Twitter Airline Sentiment.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# **Twitter US Airline Sentiment with Keras & TensorFlow**\n",
    "\n",
    "**Dataset:** Twitter US Airline Sentiment - https://www.kaggle.com/crowdflower/twitter-airline-sentiment/data#\n",
    "\n",
    "Originally came from Crowdflower's Data for Everyone library\n",
    "\n",
    "**Code:** Keras code example \"Using pre-trained word embeddings\" - https://keras.io/examples/nlp/pretrained_word_embeddings/"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Classify images within 10 categories ###"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [],
   "source": [
    "import tensorflow as tf\n",
    "from tensorflow import keras\n",
    "from tensorflow.keras import layers\n",
    "\n",
    "import numpy as np\n",
    "import pandas as pd\n",
    "\n",
    "%matplotlib inline\n",
    "import matplotlib as mpl\n",
    "import matplotlib.pyplot as plt"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "2.1.0\n"
     ]
    }
   ],
   "source": [
    "print(tf.__version__)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Get the data ##"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Dataset attributes:###"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "A sentiment analysis job about the problems of each major U.S. airline. Twitter data was scraped from February of 2015 and contributors were asked to first classify positive, negative, and neutral tweets, followed by categorizing negative reasons (such as \"late flight\" or \"rude service\")."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [],
   "source": [
    "import os\n",
    "\n",
    "TWITTER_AIRLINE_PATH = \"dataset\"\n",
    "\n",
    "def load_twitter_airline_data(twitter_airline_path=TWITTER_AIRLINE_PATH):\n",
    "    csv_path = os.path.join(twitter_airline_path, \"tweets.csv\")\n",
    "    return pd.read_csv(csv_path)"
   ]
  },
  {
   "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>tweet_id</th>\n",
       "      <th>airline_sentiment</th>\n",
       "      <th>airline_sentiment_confidence</th>\n",
       "      <th>negativereason</th>\n",
       "      <th>negativereason_confidence</th>\n",
       "      <th>airline</th>\n",
       "      <th>airline_sentiment_gold</th>\n",
       "      <th>name</th>\n",
       "      <th>negativereason_gold</th>\n",
       "      <th>retweet_count</th>\n",
       "      <th>text</th>\n",
       "      <th>tweet_coord</th>\n",
       "      <th>tweet_created</th>\n",
       "      <th>tweet_location</th>\n",
       "      <th>user_timezone</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>570306133677760513</td>\n",
       "      <td>neutral</td>\n",
       "      <td>1.0000</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>Virgin America</td>\n",
       "      <td>NaN</td>\n",
       "      <td>cairdin</td>\n",
       "      <td>NaN</td>\n",
       "      <td>0</td>\n",
       "      <td>@VirginAmerica What @dhepburn said.</td>\n",
       "      <td>NaN</td>\n",
       "      <td>2015-02-24 11:35:52 -0800</td>\n",
       "      <td>NaN</td>\n",
       "      <td>Eastern Time (US &amp; Canada)</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>570301130888122368</td>\n",
       "      <td>positive</td>\n",
       "      <td>0.3486</td>\n",
       "      <td>NaN</td>\n",
       "      <td>0.0000</td>\n",
       "      <td>Virgin America</td>\n",
       "      <td>NaN</td>\n",
       "      <td>jnardino</td>\n",
       "      <td>NaN</td>\n",
       "      <td>0</td>\n",
       "      <td>@VirginAmerica plus you've added commercials t...</td>\n",
       "      <td>NaN</td>\n",
       "      <td>2015-02-24 11:15:59 -0800</td>\n",
       "      <td>NaN</td>\n",
       "      <td>Pacific Time (US &amp; Canada)</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>570301083672813571</td>\n",
       "      <td>neutral</td>\n",
       "      <td>0.6837</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>Virgin America</td>\n",
       "      <td>NaN</td>\n",
       "      <td>yvonnalynn</td>\n",
       "      <td>NaN</td>\n",
       "      <td>0</td>\n",
       "      <td>@VirginAmerica I didn't today... Must mean I n...</td>\n",
       "      <td>NaN</td>\n",
       "      <td>2015-02-24 11:15:48 -0800</td>\n",
       "      <td>Lets Play</td>\n",
