emotiondetection.ipynb

{
  "cells": [
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "K1qqTwQ6hfQ9",
        "outputId": "fb4200b6-e2ca-4bf2-bb9f-c168ee2efd4f"
      },
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Looking in indexes: https://pypi.org/simple, https://us-python.pkg.dev/colab-wheels/public/simple/\n",
            "Requirement already satisfied: emot in /usr/local/lib/python3.7/dist-packages (3.1)\n"
          ]
        }
      ],
      "source": [
        "!pip install emot"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "t93yvEWBhoGc"
      },
      "outputs": [],
      "source": [
        "import numpy as np\n",
        "import pandas as pd\n",
        "import nltk\n",
        "import string\n",
        "import tensorflow\n",
        "import seaborn as sns\n",
        "import matplotlib.pyplot as plt\n",
        "import re, sys, os, csv, keras, pickle\n",
        "from nltk.corpus import stopwords\n",
        "from nltk.tokenize import word_tokenize\n",
        "from collections import Counter\n",
        "from sklearn.base import BaseEstimator, TransformerMixin\n",
        "from emot.emo_unicode import UNICODE_EMOJI, EMOTICONS_EMO\n",
        "from nltk.corpus import stopwords\n",
        "from nltk.corpus import wordnet\n",
        "from nltk.tokenize import word_tokenize\n",
        "from nltk.stem import PorterStemmer\n",
        "from nltk.stem import WordNetLemmatizer\n",
        "\n",
        "from keras import regularizers, initializers, optimizers, callbacks\n",
        "from keras.preprocessing.text import Tokenizer\n",
        "from keras.preprocessing.sequence import pad_sequences\n",
        "from keras.utils.np_utils import to_categorical\n",
        "from keras.layers import Conv1D, Dense, Input, Flatten, Concatenate\n",
        "from keras.layers import  MaxPooling1D, Embedding, Dropout, LSTM, GRU, Bidirectional\n",
        "from keras.models import Model\n",
        "from keras import backend as K\n",
        "from keras.layers import Layer, InputSpec\n",
        "from keras.callbacks import ModelCheckpoint\n",
        "\n",
        "\n",
        "%run ExtraFunctions.ipynb"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "n2m3W6H3hpxL",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 441
        },
        "outputId": "cd4cf19e-b9a1-4497-9718-500a26e9534d"
      },
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "shape:  (40000, 4)\n"
          ]
        },
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "         tweet_id   sentiment         author  \\\n",
              "0      1956967341       empty     xoshayzers   \n",
              "1      1956967666     sadness      wannamama   \n",
              "2      1956967696     sadness      coolfunky   \n",
              "3      1956967789  enthusiasm    czareaquino   \n",
              "4      1956968416     neutral      xkilljoyx   \n",
              "...           ...         ...            ...   \n",
              "39995  1753918954     neutral  showMe_Heaven   \n",
              "39996  1753919001        love       drapeaux   \n",
              "39997  1753919005        love       JenniRox   \n",
              "39998  1753919043   happiness       ipdaman1   \n",
              "39999  1753919049        love    Alpharalpha   \n",
              "\n",
              "                                                 content  \n",
              "0      @tiffanylue i know  i was listenin to bad habi...  \n",
              "1      Layin n bed with a headache  ughhhh...waitin o...  \n",
              "2                    Funeral ceremony...gloomy friday...  \n",
              "3                   wants to hang out with friends SOON!  \n",
              "4      @dannycastillo We want to trade with someone w...  \n",
              "...                                                  ...  \n",
              "39995                                   @JohnLloydTaylor  \n",
              "39996                     Happy Mothers Day  All my love  \n",
              "39997  Happy Mother's Day to all the mommies out ther...  \n",
              "39998  @niariley WASSUP BEAUTIFUL!!! FOLLOW ME!!  PEE...  \n",
              "39999  @mopedronin bullet train from tokyo    the gf ...  \n",
              "\n",
              "[40000 rows x 4 columns]"
            ],
            "text/html": [
              "\n",
              "  <div id=\"df-daad89de-5bf6-4bd4-aad4-2abe5ff687ed\">\n",
              "    <div class=\"colab-df-container\">\n",
              "      <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>sentiment</th>\n",
              "      <th>author</th>\n",
              "      <th>content</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>0</th>\n",
              "      <td>1956967341</td>\n",
              "      <td>empty</td>\n",
              "      <td>xoshayzers</td>\n",
              "      <td>@tiffanylue i know  i was listenin to bad habi...</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>1956967666</td>\n",
              "      <td>sadness</td>\n",
              "      <td>wannamama</td>\n",
              "      <td>Layin n bed with a headache  ughhhh...waitin o...</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2</th>\n",
              "      <td>1956967696</td>\n",
              "      <td>sadness</td>\n",
              "      <td>coolfunky</td>\n",
              "      <td>Funeral ceremony...gloomy friday...</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3</th>\n",
              "      <td>1956967789</td>\n",
              "      <td>enthusiasm</td>\n",
              "      <td>czareaquino</td>\n",
              "      <td>wants to hang out with friends SOON!</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>4</th>\n",
              "      <td>1956968416</td>\n",
