",
    "import numpy as np\n",
    "import os\n",
    "import random\n",
    "import re\n",
    "import time\n",
    "import json\n",
    "import wandb\n",
    "import pickle\n",
    "\n",
    "from matplotlib import pyplot as plt\n",
    "from glob import glob\n",
    "from PIL import Image\n",
    "from tqdm import tqdm\n",
    "\n",
    "# Scikit-learn includes many helpful utilities\n",
    "from sklearn.model_selection import train_test_split\n",
    "from sklearn.utils import shuffle\n",
    "\n",
    "tqdm.pandas()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\u001b[34m\u001b[1mwandb\u001b[0m: \u001b[32m\u001b[41mERROR\u001b[0m Not authenticated.  Copy a key from https://app.wandb.ai/authorize\n",
      "\u001b[34m\u001b[1mwandb\u001b[0m: \u001b[32m\u001b[41mERROR\u001b[0m Can't accept input in this environment, you should set WANDB_API_KEY or call wandb.login(key='YOUR_API_KEY')\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "False"
      ]
     },
     "execution_count": 2,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "wandb.login()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [],
   "source": [
    "df_captions = pd.read_csv('../input/flickr-image-dataset/flickr30k_images/results.csv', delimiter='|')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>image_name</th>\n",
       "      <th>comment_number</th>\n",
       "      <th>comment</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>1000092795.jpg</td>\n",
       "      <td>0</td>\n",
       "      <td>Two young guys with shaggy hair look at their...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>1000092795.jpg</td>\n",
       "      <td>1</td>\n",
       "      <td>Two young , White males are outside near many...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>1000092795.jpg</td>\n",
       "      <td>2</td>\n",
       "      <td>Two men in green shirts are standing in a yard .</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>1000092795.jpg</td>\n",
       "      <td>3</td>\n",
       "      <td>A man in a blue shirt standing in a garden .</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>1000092795.jpg</td>\n",
       "      <td>4</td>\n",
       "      <td>Two friends enjoy time spent together .</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "       image_name  comment_number  \\\n",
       "0  1000092795.jpg               0   \n",
       "1  1000092795.jpg               1   \n",
       "2  1000092795.jpg               2   \n",
       "3  1000092795.jpg               3   \n",
       "4  1000092795.jpg               4   \n",
       "\n",
       "                                             comment  \n",
       "0   Two young guys with shaggy hair look at their...  \n",
       "1   Two young , White males are outside near many...  \n",
       "2   Two men in green shirts are standing in a yard .  \n",
       "3       A man in a blue shirt standing in a garden .  \n",
       "4            Two friends enjoy time spent together .  "
      ]
     },
     "execution_count": 4,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_captions.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [],
   "source": [
    "df_captions.columns = ['image_name', 'comment_number', 'comment']"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 158915/158915 [00:00<00:00, 432722.64it/s]\n"
     ]
    }
   ],
   "source": [
    "df_captions['comment_new'] = df_captions['comment'].progress_apply(lambda x : '<start> ' + str(x) + ' <end>')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 158915/158915 [00:00<00:00, 492079.43it/s]\n"
     ]
    }
   ],
   "source": [
    "path = '../input/flickr-image-dataset/flickr30k_images/flickr30k_images/'\n",
    "df_captions['image_name_new'] = df_captions['image_name'].progress_apply(lambda x : path + x)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>image_name</th>\n",
       "      <th>comment_number</th>\n",
       "      <th>comment</th>\n",
       "      <th>comment_new</th>\n",
       "      <th>image_name_new</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>1000092795.jpg</td>\n",
       "      <td>0</td>\n",
       "      <td>Two young guys with shaggy hair look at their...</td>\n",
       "      <td>&lt;start&gt;  Two young guys with shaggy hair look ...</td>\n",
       "      <td>../input/flickr-image-dataset/flickr30k_images...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>1000092795.jpg</td>\n",
       "      <td>1</td>\n",
       "      <td>Two young , White males are outside near many...</td>\n",
       "      <td>&lt;start&gt;  Two young , White males are outside n...</td>\n",
       "      <td>../input/flickr-image-dataset/flickr30k_images...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>1000092795.jpg</td>\n",
       "      <td>2</td>\n",
       "      <td>Two men in green shirts are standing in a yard .</td>\n",
       "      <td>&lt;start&gt;  Two men in green shirts are standing ...</td>\n",
       "      <td>../input/flickr-image-dataset/flickr30k_images...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>1000092795.jpg</td>\n",
       "      <td>3</td>\n",
       "      <td>A man in a blue shirt standing in a garden .</td>\n",
       "      <td>&lt;start&gt;  A man in a blue shirt standing in a g...</td>\n",
       "      <td>../input/flickr-image-dataset/flickr30k_images...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>1000092795.jpg</td>\n",
       "      <td>4</td>\n",
       "      <td>Two friends enjoy time spent together .</td>\n",
       "      <td>&lt;start&gt;  Two friends enjoy time spent together...</td>\n",
       "      <td>../input/flickr-image-dataset/flickr30k_images...</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "       image_name comment_number  \\\n",
       "0  1000092795.jpg              0   \n",
       "1  1000092795.jpg              1   \n",