       "      <td>Central Time (US &amp; Canada)</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>570301031407624196</td>\n",
       "      <td>negative</td>\n",
       "      <td>1.0000</td>\n",
       "      <td>Bad Flight</td>\n",
       "      <td>0.7033</td>\n",
       "      <td>Virgin America</td>\n",
       "      <td>NaN</td>\n",
       "      <td>jnardino</td>\n",
       "      <td>NaN</td>\n",
       "      <td>0</td>\n",
       "      <td>@VirginAmerica it's really aggressive to blast...</td>\n",
       "      <td>NaN</td>\n",
       "      <td>2015-02-24 11:15:36 -0800</td>\n",
       "      <td>NaN</td>\n",
       "      <td>Pacific Time (US &amp; Canada)</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>570300817074462722</td>\n",
       "      <td>negative</td>\n",
       "      <td>1.0000</td>\n",
       "      <td>Can't Tell</td>\n",
       "      <td>1.0000</td>\n",
       "      <td>Virgin America</td>\n",
       "      <td>NaN</td>\n",
       "      <td>jnardino</td>\n",
       "      <td>NaN</td>\n",
       "      <td>0</td>\n",
       "      <td>@VirginAmerica and it's a really big bad thing...</td>\n",
       "      <td>NaN</td>\n",
       "      <td>2015-02-24 11:14:45 -0800</td>\n",
       "      <td>NaN</td>\n",
       "      <td>Pacific Time (US &amp; Canada)</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "             tweet_id airline_sentiment  airline_sentiment_confidence  \\\n",
       "0  570306133677760513           neutral                        1.0000   \n",
       "1  570301130888122368          positive                        0.3486   \n",
       "2  570301083672813571           neutral                        0.6837   \n",
       "3  570301031407624196          negative                        1.0000   \n",
       "4  570300817074462722          negative                        1.0000   \n",
       "\n",
       "  negativereason  negativereason_confidence         airline  \\\n",
       "0            NaN                        NaN  Virgin America   \n",
       "1            NaN                     0.0000  Virgin America   \n",
       "2            NaN                        NaN  Virgin America   \n",
       "3     Bad Flight                     0.7033  Virgin America   \n",
       "4     Can't Tell                     1.0000  Virgin America   \n",
       "\n",
       "  airline_sentiment_gold        name negativereason_gold  retweet_count  \\\n",
       "0                    NaN     cairdin                 NaN              0   \n",
       "1                    NaN    jnardino                 NaN              0   \n",
       "2                    NaN  yvonnalynn                 NaN              0   \n",
       "3                    NaN    jnardino                 NaN              0   \n",
       "4                    NaN    jnardino                 NaN              0   \n",
       "\n",
       "                                                text tweet_coord  \\\n",
       "0                @VirginAmerica What @dhepburn said.         NaN   \n",
       "1  @VirginAmerica plus you've added commercials t...         NaN   \n",
       "2  @VirginAmerica I didn't today... Must mean I n...         NaN   \n",
       "3  @VirginAmerica it's really aggressive to blast...         NaN   \n",
       "4  @VirginAmerica and it's a really big bad thing...         NaN   \n",
       "\n",
       "               tweet_created tweet_location               user_timezone  \n",
       "0  2015-02-24 11:35:52 -0800            NaN  Eastern Time (US & Canada)  \n",
       "1  2015-02-24 11:15:59 -0800            NaN  Pacific Time (US & Canada)  \n",
       "2  2015-02-24 11:15:48 -0800      Lets Play  Central Time (US & Canada)  \n",
       "3  2015-02-24 11:15:36 -0800            NaN  Pacific Time (US & Canada)  \n",
       "4  2015-02-24 11:14:45 -0800            NaN  Pacific Time (US & Canada)  "
      ]
     },
     "execution_count": 4,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "twitter_airline = load_twitter_airline_data()\n",
    "twitter_airline.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "negative    9178\n",
       "neutral     3099\n",
       "positive    2363\n",
       "Name: airline_sentiment, dtype: int64"
      ]
     },
     "execution_count": 5,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "twitter_airline[\"airline_sentiment\"].value_counts()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Data preparation"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [],