              "      <td>neutral</td>\n",
              "      <td>xkilljoyx</td>\n",
              "      <td>@dannycastillo We want to trade with someone w...</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>...</th>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>39995</th>\n",
              "      <td>1753918954</td>\n",
              "      <td>neutral</td>\n",
              "      <td>showMe_Heaven</td>\n",
              "      <td>@JohnLloydTaylor</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>39996</th>\n",
              "      <td>1753919001</td>\n",
              "      <td>love</td>\n",
              "      <td>drapeaux</td>\n",
              "      <td>Happy Mothers Day  All my love</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>39997</th>\n",
              "      <td>1753919005</td>\n",
              "      <td>love</td>\n",
              "      <td>JenniRox</td>\n",
              "      <td>Happy Mother's Day to all the mommies out ther...</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>39998</th>\n",
              "      <td>1753919043</td>\n",
              "      <td>happiness</td>\n",
              "      <td>ipdaman1</td>\n",
              "      <td>@niariley WASSUP BEAUTIFUL!!! FOLLOW ME!!  PEE...</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>39999</th>\n",
              "      <td>1753919049</td>\n",
              "      <td>love</td>\n",
              "      <td>Alpharalpha</td>\n",
              "      <td>@mopedronin bullet train from tokyo    the gf ...</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "<p>40000 rows × 4 columns</p>\n",
              "</div>\n",
              "      <button class=\"colab-df-convert\" onclick=\"convertToInteractive('df-daad89de-5bf6-4bd4-aad4-2abe5ff687ed')\"\n",
              "              title=\"Convert this dataframe to an interactive table.\"\n",
              "              style=\"display:none;\">\n",
              "        \n",
              "  <svg xmlns=\"http://www.w3.org/2000/svg\" height=\"24px\"viewBox=\"0 0 24 24\"\n",
              "       width=\"24px\">\n",
              "    <path d=\"M0 0h24v24H0V0z\" fill=\"none\"/>\n",
              "    <path d=\"M18.56 5.44l.94 2.06.94-2.06 2.06-.94-2.06-.94-.94-2.06-.94 2.06-2.06.94zm-11 1L8.5 8.5l.94-2.06 2.06-.94-2.06-.94L8.5 2.5l-.94 2.06-2.06.94zm10 10l.94 2.06.94-2.06 2.06-.94-2.06-.94-.94-2.06-.94 2.06-2.06.94z\"/><path d=\"M17.41 7.96l-1.37-1.37c-.4-.4-.92-.59-1.43-.59-.52 0-1.04.2-1.43.59L10.3 9.45l-7.72 7.72c-.78.78-.78 2.05 0 2.83L4 21.41c.39.39.9.59 1.41.59.51 0 1.02-.2 1.41-.59l7.78-7.78 2.81-2.81c.8-.78.8-2.07 0-2.86zM5.41 20L4 18.59l7.72-7.72 1.47 1.35L5.41 20z\"/>\n",
              "  </svg>\n",
              "      </button>\n",
              "      \n",
              "  <style>\n",
              "    .colab-df-container {\n",
              "      display:flex;\n",
              "      flex-wrap:wrap;\n",
              "      gap: 12px;\n",
              "    }\n",
              "\n",
              "    .colab-df-convert {\n",
              "      background-color: #E8F0FE;\n",
              "      border: none;\n",
              "      border-radius: 50%;\n",
              "      cursor: pointer;\n",
              "      display: none;\n",
              "      fill: #1967D2;\n",
              "      height: 32px;\n",
              "      padding: 0 0 0 0;\n",
              "      width: 32px;\n",
              "    }\n",
              "\n",
              "    .colab-df-convert:hover {\n",
              "      background-color: #E2EBFA;\n",
              "      box-shadow: 0px 1px 2px rgba(60, 64, 67, 0.3), 0px 1px 3px 1px rgba(60, 64, 67, 0.15);\n",
              "      fill: #174EA6;\n",
              "    }\n",
              "\n",
              "    [theme=dark] .colab-df-convert {\n",
              "      background-color: #3B4455;\n",
              "      fill: #D2E3FC;\n",
              "    }\n",
              "\n",
              "    [theme=dark] .colab-df-convert:hover {\n",
              "      background-color: #434B5C;\n",
              "      box-shadow: 0px 1px 3px 1px rgba(0, 0, 0, 0.15);\n",
              "      filter: drop-shadow(0px 1px 2px rgba(0, 0, 0, 0.3));\n",
              "      fill: #FFFFFF;\n",
              "    }\n",
              "  </style>\n",
              "\n",
              "      <script>\n",
              "        const buttonEl =\n",
              "          document.querySelector('#df-daad89de-5bf6-4bd4-aad4-2abe5ff687ed button.colab-df-convert');\n",
              "        buttonEl.style.display =\n",
              "          google.colab.kernel.accessAllowed ? 'block' : 'none';\n",
              "\n",
              "        async function convertToInteractive(key) {\n",
              "          const element = document.querySelector('#df-daad89de-5bf6-4bd4-aad4-2abe5ff687ed');\n",
              "          const dataTable =\n",
              "            await google.colab.kernel.invokeFunction('convertToInteractive',\n",
              "                                                     [key], {});\n",
              "          if (!dataTable) return;\n",
              "\n",
              "          const docLinkHtml = 'Like what you see? Visit the ' +\n",
              "            '<a target=\"_blank\" href=https://colab.research.google.com/notebooks/data_table.ipynb>data table notebook</a>'\n",
              "            + ' to learn more about interactive tables.';\n",
              "          element.innerHTML = '';\n",
              "          dataTable['output_type'] = 'display_data';\n",
              "          await google.colab.output.renderOutput(dataTable, element);\n",
              "          const docLink = document.createElement('div');\n",
              "          docLink.innerHTML = docLinkHtml;\n",