       "2  1000092795.jpg              2   \n",
       "3  1000092795.jpg              3   \n",
       "4  1000092795.jpg              4   \n",
       "\n",
       "                                             comment  \\\n",
       "0   Two young guys with shaggy hair look at their...   \n",
       "1   Two young , White males are outside near many...   \n",
       "2   Two men in green shirts are standing in a yard .   \n",
       "3       A man in a blue shirt standing in a garden .   \n",
       "4            Two friends enjoy time spent together .   \n",
       "\n",
       "                                         comment_new  \\\n",
       "0  <start>  Two young guys with shaggy hair look ...   \n",
       "1  <start>  Two young , White males are outside n...   \n",
       "2  <start>  Two men in green shirts are standing ...   \n",
       "3  <start>  A man in a blue shirt standing in a g...   \n",
       "4  <start>  Two friends enjoy time spent together...   \n",
       "\n",
       "                                      image_name_new  \n",
       "0  ../input/flickr-image-dataset/flickr30k_images...  \n",
       "1  ../input/flickr-image-dataset/flickr30k_images...  \n",
       "2  ../input/flickr-image-dataset/flickr30k_images...  \n",
       "3  ../input/flickr-image-dataset/flickr30k_images...  \n",
       "4  ../input/flickr-image-dataset/flickr30k_images...  "
      ]
     },
     "execution_count": 8,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_captions.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {
    "_cell_guid": "79c7e3d0-c299-4dcb-8224-4455121ee9b0",
    "_uuid": "d629ff2d2480ee46fbb7e2d37f6b5fab8052498a"
   },
   "outputs": [],
   "source": [
    "all_img_name_vector =  df_captions['image_name_new'].tolist()\n",
    "all_captions = df_captions['comment_new'].tolist()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Shuffle captions and image_names together\n",
    "# Set a random state\n",
    "train_captions, img_name_vector = shuffle(all_captions,\n",
    "                                          all_img_name_vector,\n",
    "                                          random_state=1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "('<start>  A legless man is walking next to a man in a bright yellow shirt . <end>',\n",
       " '../input/flickr-image-dataset/flickr30k_images/flickr30k_images/6321643151.jpg')"
      ]
     },
     "execution_count": 11,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "train_captions[0],img_name_vector[0]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Select the first 30000 captions from the shuffled set\n",
    "num_examples = 10000\n",
    "train_captions = train_captions[:num_examples]\n",
    "img_name_vector = img_name_vector[:num_examples]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(10000, 158915)"
      ]
     },
     "execution_count": 13,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "len(train_captions), len(all_captions)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Preprocess the images using InceptionV3\n",
    "Next, you will use InceptionV3 (which is pretrained on Imagenet) to classify each image. You will extract features from the last convolutional layer.\n",
    "\n",
    "First, you will convert the images into InceptionV3's expected format by:\n",
    "\n",
    "Resizing the image to 299px by 299px\n",
    "Preprocess the images using the preprocess_input method to normalize the image so that it contains pixels in the range of -1 to 1, which matches the format of the images used to train InceptionV3."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {},
   "outputs": [],
   "source": [
    "def load_image(image_path,cap):\n",
    "    img = tf.io.read_file(image_path)\n",
    "    img = tf.image.decode_jpeg(img, channels=3)\n",
    "    img = tf.image.resize(img, (299, 299))\n",
    "    img = tf.keras.applications.inception_v3.preprocess_input(img)\n",
    "    return img, cap"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "+ Initialize InceptionV3 and load the pretrained Imagenet weights\n",
    "\n",
    "+ Now you'll create a tf.keras model where the output layer is the last convolutional layer in the InceptionV3 architecture. The shape of the output of this layer is 8x8x2048. You use the last convolutional layer because you are using attention in this example. \n",
    "\n",
    "+ You don't perform this initialization during training because it could become a bottleneck.\n",
    "\n",
    "+ You forward each image through the network and store the resulting vector in a dictionary (image_name --> feature_vector).\n",
    "\n",
    "+ After all the images are passed through the network, you pickle the dictionary and save it to disk."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Preprocess and tokenize the captions\n",
    "\n",
    "+ First, you'll tokenize the captions (for example, by splitting on spaces). This gives us a vocabulary of all of the unique words in the data (for example, \"surfing\", \"football\", and so on).\n",
    "\n",
    "+ Next, you'll limit the vocabulary size to the top 5,000 words (to save memory). You'll replace all other words with the token \"UNK\" (unknown).\n",
    "\n",
    "+ You then create word-to-index and index-to-word mappings.\n",
    "\n",
    "+ Finally, you pad all sequences to be the same length as the longest one."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Find the maximum length of any caption in our dataset\n",
    "def calc_max_length(tensor):\n",
    "    return max(len(t) for t in tensor)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Choose the top 5000 words from the vocabulary\n",
    "top_k = 10000\n",
    "tokenizer = tf.keras.preprocessing.text.Tokenizer(num_words=top_k,\n",
    "                                                  oov_token=\"<unk>\",\n",