   "source": [
    "num_classes = 3"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [],
   "source": [
    "twitter_airline_prep = twitter_airline.replace({'positive':'0', 'neutral':'1', 'negative':'2'})"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [],
   "source": [
    "twitter_airline_prep = twitter_airline_prep.filter([\"airline_sentiment\", \"text\"])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [],
   "source": [
    "X_train = twitter_airline_prep.drop(\"airline_sentiment\", axis=1)\n",
    "y_train = twitter_airline_prep[\"airline_sentiment\"].copy()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [],
   "source": [
    "y_train = y_train.apply(pd.to_numeric) #Change the type to integer"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [],
   "source": [
    "X_test = twitter_airline_prep.drop(\"airline_sentiment\", axis=1)\n",
    "y_test = twitter_airline_prep[\"airline_sentiment\"].copy()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [],
   "source": [
    "y_test = y_test.apply(pd.to_numeric) #Change the type to integer"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Shuffle the training data:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {},
   "outputs": [],
   "source": [
    "np.random.seed(42)\n",
    "\n",
    "shuffle_index = np.random.permutation(X_train.shape[0])\n",
    "X_train, y_train = X_train.iloc[shuffle_index], y_train.iloc[shuffle_index]"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Model"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Consider the top 20,000 words, and truncate or pad sequences to be 200 tokens long:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {},
   "outputs": [],
   "source": [
    "from tensorflow.keras.layers.experimental.preprocessing import TextVectorization\n",
    "\n",
    "max_features = 20000\n",
    "embedding_dim = 128\n",
    "sequence_length = 200\n",
    "\n",
    "vectorizer = TextVectorization(\n",
    "    #standardize=custom_standardization,    Default is 'lower_and_strip_punctuation' (lowercase and remove punctuation)\n",
    "    max_tokens=max_features,\n",
    "    output_mode=\"int\",\n",
    "    output_sequence_length=sequence_length,\n",
    ")\n",
    "\n",
    "vectorizer.adapt(X_train.to_numpy())"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "TextVectorization transforms strings into a list of token integer indices, to index the vocabulary found in the dataset"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "[b'to', b'the', b'i', b'a', b'united']"
      ]
     },
     "execution_count": 15,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "vectorizer.get_vocabulary()[:5] "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([ 30, 102,  15, 133, 134], dtype=int64)"
      ]
     },
     "execution_count": 16,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "output = vectorizer(np.array([[\"This airline is really good\"]]))\n",
    "output.numpy()[0, :5]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {},
   "outputs": [],
   "source": [
    "voc = vectorizer.get_vocabulary()\n",
    "word_index = dict(zip(voc, range(2, len(voc))))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "[30, 102, 15, 133, 134]"
      ]
     },
     "execution_count": 18,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "test = [b\"this\", b\"airline\", b\"is\", b\"really\", b\"good\"]\n",
    "[word_index[w] for w in test]"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Use the pre-trained word vectors from https://nlp.stanford.edu/projects/glove/.<br>Jeffrey Pennington, Richard Socher, and Christopher D. Manning. 2014. GloVe: Global Vectors for Word Representation\n",
    "\n",
    "6B tokens, 400K vocabulary, uncased and 50, 100, 200 and 300 dimension vectors. Use the 100 dimensions vectors:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Found 400000 word vectors.\n"
     ]
    }
   ],
   "source": [
    "import io  #use io package to be able to specify the encoding when opening a file\n",
    "\n",
    "glove_path = os.path.join(TWITTER_AIRLINE_PATH, \"glove.6B.100d.txt\")\n",
    "\n",
    "embeddings_index = {}\n",
    "with io.open(glove_path, mode=\"r\", encoding=\"utf-8\") as f:\n",
    "    for line in f:\n",
    "        word, coefs = line.split(maxsplit=1)\n",