              "          element.appendChild(docLink);\n",
              "        }\n",
              "      </script>\n",
              "    </div>\n",
              "  </div>\n",
              "  "
            ]
          },
          "metadata": {},
          "execution_count": 5
        }
      ],
      "source": [
        "train_data = pd.read_csv(\"text_emotion.csv\", sep = \",\")\n",
        "\n",
        "print(\"shape: \", train_data.shape)\n",
        "train_data"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "ybyb4l-4o7uz"
      },
      "outputs": [],
      "source": [
        "def get_wordnet_pos(word):\n",
        "        tag = nltk.pos_tag([word])[0][1][0].upper()\n",
        "        tag_dict = {\"J\": wordnet.ADJ, \"N\": wordnet.NOUN, \"V\": wordnet.VERB, \"R\": wordnet.ADV}\n",
        "        return tag_dict.get(tag, wordnet.NOUN)   \n",
        "  \n",
        "def prep(tweet):\n",
        "    ps = PorterStemmer()\n",
        "    pattern = re.compile(r\"(.)\\1{2,}\")\n",
        "    tweet = pattern.sub(r\"\\1\\1\", tweet)\n",
        "    tweet = re.sub(r'http.?://[^\\s]+[\\s]?', '', tweet)\n",
        "    tweet = re.sub('\\d+', '', tweet)\n",
        "    punct = string.punctuation\n",
        "    trantab = str.maketrans(punct, len(punct) * ' ')  # Every punctuation symbol will be replaced by a space\n",
        "    tweet = tweet.translate(trantab)\n",
        "    tweet = tweet.lower()\n",
        "    words = tweet.split()\n",
        "    stemmed_words = [ps.stem(word) for word in words]\n",
        "    lemmatizer = WordNetLemmatizer()\n",
        "    lemma_words = [lemmatizer.lemmatize(word, get_wordnet_pos(word)) for word in stemmed_words]\n",
        "    tweet = \" \".join(lemma_words)\n",
        "    return tweet       \n",
        "\n",
        "def word_prob(word):\n",
        "  return dictionary[word] / total\n",
        "\n",
        "def words(text):\n",
        "  '''returns all the words in the input \"text\" as a list '''\n",
        "  return re.findall('[a-z]+', text.lower())\n",
        "  "
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "MMHKO54xpYNN",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "0bae72ef-4bc8-407f-c54f-80966d738ad9"
      },
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Max word length:  45\n",
            "total:  535565.0\n"
          ]
        }
      ],
      "source": [
        "# Counter finds the frequency of each word in the list\n",
        "dictionary = Counter(words(open('/content/merged.txt').read())) \n",
        "\n",
        "# take the max_word_length as the one with maximum frequency\n",
        "max_word_length = max(map(len, dictionary))\n",
        "print(\"Max word length: \", max_word_length)\n",
        "\n",
        "# calculate total as the sum of frequencies\n",
        "total = float(sum(dictionary.values()))\n",
        "print(\"total: \", total)"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "k_tfnnDbpqWO"
      },
      "outputs": [],
      "source": [
        "def viterbi_segment(text):\n",
        "    probs, lasts = [1.0], [0]\n",
        "    for i in range(1, len(text) + 1):\n",
        "        prob_k, k = max((probs[j] * word_prob(text[j:i]), j)\n",
        "                        for j in range(max(0, i - max_word_length), i))\n",
        "        probs.append(prob_k)\n",
        "        lasts.append(k)\n",
        "    words = []\n",
        "    i = len(text)\n",
        "    while 0 < i:\n",
        "        words.append(text[lasts[i]:i])\n",
        "        i = lasts[i]\n",
        "    words.reverse()\n",
        "    return words, probs[-1]\n",
        "\n",
        "def fix_hashtag(text):\n",
        "    text = text.group().split(\":\")[0]\n",
        "    text = text[1:] # remove '#'\n",
        "    try:\n",
        "        test = int(text[0])\n",
        "        text = text[1:]\n",
        "    except:\n",
        "        pass\n",
        "    output = ' '.join(viterbi_segment(text)[0])\n",
        "    return output\n",
        "\n",
        "def clean_tweet(tweet):\n",
        "    tweet = tweet.lower()\n",
        "    tweet = re.sub(\"(#[A-Za-z0-9]+)\", fix_hashtag, tweet)\n",
        "    return ' '.join(re.sub(\"(@[A-Za-z0-9]+)|([^0-9A-Za-z \\t])|(\\w+:\\/\\/\\S+)\", \" \", tweet).split())\n",
        "\n",
        "\n",
        "def vectorise_label(label):\n",
        "    if label == \"empty\":return 1 \n",
        "    elif label == \"sadness\":return 2 \n",
        "    elif label == \"enthusiasm\":return 3 \n",
        "    elif label == \"neutral\":return 0\n",
        "    elif label == \"worry\":return 4\n",
        "    elif label == \"surprise\":return 5 \n",
        "    elif label == \"love\":return 6 \n",
        "    elif label == \"fun\":return 7 \n",
        "    elif label == \"hate\":return 8 \n",
        "    elif label == \"happiness\":return 9\n",
        "    elif label == \"boredom\":return 10 \n",
        "    elif label == \"relief\":return 11 \n",
        "    elif label == \"anger\":return 12 "
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "FfrG-PU-qjXI",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "26b58130-7e3a-4d19-fe37-57245215e8e9"
      },
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "data.csv created !!\n"
          ]
        }
      ],
      "source": [
        "datawriter = csv.writer(open(\"data.csv\", 'w'), delimiter=',',lineterminator=\"\\n\")\n",
        "\n",
        "totalrows = 40000\n",
        "for row in range(totalrows):\n",
        "  tweet= clean_tweet(train_data.content[row])\n",
        "  datawriter.writerow([tweet, str(vectorise_label(train_data.sentiment[row]))])\n",
        "print(\"data.csv created !!\")\n"
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "def read_csv(file, lis, lis_=[]):\n",
