prediction.ipynb

{
  "cells": [
    {
      "cell_type": "markdown",
      "source": [
        "# **Sentiment Prediction**"
      ],
      "metadata": {
        "nteract": {
          "transient": {
            "deleting": false
          }
        },
        "id": "1wXJn7YfkiD3"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "import numpy as np\n",
        "import pandas as pd\n",
        "import platform\n",
        "import configparser\n",
        "from datetime import datetime, timedelta, date\n",
        "import glob\n",
        "import sys\n",
        "import os\n",
        "import pickle\n",
        "import timeit\n",
        "from tqdm import tqdm\n",
        "\n",
        "import tensorflow as tf\n",
        "import matplotlib.pyplot as plt\n",
        "\n",
        "from tensorflow.keras.preprocessing.text import Tokenizer\n",
        "from tensorflow.keras.preprocessing.sequence import pad_sequences\n",
        "from tensorflow.keras.models import model_from_json\n",
        "from sklearn.metrics import accuracy_score, f1_score, roc_auc_score"
      ],
      "outputs": [],
      "execution_count": 2,
      "metadata": {
        "gather": {
          "logged": 1688553047177
        },
        "jupyter": {
          "outputs_hidden": true
        },
        "id": "_nsfCd9ukiEA"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "colab_active = 'google.colab' in sys.modules\n",
        "print(\"Colab Active: \", colab_active)"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "jcr3OisEfQQJ",
        "outputId": "baf997b4-07b2-46b8-e537-17e1867e12da"
      },
      "execution_count": 3,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Colab Active:  True\n"
          ]
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "#**Downloads**"
      ],
      "metadata": {
        "id": "pBkhH7a4lmGf"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "if colab_active is True:\n",
        "  OS = platform.system()      # Operating System\n",
        "\n",
        "  def data_download(file_to_download, gdrive_code, OS, uncompress = True):\n",
        "    if not os.path.exists(file_to_download):\n",
        "      os.system('gdown --id \"'+gdrive_code+'\" --output '+file_to_download)\n",
        "      if OS == \"Linux\" and uncompress:\n",
        "          os.system('unzip -o -n \"./'+file_to_download+'\" -d '+os.path.dirname(file_to_download))\n",
        "      return True\n",
        "    else:\n",
        "      return None"
      ],
      "metadata": {
        "id": "3mRTnsF3llCN"
      },
      "execution_count": 4,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "if colab_active is True:\n",
        "  out = data_download(\"./Senntiment_Social_Mood_Corpus_Tweets.zip\", \"11lopgQpkQ-UjwP62icelthzISG_fn2_5\", OS)\n",
        "  out = data_download(\"./Sentiment_Social_Mood_Models.zip\", \"1id8az7G6knkXKh7m8ENnKylj_s_W54YN\", OS)"
      ],
      "metadata": {
        "id": "1Clb3-sxlshP"
      },
      "execution_count": 5,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "with open('./Sentiment_Social_Mood_Models/preprocessed_datasets.pkl', 'rb') as file:\n",
        "    (x_test, y_test, tokenizer) = pickle.load(file)\n",
        "    file.close()"
      ],
      "metadata": {
        "id": "4_4KPxIFVGSC"
      },
      "execution_count": 6,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "source": [
        "## Configs"
      ],
      "metadata": {
        "nteract": {
          "transient": {
            "deleting": false
          }
        },
        "id": "4cDi-eeskiEE"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "start_date: date = date(2020,1,1)\n",
        "end_date  : date = date(2020,5,31)\n",
        "\n",
        "small_df  : bool = True\n",
        "small_size: int  = 6000000\n",
        "\n",
        "max_tokens: int  = 20000\n",
        "k_model   : str  = 'mod_orig_onestep'\n",
        "\n",
        "tpu = True\n",
        "batch_size = 2048"
      ],
      "outputs": [],
      "execution_count": 7,
      "metadata": {
        "jupyter": {
          "source_hidden": false,
          "outputs_hidden": false
        },
        "nteract": {
          "transient": {
            "deleting": false
          }
        },
        "gather": {
          "logged": 1688553051644
        },
        "id": "CRpWt3oNkiEF"
      }
    },
    {
      "cell_type": "markdown",
      "source": [
        "Create a list of dates ranging from `start_date` to `end_date`."