    "                                                  filters='!\"#$%&()*+.,-/:;=?@[\\]^_`{|}~ ')\n",
    "tokenizer.fit_on_texts(train_captions)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {},
   "outputs": [],
   "source": [
    "tokenizer.word_index['<pad>'] = 0\n",
    "tokenizer.index_word[0] = '<pad>'"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Create the tokenized vectors\n",
    "train_seqs = tokenizer.texts_to_sequences(train_captions)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Pad each vector to the max_length of the captions\n",
    "\n",
    "# If you do not provide a max_length value, pad_sequences calculates it automatically\n",
    "cap_vector = tf.keras.preprocessing.sequence.pad_sequences(train_seqs, padding='post')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "62"
      ]
     },
     "execution_count": 20,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Calculates the max_length, which is used to store the attention weights\n",
    "max_length = calc_max_length(train_seqs)\n",
    "max_length"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Split the data into training and testing"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Create training and validation sets using an 80-20 split\n",
    "img_name_train, img_name_val, cap_train, cap_val = train_test_split(img_name_vector,\n",
    "                                                                    cap_vector,\n",
    "                                                                    test_size=0.2,\n",
    "                                                                    random_state=0)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(8000, 8000, 2000, 2000)"
      ]
     },
     "execution_count": 22,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "len(img_name_train), len(cap_train), len(img_name_val), len(cap_val)\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Create a tf.data dataset for training\n",
    "\n",
    "Our images and captions are ready! Next, let's create a tf.data dataset to use for training our model."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Feel free to change these parameters according to your system's configuration\n",
    "\n",
    "BATCH_SIZE = 128\n",
    "BUFFER_SIZE = 1000\n",
    "embedding_dim = 100\n",
    "units = 256\n",
    "vocab_size = top_k + 1\n",
    "num_steps = len(img_name_train) // BATCH_SIZE\n",
    "# Shape of the vector extracted from InceptionV3 is (64, 2048)\n",
    "# These two variables represent that vector shape\n",
    "features_shape = 2048\n",
    "attention_features_shape = 64"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "metadata": {},
   "outputs": [],
   "source": [
    "# # Load the numpy files\n",
    "# def map_func(img_name, cap):\n",
    "#     img_tensor = np.load(img_name.decode('utf-8')+'.npy')\n",
    "#     return img_tensor, cap"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "'../input/flickr-image-dataset/flickr30k_images/flickr30k_images/2511794705.jpg'"
      ]
     },
     "execution_count": 25,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "img_name_train[0]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "metadata": {},
   "outputs": [],
   "source": [
    "#Preparing the tf dataset\n",
    "dataset = tf.data.Dataset.from_tensor_slices((img_name_train, cap_train))\n",
    "\n",
    "# # Use map to load the numpy files in parallel\n",
    "dataset = dataset.map(lambda item1, item2: tf.numpy_function(\n",
    "          load_image, [item1, item2], [tf.float32, tf.int32]),\n",
    "          num_parallel_calls = tf.data.experimental.AUTOTUNE)\n",
    "\n",
    "# Shuffle and batch\n",
    "dataset = dataset.shuffle(BUFFER_SIZE).batch(BATCH_SIZE)\n",
    "dataset = dataset.prefetch(buffer_size=tf.data.experimental.AUTOTUNE)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Model\n",
    "+ The decoder below is identical to the one in the example for Neural Machine Translation with Attention.\n",
    "\n",
    "+ The model architecture is inspired by the Show, Attend and Tell paper.\n",
    "\n",
    "+ In this example, you extract the features from the lower convolutional layer of InceptionV3 giving us a vector of shape (8, 8, 2048).\n",
    "\n",
    "+ You squash that to a shape of (64, 2048).\n",
    "\n",
    "+ This vector is then passed through the CNN Encoder (which consists of a single Fully connected layer).\n",
    "\n",
    "+ The RNN (here GRU) attends over the image to predict the next word."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "metadata": {},
   "outputs": [],
   "source": [
    "class BahdanauAttention(tf.keras.Model):\n",
    "    \n",
    "    \n",
    "    def __init__(self, units):\n",
    "        super(BahdanauAttention, self).__init__()\n",
    "        self.W1 = tf.keras.layers.Dense(units)\n",
    "        self.W2 = tf.keras.layers.Dense(units)\n",
    "        self.V = tf.keras.layers.Dense(1)\n",
    "\n",
    "    def call(self, features, hidden):\n",
    "        # features(CNN_encoder output) shape == (batch_size, 64, embedding_dim)\n",
    "\n",
    "        # hidden shape == (batch_size, hidden_size)\n",
    "        # hidden_with_time_axis shape == (batch_size, 1, hidden_size)\n",
    "        hidden_with_time_axis = tf.expand_dims(hidden, 1)\n",
    "\n",
    "        # score shape == (batch_size, 64, hidden_size)\n",
    "        score = tf.nn.tanh(self.W1(features) + self.W2(hidden_with_time_axis))\n",
    "\n",
    "        # attention_weights shape == (batch_size, 64, 1)\n",
    "        # you get 1 at the last axis because you are applying score to self.V\n",
    "        attention_weights = tf.nn.softmax(self.V(score), axis=1)\n",
    "\n",
    "        # context_vector shape after sum == (batch_size, hidden_size)\n",
    "        context_vector = attention_weights * features\n",
    "        context_vector = tf.reduce_sum(context_vector, axis=1)\n",