        "    with open(file, newline='') as f:\n",
        "        reader = csv.reader(f)\n",
        "        # print(reader)\n",
        "        for i,row in enumerate(reader):\n",
        "            # print()\n",
        "            cat = row[1]\n",
        "            lis.append(row[0])\n",
        "            lis_.append(int(row[1]))\n",
        "        return  lis, lis_\n",
        "\n",
        "data, emotion =  read_csv(\"data.csv\", lis=[], lis_=[])\n",
        "print(len(data), len(emotion))"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "G_PQMKwOAcZ2",
        "outputId": "81adf87f-6249-4ea9-e64c-186350abf9bf"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "39968 39968\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "dataWriter = csv.writer(open('merged_data.csv', 'w'), delimiter=',',lineterminator=\"\\n\")\n",
        "for n,sentence in enumerate(data):\n",
        "    # print(point)\n",
        "    category = emotion[n]\n",
        "    is_happy = category == 3 or category == 5 or category == 7 or category == 9 or category == 11 \n",
        "    is_neutral = category == 0 or category == 1 or  category == 10\n",
        "    is_sad = category == 2 or category == 4\n",
        "    is_anger = category == 8 or category == 12\n",
        "    \n",
        "    if is_happy:\n",
        "        dataWriter.writerow([sentence, 1])\n",
        "    elif is_neutral:\n",
        "        dataWriter.writerow([sentence, 0])\n",
        "    elif is_sad:\n",
        "        dataWriter.writerow([sentence, 2])\n",
        "    elif is_anger:\n",
        "        dataWriter.writerow([sentence, 4])\n",
        "    elif category == 6:\n",
        "        dataWriter.writerow([sentence, 3])\n",
        "    else:\n",
        "        print(sentence,n)"
      ],
      "metadata": {
        "id": "nsAxQ4K3DOq7"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "FwRWBE2mtO6F"
      },
      "outputs": [],
      "source": [
        "MAX_NB_WORDS = 56000 \n",
        "MAX_SEQUENCE_LENGTH = 30\n",
        "VALIDATION_SPLIT = 0.20\n",
        "EMBEDDING_DIM = 200 "
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "4c-oVe7Uvwck"
      },
      "outputs": [],
      "source": [
        "# texts1, labels1 = [], []\n",
        "# data = pd.read_csv(\"merged_data.csv\", delimiter=',')\n",
        "# for a, b in data.itertuples(index=False):\n",
        "#   # print(a,b)\n",
        "  \n",
        "#   texts1.append(a)\n",
        "#   labels1.append(b)\n",
        "# print(texts1)\n",
        "# # print(texts)\n",
        "# print(len(texts1), len(labels1))"
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "data_train = pd.read_csv(\"merged_data.csv\", sep=',', encoding='utf-8' )\n",
        "\n",
        "count  = data_train.iloc[:,1].value_counts()\n",
        "\n",
        "plt.figure(figsize=(10,7))\n",
        "sns.barplot(count.index, count.values, alpha=0.6, palette=\"plasma\")\n",
        "plt.ylabel('Count', fontsize=10)\n",
        "plt.xlabel('Emotions', fontsize=10)\n",
        "plt.show()\n",
        "\n",
        "# 0 - neutral\n",
        "# 1 - happy\n",
        "# 2 - sad\n",
        "# 3 - love\n",
        "# 4 - anger "
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 497
        },
        "id": "5QQJe5zai-ZC",
        "outputId": "da04545a-ef5c-4bb4-9ad4-36ac2a9cf327"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "/usr/local/lib/python3.7/dist-packages/seaborn/_decorators.py:43: FutureWarning: Pass the following variables as keyword args: x, y. From version 0.12, the only valid positional argument will be `data`, and passing other arguments without an explicit keyword will result in an error or misinterpretation.\n",
            "  FutureWarning\n"
          ]
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 720x504 with 1 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "23pzDIxrzqXo",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "cc508a35-5565-46de-de8a-594861a6b025"
      },
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "39858 39858\n"
          ]
        }
      ],
      "source": [
        "texts, labels = [], []\n",
        "with open('/content/merged_data.csv') as csvfile:\n",
        "    readCSV = csv.reader(csvfile, delimiter=',')\n",
        "    for row in readCSV:\n",
        "        # print(row)\n",
        "        texts.append(row[0])\n",
        "        labels.append(row[1])\n",
        "print(len(texts), len(labels))"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "4GPtKcBqxMl3"
      },
      "outputs": [],
      "source": [
        "with open('/content/tokenizer.pickle', 'rb') as handle:\n",
        "    tokenizer = pickle.load(handle)"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "_VArkh0EyyYJ",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "786e7e8c-081f-48cd-af0f-e7b62828a2a5"
      },
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "unique words:  34359\n"
          ]
        }
      ],
      "source": [
        "sequences = tokenizer.texts_to_sequences(texts)\n",
        "word_index = tokenizer.word_index\n",
        "data_int = pad_sequences(sequences, padding='pre', maxlen=(MAX_SEQUENCE_LENGTH-5))\n",
        "data = pad_sequences(data_int, padding='post', maxlen=(MAX_SEQUENCE_LENGTH))\n",
        "\n",
        "print(\"unique words: \", len(word_index))"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "BuJ1rWbKzUJ6",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "a2eeb5fd-8492-47db-ab33-276995a45544"
      },
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Shape of data tensor: (39858, 30)\n",
            "Shape of label tensor: (39858, 5)\n"
          ]
        }
      ],
      "source": [
        "# convert labels into onehot vectors\n",
        "labels = to_categorical(np.asarray(labels))\n",