      ],
      "metadata": {
        "nteract": {
          "transient": {
            "deleting": false
          }
        },
        "id": "kkUkxG44kiEG"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "date_list = []\n",
        "\n",
        "current_date = start_date\n",
        "iteration = 1\n",
        "\n",
        "while current_date <= end_date:\n",
        "    date_list.append(f'{current_date.year}-{current_date.month:02d}-{current_date.day:02d}')\n",
        "    current_date += timedelta(days=1)\n",
        "    iteration += 1"
      ],
      "outputs": [],
      "execution_count": 8,
      "metadata": {
        "jupyter": {
          "source_hidden": false,
          "outputs_hidden": false
        },
        "nteract": {
          "transient": {
            "deleting": false
          }
        },
        "gather": {
          "logged": 1688553053503
        },
        "id": "2G_P-57GkiEH"
      }
    },
    {
      "cell_type": "markdown",
      "source": [
        "## Set the TPU Environment"
      ],
      "metadata": {
        "id": "SUNnGAXIZJBg"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "if tpu is True:\n",
        "  import tensorflow as tf\n",
        "  print(\"Tensorflow version \" + tf.__version__)\n",
        "\n",
        "  try:\n",
        "    tpu = tf.distribute.cluster_resolver.TPUClusterResolver()  # TPU detection\n",
        "    print('Running on TPU ', tpu.cluster_spec().as_dict()['worker'])\n",
        "  except ValueError:\n",
        "    raise BaseException('ERROR: Not connected to a TPU runtime; please see the previous cell in this notebook for instructions!')\n",
        "\n",
        "  tf.config.experimental_connect_to_cluster(tpu)\n",
        "  tf.tpu.experimental.initialize_tpu_system(tpu)\n",
        "  tpu_strategy = tf.distribute.experimental.TPUStrategy(tpu)\n",
        "\n",
        "  AUTO = tf.data.experimental.AUTOTUNE\n",
        "  batch_size = batch_size * tpu_strategy.num_replicas_in_sync"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "EBRB9xQTZJ1B",
        "outputId": "7c271f12-8efd-4ccb-fbdc-a28bb475f848"
      },
      "execution_count": 9,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Tensorflow version 2.12.0\n",
            "Running on TPU  ['10.104.170.130:8470']\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "WARNING:absl:`tf.distribute.experimental.TPUStrategy` is deprecated, please use  the non experimental symbol `tf.distribute.TPUStrategy` instead.\n"
          ]
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "<hr>"
      ],
      "metadata": {
        "nteract": {
          "transient": {
            "deleting": false
          }
        },
        "id": "mvkKe9M6kiEI"
      }
    },
    {
      "cell_type": "markdown",
      "source": [
        "## Data"
      ],
      "metadata": {
        "nteract": {
          "transient": {
            "deleting": false
          }
        },
        "id": "S8kyKbs7kiEI"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "dfs = []\n",
        "\n",
        "for date in date_list:\n",
        "    df_date = pd.read_csv(f'./Senntiment_Social_Mood_Corpus_Tweets/corpus_tweets/{date}.csv', sep=';', encoding='ISO-8859-1')\n",
        "    dfs.append(df_date)"
      ],
      "outputs": [],
      "execution_count": 10,
      "metadata": {
        "jupyter": {
          "source_hidden": false,
          "outputs_hidden": true
        },
        "nteract": {
          "transient": {
            "deleting": false
          }
        },
        "gather": {
          "logged": 1688553129996
        },
        "id": "f8acrFzGkiEL"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "raw_df = pd.concat(dfs, ignore_index=True)\n",
        "\n",
        "if small_df is True:\n",
        "    raw_df = raw_df.sample(small_size)"
      ],
      "outputs": [],
      "execution_count": 11,
      "metadata": {
        "jupyter": {
          "source_hidden": false,
          "outputs_hidden": false
        },
        "nteract": {
          "transient": {
            "deleting": false
          }
        },
        "gather": {
          "logged": 1688553138092
        },
        "id": "-laS97XMkiEM"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "len(raw_df)"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "Q50rrue0WaeC",
        "outputId": "b9ccf56d-fdd5-4080-b1b6-1926307c9cd7"
      },
      "execution_count": 12,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "6000000"
            ]
          },
          "metadata": {},
          "execution_count": 12
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "df = raw_df[['data_tweet', 'testo']]\n",
        "df['data_tweet'] = pd.to_datetime(df['data_tweet'], format = '%Y%m%d')"
      ],
      "outputs": [
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "<ipython-input-13-2b2346512ad4>:2: SettingWithCopyWarning: \n",
            "A value is trying to be set on a copy of a slice from a DataFrame.\n",
            "Try using .loc[row_indexer,col_indexer] = value instead\n",
            "\n",
            "See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy\n",
            "  df['data_tweet'] = pd.to_datetime(df['data_tweet'], format = '%Y%m%d')\n"
          ]
        }
      ],
      "execution_count": 13,
      "metadata": {
        "jupyter": {
          "source_hidden": false,
          "outputs_hidden": true
        },
        "nteract": {
          "transient": {
            "deleting": false
          }
        },
        "gather": {
          "logged": 1688553140588
        },
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "0Nk56CX6kiEO",
        "outputId": "4af9e88e-8355-4010-d1ff-3f4d92e28700"
      }
    },
    {
      "cell_type": "markdown",
      "source": [
        "remove tweets that are not type: `str`."