    "\n",
    "        return context_vector, attention_weights"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Downloading data from https://github.com/fchollet/deep-learning-models/releases/download/v0.5/inception_v3_weights_tf_dim_ordering_tf_kernels_notop.h5\n",
      "87916544/87910968 [==============================] - 2s 0us/step\n"
     ]
    }
   ],
   "source": [
    "image_model = tf.keras.applications.InceptionV3(include_top=False, weights='imagenet')\n",
    "new_input = image_model.input\n",
    "hidden_layer = image_model.layers[-1].output"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Encoder Network"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "metadata": {},
   "outputs": [],
   "source": [
    "class CNN_Encoder(tf.keras.Model):\n",
    "    \n",
    "    # Since you have already extracted the features and dumped it using pickle\n",
    "    # This encoder passes those features through a Fully connected layer\n",
    "    def __init__(self, embedding_dim):\n",
    "        super(CNN_Encoder, self).__init__()\n",
    "        self.image_features_extract_model = tf.keras.Model(new_input, hidden_layer)\n",
    "        \n",
    "        # shape after fc == (batch_size, 64, embedding_dim)\n",
    "        self.fc = tf.keras.layers.Dense(embedding_dim)\n",
    "\n",
    "    def call(self, x):\n",
    "        x = self.image_features_extract_model(x)    #shape batch_sz * 8 * 8 * 2048\n",
    "        x = tf.reshape(x, (x.shape[0], -1, x.shape[3]))  #shape batch_sz * 64 * 2048\n",
    "        x = self.fc(x)     ##shape batch_sz * 64 * embedding_dim\n",
    "        x = tf.nn.relu(x)  ##shape batch_sz * 64 * embedding_dim with relu\n",
    "        return x"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Decoder Network"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 30,
   "metadata": {},
   "outputs": [],
   "source": [
    "class RNN_Decoder(tf.keras.Model):  \n",
    "    \n",
    "    def __init__(self, embedding_dim, units, vocab_size):\n",
    "        super(RNN_Decoder, self).__init__()\n",
    "        self.units = units     \n",
    "\n",
    "        self.embedding = tf.keras.layers.Embedding(vocab_size, embedding_dim)\n",
    "        self.gru = tf.keras.layers.GRU(self.units,\n",
    "                                       return_sequences=True,\n",
    "                                       return_state=True,\n",
    "                                       recurrent_initializer='glorot_uniform')\n",
    "        self.fc1 = tf.keras.layers.Dense(self.units)\n",
    "        self.fc2 = tf.keras.layers.Dense(vocab_size)\n",
    "\n",
    "        self.attention = BahdanauAttention(self.units)\n",
    "\n",
    "        \n",
    "\n",
    "    def call(self, x, features, hidden):\n",
    "\n",
    "        # defining attention as a separate model\n",
    "        context_vector, attention_weights = self.attention(features, hidden)\n",
    "\n",
    "        # x shape after passing through embedding == (batch_size, 1, embedding_dim)\n",
    "        x = self.embedding(x)\n",
    "\n",
    "        # x shape after concatenation == (batch_size, 1, embedding_dim + hidden_size)\n",
    "        x = tf.concat([tf.expand_dims(context_vector, 1), x], axis=-1)\n",
    "\n",
    "        # passing the concatenated vector to the GRU\n",
    "        output, state = self.gru(x)\n",
    "\n",
    "        # shape == (batch_size, max_length, hidden_size)\n",
    "        x = self.fc1(output)\n",
    "\n",
    "        # x shape == (batch_size * max_length, hidden_size)\n",
    "        x = tf.reshape(x, (-1, x.shape[2]))\n",
    "\n",
    "        # output shape == (batch_size * max_length, vocab)\n",
    "        x = self.fc2(x)\n",
    "\n",
    "        return x, state, attention_weights\n",
    "\n",
    "    \n",
    "    def reset_state(self, batch_size):\n",
    "        return tf.zeros((batch_size, self.units))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 31,
   "metadata": {},
   "outputs": [],
   "source": [
    "encoder = CNN_Encoder(embedding_dim)\n",
    "decoder = RNN_Decoder(embedding_dim, units, vocab_size)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 32,
   "metadata": {},
   "outputs": [],
   "source": [
    "#Optimizer\n",
    "optimizer = tf.keras.optimizers.Adam()\n",
    "\n",
    "loss_object = tf.keras.losses.SparseCategoricalCrossentropy(from_logits=True, reduction='none')\n",
    "\n",
    "\n",
    "def loss_function(real, pred):\n",
    "    mask = tf.math.logical_not(tf.math.equal(real, 0))\n",
    "    loss_ = loss_object(real, pred)\n",
    "\n",
    "    mask = tf.cast(mask, dtype=loss_.dtype)\n",
    "    loss_ *= mask\n",
    "\n",
    "    return tf.reduce_mean(loss_)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Checkpoint"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 33,
   "metadata": {},
   "outputs": [],
   "source": [
    "checkpoint_path = \"/kaggle/working/checkpoints/train\"\n",
    "ckpt = tf.train.Checkpoint(encoder=encoder,\n",
    "                           decoder=decoder,\n",
    "                           optimizer = optimizer)\n",
    "ckpt_manager = tf.train.CheckpointManager(ckpt, checkpoint_path, max_to_keep=5)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 34,
   "metadata": {},
   "outputs": [],
   "source": [
    "start_epoch = 0\n",
    "\n",
    "if ckpt_manager.latest_checkpoint:\n",
    "    start_epoch = int(ckpt_manager.latest_checkpoint.split('-')[-1])\n",
    "    \n",
    "    # restoring the latest checkpoint in checkpoint_path\n",
    "    ckpt.restore(ckpt_manager.latest_checkpoint)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Training\n",
    "\n",
    "+ You extract the features by passing the images through the encoder.\n",
    "\n",
    "+ The encoder output, hidden state(initialized to 0) and the decoder input (which is the start token) is passed to the decoder.\n",