        "print('Shape of data tensor:', data.shape)\n",
        "print('Shape of label tensor:', labels.shape)"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "uoD4yR9E07CT"
      },
      "outputs": [],
      "source": [
        "indices = np.arange(data.shape[0])\n",
        "np.random.shuffle(indices)\n",
        "data = data[indices]\n",
        "labels = labels[indices]\n",
        "nb_validation_samples = int(VALIDATION_SPLIT * data.shape[0])\n"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "UM_jd89HCCTi"
      },
      "outputs": [],
      "source": [
        "X_train = data[:-nb_validation_samples]\n",
        "y_train = labels[:-nb_validation_samples]\n",
        "X_val = data[-nb_validation_samples:]\n",
        "y_val = labels[-nb_validation_samples:]"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "hMTwukfaCIat",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "66b85d1c-c0b6-4375-d21d-d8a7348d1762"
      },
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "\n",
            "Done!\n"
          ]
        }
      ],
      "source": [
        "embeddings_index = {}\n",
        "f = open(\"/content/glove.twitter.27B.200d.txt\", encoding=\"utf8\")\n",
        "for line in f:\n",
        "    values = line.split()\n",
        "    word = values[0]\n",
        "    embeddings_index[word] = np.asarray(values[1:], dtype='float32')\n",
        "f.close()\n",
        "embedding_matrix = np.random.random((len(word_index) + 1, EMBEDDING_DIM))\n",
        "for word, i in word_index.items():\n",
        "    embedding_vector = embeddings_index.get(word)\n",
        "    if embedding_vector is not None:\n",
        "        # words not found in embedding index will be all-zeros.\n",
        "        embedding_matrix[i] = embedding_vector\n",
        "print(\"\\nDone!\")"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "Ctgym6kwCOB_",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "fef25e16-7a76-4690-c5c2-c08d1a40ead1"
      },
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "done!\n"
          ]
        }
      ],
      "source": [
        "# second embedding matrix for non-static channel\n",
        "embedding_matrix_ns = np.random.random((len(word_index) + 1, EMBEDDING_DIM))\n",
        "for word, i in word_index.items():\n",
        "    embedding_vector = embeddings_index.get(word)\n",
        "    if embedding_vector is not None:\n",
        "        # words not found in embedding index will be all-zeros.\n",
        "        embedding_matrix_ns[i] = embedding_vector\n",
        "print(\"done!\")"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "tFAEzyoTDPHf"
      },
      "outputs": [],
      "source": [
        "sequence_input = Input(shape=(MAX_SEQUENCE_LENGTH,), dtype='int32')\n",
        "\n",
        "# static channel\n",
        "embedding_layer_frozen = Embedding(len(word_index) + 1, EMBEDDING_DIM, weights=[embedding_matrix], input_length=MAX_SEQUENCE_LENGTH, trainable=False)\n",
        "embedded_sequences_frozen = embedding_layer_frozen(sequence_input)\n",
        "\n",
        "# non-static channel\n",
        "embedding_layer_train = Embedding(len(word_index) + 1, EMBEDDING_DIM, weights=[embedding_matrix_ns], input_length=MAX_SEQUENCE_LENGTH, trainable=True)\n",
        "embedded_sequences_train = embedding_layer_train(sequence_input)"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "-lqknEcbDTCx"
      },
      "outputs": [],
      "source": [
        "l_lstm1f = Bidirectional(LSTM(6,return_sequences=True,dropout=0.3, recurrent_dropout=0.0))(embedded_sequences_frozen)\n",
        "l_lstm1t = Bidirectional(LSTM(6,return_sequences=True,dropout=0.3, recurrent_dropout=0.0))(embedded_sequences_train)\n",
        "l_lstm1 = Concatenate(axis=1)([l_lstm1f, l_lstm1t])\n",
        "\n",
        "l_conv_2 = Conv1D(filters=24,kernel_size=2,activation='relu')(l_lstm1)\n",
        "l_conv_2 = Dropout(0.3)(l_conv_2)\n",
        "l_conv_3 = Conv1D(filters=24,kernel_size=3,activation='relu')(l_lstm1)\n",
        "l_conv_3 = Dropout(0.3)(l_conv_3)\n",
        "\n",
        "l_conv_5 = Conv1D(filters=24,kernel_size=5,activation='relu',)(l_lstm1)\n",
        "l_conv_5 = Dropout(0.3)(l_conv_5)\n",
        "l_conv_6 = Conv1D(filters=24,kernel_size=6,activation='relu',kernel_regularizer=regularizers.l2(0.0001))(l_lstm1)\n",
        "l_conv_6 = Dropout(0.3)(l_conv_6)\n",
        "\n",
        "l_conv_8 = Conv1D(filters=24,kernel_size=8,activation='relu',kernel_regularizer=regularizers.l2(0.0001))(l_lstm1)\n",
        "l_conv_8 = Dropout(0.3)(l_conv_8)\n",
        "\n",
        "conv_1 = [l_conv_6,l_conv_5, l_conv_8,l_conv_2,l_conv_3]\n",
        "\n",
        "l_lstm_c = Concatenate(axis=1)(conv_1)\n",
        "\n",
        "l_conv_4f = Conv1D(filters=12,kernel_size=4,activation='relu',kernel_regularizer=regularizers.l2(0.0001))(embedded_sequences_frozen)\n",
        "l_conv_4f = Dropout(0.3)(l_conv_4f)\n",
        "l_conv_4t = Conv1D(filters=12,kernel_size=4,activation='relu',kernel_regularizer=regularizers.l2(0.0001))(embedded_sequences_train)\n",
        "l_conv_4t = Dropout(0.3)(l_conv_4t)\n",
        "\n",
        "l_conv_3f = Conv1D(filters=12,kernel_size=3,activation='relu',)(embedded_sequences_frozen)\n",
        "l_conv_3f = Dropout(0.3)(l_conv_3f)\n",
        "l_conv_3t = Conv1D(filters=12,kernel_size=3,activation='relu',)(embedded_sequences_train)\n",
        "l_conv_3t = Dropout(0.3)(l_conv_3t)\n",
        "\n",
        "l_conv_2f = Conv1D(filters=12,kernel_size=2,activation='relu')(embedded_sequences_frozen)\n",
        "l_conv_2f = Dropout(0.3)(l_conv_2f)\n",
        "l_conv_2t = Conv1D(filters=12,kernel_size=2,activation='relu')(embedded_sequences_train)\n",
        "l_conv_2t = Dropout(0.3)(l_conv_2t)\n",
        "\n",
        "conv_2 = [l_conv_4f, l_conv_4t,l_conv_3f, l_conv_3t, l_conv_2f, l_conv_2t]\n",