      ],
      "metadata": {
        "nteract": {
          "transient": {
            "deleting": false
          }
        },
        "id": "gptRSdLukiEO"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "df = df.loc[df['testo'].apply(lambda x: isinstance(x, str))]"
      ],
      "outputs": [],
      "execution_count": 14,
      "metadata": {
        "jupyter": {
          "source_hidden": false,
          "outputs_hidden": false
        },
        "nteract": {
          "transient": {
            "deleting": false
          }
        },
        "gather": {
          "logged": 1688553148585
        },
        "id": "Wuhb33VKkiEP"
      }
    },
    {
      "cell_type": "markdown",
      "source": [
        "## Pre-processing"
      ],
      "metadata": {
        "nteract": {
          "transient": {
            "deleting": false
          }
        },
        "id": "pTRJRpVjkiEQ"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "tweets = np.array(df['testo'])"
      ],
      "outputs": [],
      "execution_count": 15,
      "metadata": {
        "jupyter": {
          "source_hidden": false,
          "outputs_hidden": false
        },
        "nteract": {
          "transient": {
            "deleting": false
          }
        },
        "gather": {
          "logged": 1687770419126
        },
        "id": "G-2hpCbFkiER"
      }
    },
    {
      "cell_type": "markdown",
      "source": [
        "Transform words into tokens"
      ],
      "metadata": {
        "nteract": {
          "transient": {
            "deleting": false
          }
        },
        "id": "Odp0CrZRkiER"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "#tokenizer = Tokenizer(num_words = max_tokens)\n",
        "#tokenizer.fit_on_texts(tweets)"
      ],
      "outputs": [],
      "execution_count": 16,
      "metadata": {
        "jupyter": {
          "source_hidden": false,
          "outputs_hidden": false
        },
        "nteract": {
          "transient": {
            "deleting": false
          }
        },
        "gather": {
          "logged": 1687770803955
        },
        "id": "PeuLDP9mkiES"
      }
    },
    {
      "cell_type": "markdown",
      "source": [
        "Transform word sequences into indexed token sequences"
      ],
      "metadata": {
        "nteract": {
          "transient": {
            "deleting": false
          }
        },
        "id": "b6xDyRdHkiES"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "sequences = tokenizer.texts_to_sequences(tweets)"
      ],
      "outputs": [],
      "execution_count": 17,
      "metadata": {
        "jupyter": {
          "source_hidden": false,
          "outputs_hidden": false
        },
        "nteract": {
          "transient": {
            "deleting": false
          }
        },
        "gather": {
          "logged": 1687771099997
        },
        "id": "_r46Ss6akiET"
      }
    },
    {
      "cell_type": "markdown",
      "source": [
        "<hr>"
      ],
      "metadata": {
        "nteract": {
          "transient": {
            "deleting": false
          }
        },
        "id": "o3K9w19ckiET"
      }
    },
    {
      "cell_type": "markdown",
      "source": [
        "## Import Model"
      ],
      "metadata": {
        "nteract": {
          "transient": {
            "deleting": false
          }
        },
        "id": "GXOjBB9xkiEU"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "#model = tf.keras.models.load_model(f'models/{k_model}')\n",
        "#model = models.load_weights(f'sentiment_bilstm_best.hdf5')\n",
        "if tpu is True:\n",
        "  with strategy.scope():\n",
        "    json_file = open('./Sentiment_Social_Mood_Models/sentiment_bilstm_topology.json', 'r')\n",
        "    loaded_model_json = json_file.read()\n",
        "    json_file.close()\n",
        "    model = model_from_json(loaded_model_json)\n",
        "    model.load_weights(\"./Sentiment_Social_Mood_Models/sentiment_bilstm_best.hdf5\")\n",
        "else:\n",
        "  json_file = open('./Sentiment_Social_Mood_Models/sentiment_bilstm_topology.json', 'r')\n",
        "  loaded_model_json = json_file.read()\n",
        "  json_file.close()\n",
        "  model = model_from_json(loaded_model_json)\n",
        "  model.load_weights(\"./Sentiment_Social_Mood_Models/sentiment_bilstm_best.hdf5\")"
      ],
      "outputs": [],
      "execution_count": 18,
      "metadata": {
        "jupyter": {
          "source_hidden": false,
          "outputs_hidden": true
        },
        "nteract": {
          "transient": {
            "deleting": false
          }
        },
        "gather": {
          "logged": 1687771123290
        },
        "id": "gcTeAd-5kiEU"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "model.summary()"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "F8kxKVH6YJBy",