    "\n",
    "+ The decoder returns the predictions and the decoder hidden state.\n",
    "\n",
    "+ The decoder hidden state is then passed back into the model and the predictions are used to calculate the loss.\n",
    "\n",
    "+ Use teacher forcing to decide the next input to the decoder.\n",
    "\n",
    "+ Teacher forcing is the technique where the target word is passed as the next input to the decoder.\n",
    "\n",
    "+ The final step is to calculate the gradients and apply it to the optimizer and backpropagate."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 35,
   "metadata": {},
   "outputs": [],
   "source": [
    "# adding this in a separate cell because if you run the training cell\n",
    "# many times, the loss_plot array will be reset\n",
    "loss_plot = []"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 36,
   "metadata": {},
   "outputs": [],
   "source": [
    "@tf.function\n",
    "def train_step(img_tensor, target):\n",
    "    loss = 0\n",
    "\n",
    "    # initializing the hidden state for each batch\n",
    "    # because the captions are not related from image to image\n",
    "    hidden = decoder.reset_state(batch_size=target.shape[0])\n",
    "\n",
    "    dec_input = tf.expand_dims([tokenizer.word_index['<start>']] * target.shape[0], 1)\n",
    "\n",
    "    with tf.GradientTape() as tape:\n",
    "        \n",
    "        features = encoder(img_tensor)\n",
    "        \n",
    "        for i in range(1, target.shape[1]):\n",
    "            \n",
    "            # passing the features through the decoder\n",
    "            predictions, hidden, _ = decoder(dec_input, features, hidden)\n",
    "            \n",
    "            loss += loss_function(target[:, i], predictions)\n",
    "            \n",
    "            # using teacher forcing\n",
    "            dec_input = tf.expand_dims(target[:, i], 1)\n",
    "\n",
    "    #Total loss for batchi\n",
    "    total_loss = (loss / int(target.shape[1]))  #loss for batch_i per word\n",
    "\n",
    "\n",
    "    #trainable_variables ????? -> Getting all th weights for the encoder and decoder for gradient descent\n",
    "    trainable_variables = encoder.trainable_variables + decoder.trainable_variables\n",
    "    \n",
    "    #Performing the gradient descent\n",
    "    gradients = tape.gradient(loss, trainable_variables)\n",
    "    \n",
    "    #applying the optimization\n",
    "    optimizer.apply_gradients(zip(gradients, trainable_variables))\n",
    "\n",
    "    return loss, total_loss"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 37,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\u001b[34m\u001b[1mwandb\u001b[0m: \u001b[32m\u001b[41mERROR\u001b[0m Not authenticated.  Copy a key from https://app.wandb.ai/authorize\n",
      "\u001b[34m\u001b[1mwandb\u001b[0m: \u001b[32m\u001b[41mERROR\u001b[0m Can't accept input in this environment, you should set WANDB_API_KEY or call wandb.login(key='YOUR_API_KEY')\n"
     ]
    },
    {
     "data": {
      "text/html": [
       "\n",
       "                <b>Could not authenticate.</b><br/>\n",
       "            "
      ],
      "text/plain": [
       "<IPython.core.display.HTML object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "\n",
       "            Using <a href=\"https://wandb.com\" target=\"_blank\">Weights & Biases</a> in dryrun mode. Not logging results to the cloud.<br/>\n",
       "            Call wandb.login() to authenticate this machine.<br/>\n",
       "        "
      ],
      "text/plain": [
       "<IPython.core.display.HTML object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\u001b[34m\u001b[1mwandb\u001b[0m: Wandb version 0.9.1 is available!  To upgrade, please run:\n",
      "\u001b[34m\u001b[1mwandb\u001b[0m:  $ pip install wandb --upgrade\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "W&B Error: Can't find API key, run wandb login or set WANDB_API_KEY"
      ]
     },
     "execution_count": 37,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "### id is for run name\n",
    "wandb.init(project=\"image-captioning\", id=\"initial-run\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 38,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 1 Batch 0 Loss 1.9233\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\u001b[34m\u001b[1mwandb\u001b[0m: Wandb version 0.9.1 is available!  To upgrade, please run:\n",
      "\u001b[34m\u001b[1mwandb\u001b[0m:  $ pip install wandb --upgrade\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 1 Loss 1.350736\n",
      "Time taken for 1 epoch 272.01242208480835 sec\n",
      "\n",
      "Epoch 2 Batch 0 Loss 1.2290\n",
      "Epoch 2 Loss 1.231870\n",
      "Time taken for 1 epoch 110.4753987789154 sec\n",
      "\n",
      "Epoch 3 Batch 0 Loss 1.2040\n",
      "Epoch 3 Loss 1.229931\n",
      "Time taken for 1 epoch 111.61562943458557 sec\n",
      "\n",
      "Epoch 4 Batch 0 Loss 1.2644\n",
      "Epoch 4 Loss 1.229338\n",
      "Time taken for 1 epoch 110.47138857841492 sec\n",
      "\n",
      "Epoch 5 Batch 0 Loss 1.2954\n",
      "Epoch 5 Loss 1.227331\n",
      "Time taken for 1 epoch 110.67935037612915 sec\n",
      "\n",
      "Epoch 6 Batch 0 Loss 1.2302\n",
      "Epoch 6 Loss 1.226623\n",
      "Time taken for 1 epoch 111.2239294052124 sec\n",
      "\n",
      "Epoch 7 Batch 0 Loss 1.1759\n",
      "Epoch 7 Loss 1.225484\n",
      "Time taken for 1 epoch 110.1174669265747 sec\n",
      "\n",
      "Epoch 8 Batch 0 Loss 1.2342\n",
      "Epoch 8 Loss 1.226100\n",
      "Time taken for 1 epoch 109.49009132385254 sec\n",
      "\n",
      "Epoch 9 Batch 0 Loss 1.2136\n",
      "Epoch 9 Loss 1.225724\n",
      "Time taken for 1 epoch 110.77115035057068 sec\n",
      "\n",
      "Epoch 10 Batch 0 Loss 1.2571\n",
      "Epoch 10 Loss 1.224396\n",
      "Time taken for 1 epoch 110.42040777206421 sec\n",