    "        coefs = np.fromstring(coefs, \"f\", sep=\" \")\n",
    "        embeddings_index[word] = coefs\n",
    "\n",
    "print(\"Found %s word vectors.\" % len(embeddings_index))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "The embedding matrix is a NumPy matrix where entry at index i is the pre-trained vector for the word of index i in the vectorizer's vocabulary."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Converted 10113 words (7017 misses)\n"
     ]
    }
   ],
   "source": [
    "num_tokens = len(voc) + 2\n",
    "embedding_dim = 100\n",
    "hits = 0\n",
    "misses = 0\n",
    "\n",
    "# Prepare embedding matrix\n",
    "embedding_matrix = np.zeros((num_tokens, embedding_dim))\n",
    "for word, i in word_index.items():\n",
    "    embedding_vector = embeddings_index.get(word.decode(\"utf-8\"))\n",
    "    if embedding_vector is not None:\n",
    "        # Words not found in embedding index will be all-zeros.\n",
    "        # This includes the representation for \"padding\" and \"OOV\"\n",
    "        embedding_matrix[i] = embedding_vector\n",
    "        hits += 1\n",
    "    else:\n",
    "        misses += 1\n",
    "print(\"Converted %d words (%d misses)\" % (hits, misses))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Visualize the dataset vocabulary vs. GloVe vocabulary"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "metadata": {},
   "outputs": [],
   "source": [
    "embedding_matrix_full = np.zeros((len(embeddings_index), embedding_dim))\n",
    "i = 0\n",
    "for embedding_vector in embeddings_index.values():\n",
    "    embedding_matrix_full[i] = embedding_vector\n",
    "    i += 1"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "metadata": {},
   "outputs": [],
   "source": [
    "from sklearn.decomposition import PCA\n",
    "\n",
    "pca = PCA(n_components=2, random_state=42)\n",
    "X2D = pca.fit_transform(embedding_matrix)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([0.1317866 , 0.04222822])"
      ]
     },
     "execution_count": 23,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "pca.explained_variance_ratio_"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "metadata": {},
   "outputs": [],
   "source": [
    "X2D_full = pca.fit_transform(embedding_matrix_full)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([0.06054711, 0.03435695])"
      ]
     },
     "execution_count": 25,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "pca.explained_variance_ratio_"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 720x504 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.figure(figsize=(10, 7))\n",
    "\n",
    "plt.scatter(X2D_full[:, 0], X2D_full[:, 1], color='blue', s=10, label=\"GloVe\")\n",
    "plt.scatter(X2D[:, 0], X2D[:, 1], color='red', s=3, label=\"Twitter Dataset\")\n",
    "plt.legend(loc=\"upper left\")\n",
    "\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Create a Keras layer with the embedding matrix:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "metadata": {},
   "outputs": [],
   "source": [
    "from tensorflow.keras.layers import Embedding\n",
    "\n",
    "embedding_layer = Embedding(\n",
    "    num_tokens,\n",
    "    embedding_dim,\n",
    "    embeddings_initializer=keras.initializers.Constant(embedding_matrix),\n",
    "    trainable=False,\n",
    ")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Model: \"model\"\n",
      "_________________________________________________________________\n",
      "Layer (type)                 Output Shape              Param #   \n",
      "=================================================================\n",
      "input_1 (InputLayer)         [(None, None)]            0         \n",
      "_________________________________________________________________\n",
      "embedding (Embedding)        (None, None, 100)         1713400   \n",
      "_________________________________________________________________\n",
      "conv1d (Conv1D)              (None, None, 128)         64128     \n",
      "_________________________________________________________________\n",
      "max_pooling1d (MaxPooling1D) (None, None, 128)         0         \n",
      "_________________________________________________________________\n",
      "conv1d_1 (Conv1D)            (None, None, 128)         82048     \n",