        "\n",
        "l_merge_2 = Concatenate(axis=1)(conv_2)\n",
        "l_c_lstm = Bidirectional(LSTM(12,return_sequences=True,dropout=0.3, recurrent_dropout=0.0))(l_merge_2)\n",
        "\n",
        "\n",
        "l_merge = Concatenate(axis=1)([l_lstm_c, l_c_lstm])\n",
        "l_pool = MaxPooling1D(4)(l_merge)\n",
        "l_drop = Dropout(0.5)(l_pool)\n",
        "l_flat = Flatten()(l_drop)\n",
        "l_dense = Dense(26, activation='relu')(l_flat)\n",
        "preds = Dense(5, activation='softmax')(l_dense)"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "-YHn4biqDj-c",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "48b10b2f-05df-4948-a70c-501cd3121e5d"
      },
      "outputs": [
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "/usr/local/lib/python3.7/dist-packages/keras/optimizer_v2/adadelta.py:74: UserWarning: The `lr` argument is deprecated, use `learning_rate` instead.\n",
            "  super(Adadelta, self).__init__(name, **kwargs)\n"
          ]
        }
      ],
      "source": [
        "model = Model(sequence_input, preds)\n",
        "adadelta = tensorflow.keras.optimizers.Adadelta(lr=0.9, rho=0.95, epsilon=None, decay=0.002)\n",
        "lr_metric = get_lr_metric(adadelta)\n",
        "model.compile(loss='categorical_crossentropy', optimizer=adadelta, metrics=['acc'])"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "Xb8wAwhUDqU8",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "c0e01c66-52a5-46ab-d807-c191b3ca6bd8"
      },
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "WARNING:tensorflow:`write_grads` will be ignored in TensorFlow 2.0 for the `TensorBoard` Callback.\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "WARNING:tensorflow:`write_grads` will be ignored in TensorFlow 2.0 for the `TensorBoard` Callback.\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "WARNING:tensorflow:`batch_size` is no longer needed in the `TensorBoard` Callback and will be ignored in TensorFlow 2.0.\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "WARNING:tensorflow:`batch_size` is no longer needed in the `TensorBoard` Callback and will be ignored in TensorFlow 2.0.\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "WARNING:tensorflow:`period` argument is deprecated. Please use `save_freq` to specify the frequency in number of batches seen.\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "WARNING:tensorflow:`period` argument is deprecated. Please use `save_freq` to specify the frequency in number of batches seen.\n"
          ]
        }
      ],
      "source": [
        "tensorboard = callbacks.TensorBoard(log_dir='./logs', histogram_freq=0, batch_size=16, write_grads=True , write_graph=True)\n",
        "model_checkpoints = ModelCheckpoint(\"checkpoint-{val_acc:.4f}.h5\", monitor='val_acc', verbose=0, save_best_only=True, save_weights_only=False, mode='auto', period=0)\n",
        "lr_schedule = callbacks.LearningRateScheduler(initial_boost)"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "FdGWxuI_EQjp"
      },
      "outputs": [],
      "source": [
        "model.save('BalanceNet.h5')"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "c1hpJWbmEXzB"
      },
      "outputs": [],
      "source": [
        "model = keras.models.load_model(\"/content/BalanceNet.h5\")"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "qqgOemzKEhFH",
        "outputId": "989f9c7c-9642-447a-bac2-f340a6c43f9e"
      },
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Epoch 1/10\n",
            "319/319 [==============================] - 109s 301ms/step - loss: 1.3775 - acc: 0.3933 - val_loss: 1.2901 - val_acc: 0.4602\n",
            "Epoch 2/10\n",
            "319/319 [==============================] - 91s 286ms/step - loss: 1.2598 - acc: 0.4685 - val_loss: 1.2275 - val_acc: 0.5018\n",
            "Epoch 3/10\n",
            "319/319 [==============================] - 90s 281ms/step - loss: 1.2217 - acc: 0.4916 - val_loss: 1.2173 - val_acc: 0.4937\n",
            "Epoch 4/10\n",
            "319/319 [==============================] - 90s 283ms/step - loss: 1.2015 - acc: 0.5072 - val_loss: 1.1962 - val_acc: 0.5116\n",
            "Epoch 5/10\n",
            "319/319 [==============================] - 91s 286ms/step - loss: 1.1883 - acc: 0.5090 - val_loss: 1.1750 - val_acc: 0.5179\n",
            "Epoch 6/10\n",
            "319/319 [==============================] - 90s 282ms/step - loss: 1.1778 - acc: 0.5150 - val_loss: 1.1749 - val_acc: 0.5234\n",
            "Epoch 7/10\n",
            "319/319 [==============================] - 90s 283ms/step - loss: 1.1700 - acc: 0.5172 - val_loss: 1.1727 - val_acc: 0.5193\n",
            "Epoch 8/10\n",
            "319/319 [==============================] - 89s 280ms/step - loss: 1.1633 - acc: 0.5217 - val_loss: 1.1644 - val_acc: 0.5228\n",
            "Epoch 9/10\n",
            "319/319 [==============================] - 90s 283ms/step - loss: 1.1557 - acc: 0.5251 - val_loss: 1.1616 - val_acc: 0.5255\n",
            "Epoch 10/10\n",
            "319/319 [==============================] - 90s 283ms/step - loss: 1.1523 - acc: 0.5268 - val_loss: 1.1625 - val_acc: 0.5254\n"
          ]
        }
      ],
      "source": [
        "model_log = model.fit(X_train, y_train, validation_data=(X_val, y_val), epochs=10, batch_size=100, callbacks=[tensorboard, model_checkpoints])\n",
        "pd.DataFrame(model_log.history).to_csv(\"history-balance.csv\")"
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "import matplotlib.pyplot as plt\n",
        "\n",
        "accuracy = model_log.history['acc']\n",
        "val_accuracy = model_log.history['val_acc']\n",
        "loss = model_log.history['loss']\n",
        "val_loss = model_log.history['val_loss']\n",
        "\n",
        "epochs = range(1, len(accuracy)+1)\n",
        "\n",