      "\n",
      "Epoch 11 Batch 0 Loss 1.2848\n",
      "Epoch 11 Loss 1.225491\n",
      "Time taken for 1 epoch 112.33957767486572 sec\n",
      "\n",
      "Epoch 12 Batch 0 Loss 1.2085\n",
      "Epoch 12 Loss 1.224564\n",
      "Time taken for 1 epoch 113.59766006469727 sec\n",
      "\n",
      "Epoch 13 Batch 0 Loss 1.2501\n",
      "Epoch 13 Loss 1.224405\n",
      "Time taken for 1 epoch 111.4470591545105 sec\n",
      "\n",
      "Epoch 14 Batch 0 Loss 1.2086\n",
      "Epoch 14 Loss 1.224868\n",
      "Time taken for 1 epoch 111.36971712112427 sec\n",
      "\n",
      "Epoch 15 Batch 0 Loss 1.2545\n",
      "Epoch 15 Loss 1.223539\n",
      "Time taken for 1 epoch 110.60456347465515 sec\n",
      "\n",
      "Epoch 16 Batch 0 Loss 1.2447\n",
      "Epoch 16 Loss 1.224191\n",
      "Time taken for 1 epoch 111.00417232513428 sec\n",
      "\n",
      "Epoch 17 Batch 0 Loss 1.2269\n",
      "Epoch 17 Loss 1.223083\n",
      "Time taken for 1 epoch 110.20814990997314 sec\n",
      "\n",
      "Epoch 18 Batch 0 Loss 1.2309\n",
      "Epoch 18 Loss 1.224400\n",
      "Time taken for 1 epoch 110.50287246704102 sec\n",
      "\n",
      "Epoch 19 Batch 0 Loss 1.2306\n",
      "Epoch 19 Loss 1.224800\n",
      "Time taken for 1 epoch 110.2268614768982 sec\n",
      "\n",
      "Epoch 20 Batch 0 Loss 1.2836\n",
      "Epoch 20 Loss 1.224016\n",
      "Time taken for 1 epoch 110.76518869400024 sec\n",
      "\n",
      "Epoch 21 Batch 0 Loss 1.2028\n",
      "Epoch 21 Loss 1.223463\n",
      "Time taken for 1 epoch 112.6316339969635 sec\n",
      "\n",
      "Epoch 22 Batch 0 Loss 1.3002\n",
      "Epoch 22 Loss 1.222823\n",
      "Time taken for 1 epoch 111.62313795089722 sec\n",
      "\n",
      "Epoch 23 Batch 0 Loss 1.2750\n",
      "Epoch 23 Loss 1.223393\n",
      "Time taken for 1 epoch 111.4014220237732 sec\n",
      "\n",
      "Epoch 24 Batch 0 Loss 1.1865\n",
      "Epoch 24 Loss 1.223527\n",
      "Time taken for 1 epoch 114.42125463485718 sec\n",
      "\n",
      "Epoch 25 Batch 0 Loss 1.2362\n",
      "Epoch 25 Loss 1.223206\n",
      "Time taken for 1 epoch 111.32345604896545 sec\n",
      "\n",
      "Epoch 26 Batch 0 Loss 1.2385\n",
      "Epoch 26 Loss 1.222319\n",
      "Time taken for 1 epoch 113.09670639038086 sec\n",
      "\n",
      "Epoch 27 Batch 0 Loss 1.2666\n",
      "Epoch 27 Loss 1.223910\n",
      "Time taken for 1 epoch 111.94857025146484 sec\n",
      "\n",
      "Epoch 28 Batch 0 Loss 1.1125\n",
      "Epoch 28 Loss 1.224657\n",
      "Time taken for 1 epoch 111.04509663581848 sec\n",
      "\n",
      "Epoch 29 Batch 0 Loss 1.1756\n",
      "Epoch 29 Loss 1.223562\n",
      "Time taken for 1 epoch 110.48151445388794 sec\n",
      "\n",
      "Epoch 30 Batch 0 Loss 1.1773\n",
      "Epoch 30 Loss 1.223339\n",
      "Time taken for 1 epoch 110.98088335990906 sec\n",
      "\n",
      "Epoch 31 Batch 0 Loss 1.2511\n",
      "Epoch 31 Loss 1.222870\n",
      "Time taken for 1 epoch 111.68931031227112 sec\n",
      "\n",
      "Epoch 32 Batch 0 Loss 1.1725\n",
      "Epoch 32 Loss 1.222943\n",
      "Time taken for 1 epoch 111.63297533988953 sec\n",
      "\n",
      "Epoch 33 Batch 0 Loss 1.2755\n",
      "Epoch 33 Loss 1.223628\n",
      "Time taken for 1 epoch 111.41829061508179 sec\n",
      "\n",
      "Epoch 34 Batch 0 Loss 1.2099\n",
      "Epoch 34 Loss 1.222759\n",
      "Time taken for 1 epoch 111.83028554916382 sec\n",
      "\n",
      "Epoch 35 Batch 0 Loss 1.2716\n",
      "Epoch 35 Loss 1.222444\n",
      "Time taken for 1 epoch 111.38392519950867 sec\n",
      "\n",
      "Epoch 36 Batch 0 Loss 1.2679\n",
      "Epoch 36 Loss 1.222960\n",
      "Time taken for 1 epoch 111.11568236351013 sec\n",
      "\n",
      "Epoch 37 Batch 0 Loss 1.1969\n",
      "Epoch 37 Loss 1.223696\n",
      "Time taken for 1 epoch 110.65256857872009 sec\n",
      "\n",
      "Epoch 38 Batch 0 Loss 1.2262\n",
      "Epoch 38 Loss 1.223758\n",
      "Time taken for 1 epoch 109.91590690612793 sec\n",
      "\n",
      "Epoch 39 Batch 0 Loss 1.1424\n",
      "Epoch 39 Loss 1.223754\n",
      "Time taken for 1 epoch 110.29620051383972 sec\n",
      "\n",
      "Epoch 40 Batch 0 Loss 1.2141\n",
      "Epoch 40 Loss 1.223446\n",
      "Time taken for 1 epoch 110.25956344604492 sec\n",
      "\n",
      "Epoch 41 Batch 0 Loss 1.1718\n",
      "Epoch 41 Loss 1.224398\n",
      "Time taken for 1 epoch 110.84946417808533 sec\n",
      "\n",
      "Epoch 42 Batch 0 Loss 1.2910\n",
      "Epoch 42 Loss 1.223129\n",
      "Time taken for 1 epoch 109.94319534301758 sec\n",
      "\n",
      "Epoch 43 Batch 0 Loss 1.2477\n",
      "Epoch 43 Loss 1.223064\n",
      "Time taken for 1 epoch 110.36293601989746 sec\n",
      "\n",
      "Epoch 44 Batch 0 Loss 1.1895\n",
      "Epoch 44 Loss 1.223792\n",
      "Time taken for 1 epoch 110.76167416572571 sec\n",
      "\n",
      "Epoch 45 Batch 0 Loss 1.1518\n",
      "Epoch 45 Loss 1.223401\n",
      "Time taken for 1 epoch 110.07666873931885 sec\n",
      "\n",
      "Epoch 46 Batch 0 Loss 1.2362\n",
      "Epoch 46 Loss 1.222863\n",
      "Time taken for 1 epoch 111.05181765556335 sec\n",
      "\n",
      "Epoch 47 Batch 0 Loss 1.1552\n",
      "Epoch 47 Loss 1.223373\n",
      "Time taken for 1 epoch 109.51539611816406 sec\n",
      "\n",
      "Epoch 48 Batch 0 Loss 1.2346\n",
      "Epoch 48 Loss 1.223980\n",
      "Time taken for 1 epoch 109.22075891494751 sec\n",
      "\n",
      "Epoch 49 Batch 0 Loss 1.1647\n",
      "Epoch 49 Loss 1.223316\n",
      "Time taken for 1 epoch 109.21959114074707 sec\n",
      "\n",
      "Epoch 50 Batch 0 Loss 1.2224\n",
      "Epoch 50 Loss 1.223395\n",
      "Time taken for 1 epoch 109.69105195999146 sec\n",
      "\n",
      "Epoch 51 Batch 0 Loss 1.2088\n",
      "Epoch 51 Loss 1.224054\n",
      "Time taken for 1 epoch 114.69607353210449 sec\n",
      "\n",
      "Epoch 52 Batch 0 Loss 1.3268\n",