      "_________________________________________________________________\n",
      "max_pooling1d_1 (MaxPooling1 (None, None, 128)         0         \n",
      "_________________________________________________________________\n",
      "conv1d_2 (Conv1D)            (None, None, 128)         82048     \n",
      "_________________________________________________________________\n",
      "global_max_pooling1d (Global (None, 128)               0         \n",
      "_________________________________________________________________\n",
      "dense (Dense)                (None, 128)               16512     \n",
      "_________________________________________________________________\n",
      "dropout (Dropout)            (None, 128)               0         \n",
      "_________________________________________________________________\n",
      "dense_1 (Dense)              (None, 3)                 387       \n",
      "=================================================================\n",
      "Total params: 1,958,523\n",
      "Trainable params: 245,123\n",
      "Non-trainable params: 1,713,400\n",
      "_________________________________________________________________\n"
     ]
    }
   ],
   "source": [
    "from tensorflow.keras import layers\n",
    "\n",
    "int_sequences_input = keras.Input(shape=(None,), dtype=\"int64\")\n",
    "embedded_sequences = embedding_layer(int_sequences_input)\n",
    "x = layers.Conv1D(128, 5, activation=\"relu\")(embedded_sequences)\n",
    "x = layers.MaxPooling1D(5)(x)\n",
    "x = layers.Conv1D(128, 5, activation=\"relu\")(x)\n",
    "x = layers.MaxPooling1D(5)(x)\n",
    "x = layers.Conv1D(128, 5, activation=\"relu\")(x)\n",
    "x = layers.GlobalMaxPooling1D()(x)\n",
    "x = layers.Dense(128, activation=\"relu\")(x)\n",
    "x = layers.Dropout(0.5)(x)\n",
    "preds = layers.Dense(num_classes, activation=\"softmax\")(x)\n",
    "model = keras.Model(int_sequences_input, preds)\n",
    "model.summary()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "The training input is vectorized:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "metadata": {},
   "outputs": [],
   "source": [
    "X_train_prep = vectorizer(X_train.to_numpy()).numpy()\n",
    "\n",
    "y_train_prep = np.array(y_train)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 30,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Train on 13176 samples, validate on 1464 samples\n",
      "Epoch 1/20\n",
      "13176/13176 [==============================] - 18s 1ms/sample - loss: 0.8028 - accuracy: 0.6655 - val_loss: 0.6575 - val_accuracy: 0.7193\n",
      "Epoch 2/20\n",
      "13176/13176 [==============================] - 17s 1ms/sample - loss: 0.6228 - accuracy: 0.7577 - val_loss: 0.7143 - val_accuracy: 0.7206\n",
      "Epoch 3/20\n",
      "13176/13176 [==============================] - 17s 1ms/sample - loss: 0.5395 - accuracy: 0.7895 - val_loss: 0.5517 - val_accuracy: 0.7725\n",
      "Epoch 4/20\n",
      "13176/13176 [==============================] - 17s 1ms/sample - loss: 0.4810 - accuracy: 0.8163 - val_loss: 0.6436 - val_accuracy: 0.7643\n",
      "Epoch 5/20\n",
      "13176/13176 [==============================] - 17s 1ms/sample - loss: 0.4216 - accuracy: 0.8432 - val_loss: 0.5353 - val_accuracy: 0.7862\n",
      "Epoch 6/20\n",
      "13176/13176 [==============================] - 17s 1ms/sample - loss: 0.3574 - accuracy: 0.8698 - val_loss: 0.6130 - val_accuracy: 0.7719\n",
      "Epoch 7/20\n",
      "13176/13176 [==============================] - 18s 1ms/sample - loss: 0.2984 - accuracy: 0.8941 - val_loss: 0.8694 - val_accuracy: 0.7411\n",
      "Epoch 8/20\n",
      "13176/13176 [==============================] - 19s 1ms/sample - loss: 0.2572 - accuracy: 0.9116 - val_loss: 0.9866 - val_accuracy: 0.6851\n",
      "Epoch 9/20\n",
      "13176/13176 [==============================] - 18s 1ms/sample - loss: 0.2219 - accuracy: 0.9240 - val_loss: 1.3939 - val_accuracy: 0.7370\n",
      "Epoch 10/20\n",
      "13176/13176 [==============================] - 17s 1ms/sample - loss: 0.1878 - accuracy: 0.9404 - val_loss: 0.8650 - val_accuracy: 0.7807\n",
      "Epoch 11/20\n",
      "13176/13176 [==============================] - 18s 1ms/sample - loss: 0.1871 - accuracy: 0.9446 - val_loss: 0.8943 - val_accuracy: 0.7609\n",
      "Epoch 12/20\n",