        "plt.plot(epochs, accuracy, 'g', label='Training accuracy')\n",
        "plt.plot(epochs, val_accuracy, 'r', label='Validation accuracy')\n",
        "plt.title('Training and validation accuracy')\n",
        "plt.legend()\n",
        "\n",
        "plt.figure()\n",
        "\n",
        "plt.plot(epochs, loss, 'g', label='Training loss')\n",
        "plt.plot(epochs, val_loss, 'r', label='Validation loss')\n",
        "plt.title('Training and validation loss')\n",
        "plt.legend()\n",
        "\n",
        "plt.show()"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 545
        },
        "id": "VbR-2i0lJUGF",
        "outputId": "49005422-a137-4a6a-b8d4-8ade5f8b1e5b"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {
            "needs_background": "light"
          }
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "from keras.models import load_model\n",
        "from sklearn.metrics import classification_report, confusion_matrix\n",
        "import matplotlib.pyplot as plt\n",
        "import numpy as np\n",
        "%config InlineBackend.figure_format = 'retina'\n",
        "import itertools, pickle\n",
        "\n",
        "with open('tokenizer.pickle', 'rb') as handle:\n",
        "    tokenizer = pickle.load(handle)\n",
        "\n",
        "classes = [\"Neutral\", \"Happy\", \"Sad\", \"Love\", \"Anger\"]"
      ],
      "metadata": {
        "id": "wXa1UJPvJVHG"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "model_test = load_model('/content/checkpoint-0.5255.h5')\n",
        "Y_test = np.argmax(y_val, axis=1) # Convert one-hot to index\n",
        "y_pred = model_test.predict(X_val)\n",
        "y_pred_class = np.argmax(y_pred,axis=1)\n",
        "cnf_matrix = confusion_matrix(Y_test, y_pred_class)"
      ],
      "metadata": {
        "id": "tDss0En0Jc0b"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "print(classification_report(Y_test, y_pred_class, target_names=classes))"
      ],
      "metadata": {
        "id": "-nmxWWtrJnnZ",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "fbfd0454-0999-4c9a-e697-3711e8541e39"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "              precision    recall  f1-score   support\n",
            "\n",
            "     Neutral       0.51      0.40      0.45      1957\n",
            "       Happy       0.50      0.54      0.52      2283\n",
            "         Sad       0.56      0.68      0.62      2730\n",
            "        Love       0.50      0.39      0.44       730\n",
            "       Anger       0.33      0.10      0.15       271\n",
            "\n",
            "    accuracy                           0.53      7971\n",
            "   macro avg       0.48      0.42      0.44      7971\n",
            "weighted avg       0.52      0.53      0.51      7971\n",
            "\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "def plot_confusion_matrix(cm, labels, normalize=True, title='Confusion Matrix (Validation Set)', cmap=plt.cm.Blues):\n",
        "    \"\"\"\n",
        "    This function prints and plots the confusion matrix.\n",
        "    Normalization can be applied by setting `normalize=True`.\n",
        "    \"\"\"\n",
        "    if normalize:\n",
        "        cm = cm.astype('float') / cm.sum(axis=1)[:, np.newaxis]\n",
        "        #print(\"Normalized confusion matrix\")\n",
        "    else:\n",
        "        #print('Confusion matrix, without normalization')\n",
        "        pass\n",
        "\n",
        "    #print(cm)\n",
        "\n",
        "    plt.imshow(cm, interpolation='nearest', cmap=cmap)\n",
        "    plt.title(title)\n",
        "    plt.colorbar()\n",
        "    tick_marks = np.arange(len(labels))\n",
        "    plt.xticks(tick_marks, labels, rotation=45)\n",
        "    plt.yticks(tick_marks, labels)\n",
        "\n",
        "    fmt = '.2f' if normalize else 'd'\n",
        "    thresh = cm.max() / 2.\n",
        "    for i, j in itertools.product(range(cm.shape[0]), range(cm.shape[1])):\n",
        "        plt.text(j, i, format(cm[i, j], fmt),\n",
        "                 horizontalalignment=\"center\",\n",
        "                 color=\"white\" if cm[i, j] > thresh else \"black\")\n",
        "\n",
        "    plt.tight_layout()\n",
        "    plt.ylabel('TRUE LABEL')\n",
        "    plt.xlabel('PPREDICTED LABEL')\n",
        "\n",
        "plt.figure(figsize=(10,6))\n",
        "plot_confusion_matrix(cnf_matrix, labels=classes)"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 454
        },
        "id": "BmQaQjLuwxoU",
        "outputId": "be2af7c1-8171-4a0b-9291-2b2cfbf0b9ee"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 720x432 with 2 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {
            "image/png": {
              "width": 501,
              "height": 437
            },
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "#Predicting emotions from paragraphs and sentences.\n",
        "\n",
        "text = [\"For instance, giving a kiss to your younger sibling daily after waking up in the morning and showing him how much you love them. For some happiness means loving life and seeing others happy. While some finds happiness in writing stories. Some conquer happiness in being simple yet the best person they can ever be. Everyone has their own unique way to feel happy by finding things that they never expected to find.\", # happy\n",
        "        \"Love is the key to happiness. We all want to lead a happy life. People look around for happiness in power, fashion, wealth, drugs etc. But these things can only give temporary pleasures. The power of love can create miracles. Love can create unity among nations and its citizens. Love is the most beautiful feeling in the world. Love has given different meaning by different people depending upon how they have experienced this wonderful feeling.\", # love\n",