      "Epoch 52 Loss 1.223179\n",
      "Time taken for 1 epoch 115.09964036941528 sec\n",
      "\n",
      "Epoch 53 Batch 0 Loss 1.2215\n",
      "Epoch 53 Loss 1.224096\n",
      "Time taken for 1 epoch 110.26670742034912 sec\n",
      "\n",
      "Epoch 54 Batch 0 Loss 1.1875\n",
      "Epoch 54 Loss 1.223164\n",
      "Time taken for 1 epoch 109.5617024898529 sec\n",
      "\n",
      "Epoch 55 Batch 0 Loss 1.2189\n",
      "Epoch 55 Loss 1.223075\n",
      "Time taken for 1 epoch 109.40933322906494 sec\n",
      "\n",
      "Epoch 56 Batch 0 Loss 1.2539\n",
      "Epoch 56 Loss 1.223186\n",
      "Time taken for 1 epoch 110.06198167800903 sec\n",
      "\n",
      "Epoch 57 Batch 0 Loss 1.1783\n",
      "Epoch 57 Loss 1.223593\n",
      "Time taken for 1 epoch 109.08033227920532 sec\n",
      "\n",
      "Epoch 58 Batch 0 Loss 1.2558\n",
      "Epoch 58 Loss 1.224322\n",
      "Time taken for 1 epoch 109.38525176048279 sec\n",
      "\n",
      "Epoch 59 Batch 0 Loss 1.2450\n",
      "Epoch 59 Loss 1.222839\n",
      "Time taken for 1 epoch 109.02149891853333 sec\n",
      "\n",
      "Epoch 60 Batch 0 Loss 1.2242\n",
      "Epoch 60 Loss 1.223802\n",
      "Time taken for 1 epoch 108.82734680175781 sec\n",
      "\n",
      "Epoch 61 Batch 0 Loss 1.2114\n",
      "Epoch 61 Loss 1.223743\n",
      "Time taken for 1 epoch 117.77980065345764 sec\n",
      "\n",
      "Epoch 62 Batch 0 Loss 1.2213\n",
      "Epoch 62 Loss 1.222995\n",
      "Time taken for 1 epoch 109.0010449886322 sec\n",
      "\n",
      "Epoch 63 Batch 0 Loss 1.2814\n",
      "Epoch 63 Loss 1.224577\n",
      "Time taken for 1 epoch 130.17737007141113 sec\n",
      "\n",
      "Epoch 64 Batch 0 Loss 1.1797\n",
      "Epoch 64 Loss 1.223356\n",
      "Time taken for 1 epoch 109.2139732837677 sec\n",
      "\n",
      "Epoch 65 Batch 0 Loss 1.2487\n",
      "Epoch 65 Loss 1.223319\n",
      "Time taken for 1 epoch 108.9954206943512 sec\n",
      "\n",
      "Epoch 66 Batch 0 Loss 1.2419\n",
      "Epoch 66 Loss 1.222709\n",
      "Time taken for 1 epoch 111.80069851875305 sec\n",
      "\n",
      "Epoch 67 Batch 0 Loss 1.2681\n",
      "Epoch 67 Loss 1.223538\n",
      "Time taken for 1 epoch 109.62046813964844 sec\n",
      "\n",
      "Epoch 68 Batch 0 Loss 1.2347\n",
      "Epoch 68 Loss 1.222831\n",
      "Time taken for 1 epoch 109.62162899971008 sec\n",
      "\n",
      "Epoch 69 Batch 0 Loss 1.2187\n",
      "Epoch 69 Loss 1.222010\n",
      "Time taken for 1 epoch 108.67753887176514 sec\n",
      "\n",
      "Epoch 70 Batch 0 Loss 1.2283\n",
      "Epoch 70 Loss 1.222347\n",
      "Time taken for 1 epoch 108.66331219673157 sec\n",
      "\n",
      "Epoch 71 Batch 0 Loss 1.1283\n",
      "Epoch 71 Loss 1.222100\n",
      "Time taken for 1 epoch 109.6318998336792 sec\n",
      "\n",
      "Epoch 72 Batch 0 Loss 1.3548\n",
      "Epoch 72 Loss 1.222869\n",
      "Time taken for 1 epoch 109.95491671562195 sec\n",
      "\n",
      "Epoch 73 Batch 0 Loss 1.2206\n",
      "Epoch 73 Loss 1.222496\n",
      "Time taken for 1 epoch 109.57287120819092 sec\n",
      "\n",
      "Epoch 74 Batch 0 Loss 1.1221\n",
      "Epoch 74 Loss 1.222779\n",
      "Time taken for 1 epoch 109.09430360794067 sec\n",
      "\n",
      "Epoch 75 Batch 0 Loss 1.3100\n",
      "Epoch 75 Loss 1.222101\n",
      "Time taken for 1 epoch 108.3053708076477 sec\n",
      "\n",
      "Epoch 76 Batch 0 Loss 1.2447\n",
      "Epoch 76 Loss 1.221903\n",
      "Time taken for 1 epoch 110.15840458869934 sec\n",
      "\n",
      "Epoch 77 Batch 0 Loss 1.2200\n",
      "Epoch 77 Loss 1.223107\n",
      "Time taken for 1 epoch 109.12903809547424 sec\n",
      "\n",
      "Epoch 78 Batch 0 Loss 1.2693\n",
      "Epoch 78 Loss 1.222813\n",
      "Time taken for 1 epoch 108.73918628692627 sec\n",
      "\n",
      "Epoch 79 Batch 0 Loss 1.2880\n",
      "Epoch 79 Loss 1.222715\n",
      "Time taken for 1 epoch 109.50655508041382 sec\n",
      "\n",
      "Epoch 80 Batch 0 Loss 1.1973\n",
      "Epoch 80 Loss 1.223155\n",
      "Time taken for 1 epoch 109.2989935874939 sec\n",
      "\n",
      "Epoch 81 Batch 0 Loss 1.1809\n",
      "Epoch 81 Loss 1.223580\n",
      "Time taken for 1 epoch 109.51751708984375 sec\n",
      "\n",
      "Epoch 82 Batch 0 Loss 1.1368\n",
      "Epoch 82 Loss 1.222748\n",
      "Time taken for 1 epoch 109.29147100448608 sec\n",
      "\n",
      "Epoch 83 Batch 0 Loss 1.1203\n",
      "Epoch 83 Loss 1.223517\n",
      "Time taken for 1 epoch 108.54981923103333 sec\n",
      "\n",
      "Epoch 84 Batch 0 Loss 1.2533\n",
      "Epoch 84 Loss 1.223273\n",
      "Time taken for 1 epoch 108.49173402786255 sec\n",
      "\n",
      "Epoch 85 Batch 0 Loss 1.1391\n",
      "Epoch 85 Loss 1.220787\n",
      "Time taken for 1 epoch 108.3560140132904 sec\n",
      "\n",
      "Epoch 86 Batch 0 Loss 1.3317\n",
      "Epoch 86 Loss 1.223900\n",
      "Time taken for 1 epoch 109.2793653011322 sec\n",
      "\n",
      "Epoch 87 Batch 0 Loss 1.1885\n",
      "Epoch 87 Loss 1.223268\n",
      "Time taken for 1 epoch 109.17099595069885 sec\n",
      "\n",
      "Epoch 88 Batch 0 Loss 1.2182\n",
      "Epoch 88 Loss 1.222166\n",
      "Time taken for 1 epoch 116.48672461509705 sec\n",
      "\n",
      "Epoch 89 Batch 0 Loss 1.1629\n",
      "Epoch 89 Loss 1.222568\n",
      "Time taken for 1 epoch 110.55365562438965 sec\n",
      "\n",
      "Epoch 90 Batch 0 Loss 1.1608\n",
      "Epoch 90 Loss 1.222275\n",
      "Time taken for 1 epoch 108.93892455101013 sec\n",
      "\n",
      "Epoch 91 Batch 0 Loss 1.2223\n",
      "Epoch 91 Loss 1.221730\n",
      "Time taken for 1 epoch 109.33287787437439 sec\n",
      "\n",
      "Epoch 92 Batch 0 Loss 1.2096\n",
      "Epoch 92 Loss 1.223100\n",
      "Time taken for 1 epoch 110.56879830360413 sec\n",
      "\n",
      "Epoch 93 Batch 0 Loss 1.1248\n",
      "Epoch 93 Loss 1.223442\n",