      "13176/13176 [==============================] - 17s 1ms/sample - loss: 0.1391 - accuracy: 0.9587 - val_loss: 1.0250 - val_accuracy: 0.7705\n",
      "Epoch 13/20\n",
      "13176/13176 [==============================] - 18s 1ms/sample - loss: 0.1546 - accuracy: 0.9584 - val_loss: 0.9562 - val_accuracy: 0.7650\n",
      "Epoch 14/20\n",
      "13176/13176 [==============================] - 18s 1ms/sample - loss: 0.1161 - accuracy: 0.9638 - val_loss: 1.1013 - val_accuracy: 0.7719\n",
      "Epoch 15/20\n",
      "13176/13176 [==============================] - 18s 1ms/sample - loss: 0.1136 - accuracy: 0.9667 - val_loss: 1.3730 - val_accuracy: 0.7705\n",
      "Epoch 16/20\n",
      "13176/13176 [==============================] - 18s 1ms/sample - loss: 0.1035 - accuracy: 0.9692 - val_loss: 1.3575 - val_accuracy: 0.7698\n",
      "Epoch 17/20\n",
      "13176/13176 [==============================] - 19s 1ms/sample - loss: 0.1190 - accuracy: 0.9667 - val_loss: 1.1686 - val_accuracy: 0.7678\n",
      "Epoch 18/20\n",
      "13176/13176 [==============================] - 18s 1ms/sample - loss: 0.1043 - accuracy: 0.9737 - val_loss: 1.2161 - val_accuracy: 0.7760\n",
      "Epoch 19/20\n",
      "13176/13176 [==============================] - 18s 1ms/sample - loss: 0.0844 - accuracy: 0.9769 - val_loss: 1.3855 - val_accuracy: 0.7698\n",
      "Epoch 20/20\n",
      "13176/13176 [==============================] - 18s 1ms/sample - loss: 0.0865 - accuracy: 0.9766 - val_loss: 1.0405 - val_accuracy: 0.7561\n"
     ]
    }
   ],
   "source": [
    "batch_size = 128\n",
    "epochs = 20\n",
    "\n",
    "model.compile(\n",
    "    loss=\"sparse_categorical_crossentropy\", optimizer=\"rmsprop\", metrics=[\"accuracy\"]\n",
    ")\n",
    "history = model.fit(X_train_prep, y_train_prep, batch_size=batch_size, epochs=epochs, validation_split = 0.1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 31,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 576x360 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "pd.DataFrame(history.history).plot(figsize=(8, 5))\n",
    "plt.grid(True)\n",
    "plt.gca().set_ylim(0, 1)\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 32,
   "metadata": {},
   "outputs": [],
   "source": [
    "X_test_prep = vectorizer(X_test.to_numpy()).numpy()\n",
    "\n",
    "y_test_prep = np.array(y_test)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 33,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Test loss: 0.147441007135324\n",
      "Test accuracy: 0.96530056\n"
     ]
    }
   ],
   "source": [
    "score = model.evaluate(X_test_prep, y_test_prep, verbose=0)\n",
    "print(\"Test loss:\", score[0])\n",
    "print(\"Test accuracy:\", score[1])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 34,
   "metadata": {},
   "outputs": [],
   "source": [
    "def predict_sentiment(text = \"OK\"):\n",
    "    \n",
    "    sentiment_mapping = ['positive', 'neutral', 'negative']\n",
    "\n",
    "    string_input = keras.Input(shape=(1,), dtype=\"string\")\n",
    "    x = vectorizer(string_input)\n",
    "    preds = model(x)\n",
    "    end_to_end_model = keras.Model(string_input, preds)\n",
    "\n",
    "    probabilities = end_to_end_model.predict(\n",
    "        [[text]]\n",
    "    )\n",
    "\n",
    "    return sentiment_mapping[np.argmax(probabilities[0])]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 35,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "negative\n"
     ]
    }
   ],
   "source": [
    "print(predict_sentiment(\"The flight was not so great, due to the long delay.\"))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 36,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "neutral\n"
     ]
    }
   ],
   "source": [
    "print(predict_sentiment(\"The flight was OK.\"))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 37,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "positive\n"
     ]
    }
   ],
   "source": [
    "print(predict_sentiment(\"What a nice experience all along!\"))"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.7.7"
  },
  "nav_menu": {
   "height": "279px",
   "width": "309px"
  },
  "toc": {
   "nav_menu": {},
   "number_sections": true,
   "sideBar": true,
   "skip_h1_title": false,
   "toc_cell": false,
   "toc_position": {},
   "toc_section_display": "block",
   "toc_window_display": false
  }
 },
 "nbformat": 4,
 "nbformat_minor": 1
}