        "        \"One day I was studying in my room when, all of a sudden, i heard hot words being exchanged between two persons in the street. I paid no attention, thinking it would be a minor quarrel but soon I heard the voices of a large number of people. I peeped from the window and saw that there was a street quarrel. I went downstairs and reached the spot in the twinkling of an eyes. I was at my wits end on seeing that both of them had come to blows. The people were standing around them and enjoying their quarrel but none tried to pacify them.\", # sad\n",
        "        \"I am so angry at you!!!!!\", # anger\n",
        "        \"you ve hit a new low with a danger of blm fascist slogan please stop it before too late stop\", # anger\n",
        "        \"I love my doggg\", # love\n",
        "        \"I think i'm gonna be sick :'‑(\", # sad\n",
        "        \"I hate you so much\", # anger\n",
        "        \"I'm at work\", # neutral\n",
        "        \"@TheTombert i was watching Harpers Island, lol... there was no vodka involved\", # neutral\n",
        "        \"sometimes i wish things could go back to the way they were the beginning of last summer\", # sad\n",
        "        \"it's your 18th birthday finally!!! yippeeeee\", # happy\n",
        "        \"still waiting in line\", # neutral\n",
        "        \"aarrgghh - fu*k.....a hose has leaked water all over the new floating floor\", # anger\n",
        "        \"that b*tch is so ugly\", # anger\n",
        "        \"oh no he is hospitalised!!!\", # sad\n",
        "        ]"
      ],
      "metadata": {
        "id": "Zk2qdPnvJuKC"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "sequences_test = tokenizer.texts_to_sequences(text)\n",
        "data_int_t = pad_sequences(sequences_test, padding='pre', maxlen=(MAX_SEQUENCE_LENGTH-5))\n",
        "data_test = pad_sequences(data_int_t, padding='post', maxlen=(MAX_SEQUENCE_LENGTH))\n",
        "y_prob = model.predict(data_test)\n",
        "for n, prediction in enumerate(y_prob):\n",
        "    pred = y_prob.argmax(axis=-1)[n]\n",
        "    print(text[n],\"\\nPREDICTION:\",classes[pred],\"\\n\")"
      ],
      "metadata": {
        "id": "SggLzM34J2w5",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "dce412ec-6b09-4587-e10d-c0d0d6735241"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "For instance, giving a kiss to your younger sibling daily after waking up in the morning and showing him how much you love them. For some happiness means loving life and seeing others happy. While some finds happiness in writing stories. Some conquer happiness in being simple yet the best person they can ever be. Everyone has their own unique way to feel happy by finding things that they never expected to find. \n",
            "PREDICTION: Happy \n",
            "\n",
            "Love is the key to happiness. We all want to lead a happy life. People look around for happiness in power, fashion, wealth, drugs etc. But these things can only give temporary pleasures. The power of love can create miracles. Love can create unity among nations and its citizens. Love is the most beautiful feeling in the world. Love has given different meaning by different people depending upon how they have experienced this wonderful feeling. \n",
            "PREDICTION: Love \n",
            "\n",
            "One day I was studying in my room when, all of a sudden, i heard hot words being exchanged between two persons in the street. I paid no attention, thinking it would be a minor quarrel but soon I heard the voices of a large number of people. I peeped from the window and saw that there was a street quarrel. I went downstairs and reached the spot in the twinkling of an eyes. I was at my wits end on seeing that both of them had come to blows. The people were standing around them and enjoying their quarrel but none tried to pacify them. \n",
            "PREDICTION: Sad \n",
            "\n",
            "I am so angry at you!!!!! \n",
            "PREDICTION: Sad \n",
            "\n",
            "you ve hit a new low with a danger of blm fascist slogan please stop it before too late stop \n",
            "PREDICTION: Sad \n",
            "\n",
            "I love my doggg \n",
            "PREDICTION: Love \n",
            "\n",
            "I think i'm gonna be sick :'‑( \n",
            "PREDICTION: Sad \n",
            "\n",
            "I hate you so much \n",
            "PREDICTION: Sad \n",
            "\n",
            "I'm at work \n",
            "PREDICTION: Neutral \n",
            "\n",
            "@TheTombert i was watching Harpers Island, lol... there was no vodka involved \n",
            "PREDICTION: Happy \n",
            "\n",
            "sometimes i wish things could go back to the way they were the beginning of last summer \n",
            "PREDICTION: Sad \n",
            "\n",
            "it's your 18th birthday finally!!! yippeeeee \n",
            "PREDICTION: Happy \n",
            "\n",
            "still waiting in line \n",
            "PREDICTION: Neutral \n",
            "\n",
            "aarrgghh - fu*k.....a hose has leaked water all over the new floating floor \n",
            "PREDICTION: Neutral \n",
            "\n",
            "that b*tch is so ugly \n",
            "PREDICTION: Anger \n",
            "\n",
            "oh no he is hospitalised!!! \n",
            "PREDICTION: Sad \n",
            "\n"
          ]
        }
      ]
    }
  ],
  "metadata": {
    "colab": {
      "name": "emotiondetection-deeplearning.ipynb",
      "provenance": []
    },
    "kernelspec": {
      "display_name": "Python 3",
      "name": "python3"
    },
    "language_info": {
      "name": "python"
    }
  },
  "nbformat": 4,
  "nbformat_minor": 0
}