      "Time taken for 1 epoch 108.98738813400269 sec\n",
      "\n",
      "Epoch 94 Batch 0 Loss 1.2186\n",
      "Epoch 94 Loss 1.221677\n",
      "Time taken for 1 epoch 108.8584372997284 sec\n",
      "\n",
      "Epoch 95 Batch 0 Loss 1.1785\n",
      "Epoch 95 Loss 1.222839\n",
      "Time taken for 1 epoch 108.53772759437561 sec\n",
      "\n",
      "Epoch 96 Batch 0 Loss 1.2531\n",
      "Epoch 96 Loss 1.220862\n",
      "Time taken for 1 epoch 110.21240639686584 sec\n",
      "\n",
      "Epoch 97 Batch 0 Loss 1.2305\n",
      "Epoch 97 Loss 1.221546\n",
      "Time taken for 1 epoch 109.98406767845154 sec\n",
      "\n",
      "Epoch 98 Batch 0 Loss 1.1573\n",
      "Epoch 98 Loss 1.223164\n",
      "Time taken for 1 epoch 109.45496225357056 sec\n",
      "\n",
      "Epoch 99 Batch 0 Loss 1.1986\n",
      "Epoch 99 Loss 1.222393\n",
      "Time taken for 1 epoch 109.94515800476074 sec\n",
      "\n",
      "Epoch 100 Batch 0 Loss 1.2111\n",
      "Epoch 100 Loss 1.221190\n",
      "Time taken for 1 epoch 109.31444191932678 sec\n",
      "\n"
     ]
    }
   ],
   "source": [
    "EPOCHS = 100\n",
    "\n",
    "\n",
    "for epoch in range(start_epoch, EPOCHS):\n",
    "    start = time.time()\n",
    "    total_loss = 0\n",
    "    \n",
    "    for (batch, (img_tensor, target)) in enumerate(dataset):\n",
    "        batch_loss, t_loss = train_step(img_tensor, target)\n",
    "        total_loss += t_loss\n",
    "        \n",
    "        if batch % 100 == 0:\n",
    "            print ('Epoch {} Batch {} Loss {:.4f}'.format(\n",
    "              epoch + 1, batch, batch_loss.numpy() / int(target.shape[1])))\n",
    "    \n",
    "    # storing the epoch end loss value to plot later\n",
    "    loss_plot.append(total_loss / num_steps)\n",
    "    wandb.log({\"loss\": total_loss / num_steps})\n",
    "    \n",
    "    \n",
    "    if epoch % 5 == 0:\n",
    "        ckpt_manager.save()\n",
    "    \n",
    "    \n",
    "    print ('Epoch {} Loss {:.6f}'.format(epoch + 1,\n",
    "                                         total_loss/num_steps))\n",
    "    print ('Time taken for 1 epoch {} sec\\n'.format(time.time() - start))\n",
    "    \n",
    "\n",
    "    "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 39,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.plot(loss_plot)\n",
    "plt.xlabel('Epochs')\n",
    "plt.ylabel('Loss')\n",
    "plt.title('Loss Plot')\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 40,
   "metadata": {},
   "outputs": [],
   "source": [
    "def evaluate(image):\n",
    "    \n",
    "    #Plotting the attention weights\n",
    "    attention_plot = np.zeros((max_length, attention_features_shape))\n",
    "\n",
    "    hidden = decoder.reset_state(batch_size=1)\n",
    "    \n",
    "    #loading image with random caption\n",
    "    img,_ = load_image(image,'')\n",
    "    temp_input = tf.expand_dims(img, 0)\n",
    "    \n",
    "    #extracting image features\n",
    "    features = encoder(temp_input)\n",
    "    \n",
    "    #dec imput\n",
    "    dec_input = tf.expand_dims([tokenizer.word_index['<start>']], 0)\n",
    "    result = []\n",
    "\n",
    "    for i in range(max_length):\n",
    "        predictions, hidden, attention_weights = decoder(dec_input, features, hidden)\n",
    "\n",
    "        attention_plot[i] = tf.reshape(attention_weights, (-1, )).numpy()\n",
    "\n",
    "        predicted_id = tf.random.categorical(predictions, 1)[0][0].numpy()\n",
    "        result.append(tokenizer.index_word[predicted_id])\n",
    "\n",
    "        if tokenizer.index_word[predicted_id] == '<end>':\n",
    "            return result, attention_plot\n",
    "\n",
    "        dec_input = tf.expand_dims([predicted_id], 0)\n",
    "\n",
    "    attention_plot = attention_plot[:len(result), :]\n",
    "    return result, attention_plot"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 41,
   "metadata": {},
   "outputs": [],
   "source": [
    "def plot_attention(image, result, attention_plot):\n",
    "    temp_image = np.array(Image.open(image))\n",
    "\n",
    "    fig = plt.figure(figsize=(10, 10))\n",
    "\n",
    "    len_result = len(result)\n",
    "    for l in range(len_result):\n",
    "        temp_att = np.resize(attention_plot[l], (8, 8))\n",
    "        ax = fig.add_subplot(len_result//2, len_result//2, l+1)\n",
    "        ax.set_title(result[l])\n",
    "        img = ax.imshow(temp_image)\n",
    "        ax.imshow(temp_att, cmap='gray', alpha=0.6, extent=img.get_extent())\n",
    "\n",
    "    plt.tight_layout()\n",
    "    plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 42,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Real Caption: <start> a tan dog and a cream colored dog run through grasslands <end>\n",
      "Prediction Caption: play kiss blue a black wearing <end>\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 720x720 with 7 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "\n",
    "# captions on the validation set\n",
    "rid = np.random.randint(0, len(img_name_val))\n",
    "image = img_name_val[rid]\n",
    "real_caption = ' '.join([tokenizer.index_word[i] for i in cap_val[rid] if i not in [0]])\n",
    "result, attention_plot = evaluate(image)\n",
    "\n",
    "print ('Real Caption:', real_caption)\n",
    "print ('Prediction Caption:', ' '.join(result))\n",
    "plot_attention(image, result, attention_plot)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "### include more data to get better captions for the image"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": 43,
   "metadata": {},
   "outputs": [],
   "source": [
    "# x = tf.random.normal(shape = (100,299,299,3))\n",
    "# x.shape\n",
    "# enc = CNN_Encoder(256)\n",
    "# enc(x).shape"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.6.9"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 4
}