        "outputId": "0744d7a0-a4e7-4be8-b8e7-648c6f657c8e"
      },
      "execution_count": 19,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Model: \"model\"\n",
            "_________________________________________________________________\n",
            " Layer (type)                Output Shape              Param #   \n",
            "=================================================================\n",
            " input_1 (InputLayer)        [(None, None)]            0         \n",
            "                                                                 \n",
            " embedding (Embedding)       (None, None, 300)         6248100   \n",
            "                                                                 \n",
            " bidirectional (Bidirectiona  (None, None, 256)        439296    \n",
            " l)                                                              \n",
            "                                                                 \n",
            " bidirectional_1 (Bidirectio  (None, None, 64)         73984     \n",
            " nal)                                                            \n",
            "                                                                 \n",
            " global_max_pooling1d (Globa  (None, 64)               0         \n",
            " lMaxPooling1D)                                                  \n",
            "                                                                 \n",
            " dense (Dense)               (None, 1)                 65        \n",
            "                                                                 \n",
            "=================================================================\n",
            "Total params: 6,761,445\n",
            "Trainable params: 6,761,445\n",
            "Non-trainable params: 0\n",
            "_________________________________________________________________\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "del sequences"
      ],
      "metadata": {
        "id": "fmDHjB4Y8-tG"
      },
      "execution_count": 20,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "del tweets"
      ],
      "metadata": {
        "id": "E0R-6xUf9IAO"
      },
      "execution_count": 21,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "del raw_df"
      ],
      "metadata": {
        "id": "wZ9ATli19UCJ"
      },
      "execution_count": 22,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "test_tw = pad_sequences(tokenizer.texts_to_sequences(df['testo']), padding = \"pre\")"
      ],
      "outputs": [],
      "execution_count": 23,
      "metadata": {
        "jupyter": {
          "source_hidden": false,
          "outputs_hidden": false
        },
        "nteract": {
          "transient": {
            "deleting": false
          }
        },
        "gather": {
          "logged": 1687448884502
        },
        "id": "77PSDJ7jkiEV"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "len(test_tw)"
      ],
      "metadata": {
        "id": "GJYzxaRdYyXh",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "5486dc15-f461-48df-fdcb-635031c0349d"
      },
      "execution_count": 24,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "5999814"
            ]
          },
          "metadata": {},
          "execution_count": 24
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "## Predictions"
      ],
      "metadata": {
        "nteract": {
          "transient": {
            "deleting": false
          }
        },
        "id": "KWByGrq-kiEV"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "# Prediction of The Model\n",
        "pred_test = model.predict(x_test)\n",
        "pred_test = np.where(pred_test > 0.5, 1, 0)\n",
        "# Performance of The Model\n",
        "accuracy_test = accuracy_score(y_test.values, pred_test)\n",
        "auc_roc_score = roc_auc_score(y_test.values, pred_test)\n",
        "f1_score_test = f1_score(y_test.values, pred_test)\n",
        "print('\\nAccuracy of The Neural Model on Test Set : ', accuracy_test)\n",
        "print('ROC AUC Score of The Neural Model on Test Set', auc_roc_score)\n",
        "print('F1_score of The Neural Model on Test Set : ', f1_score_test)"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "8sZljrWyUvEQ",
        "outputId": "8cd8e706-8047-497c-d2c7-73c5045ae122"
      },
      "execution_count": 25,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "24/24 [==============================] - 3s 37ms/step\n",
            "\n",
            "Accuracy of The Neural Model on Test Set :  0.8104575163398693\n",
            "ROC AUC Score of The Neural Model on Test Set 0.7797582767181531\n",
            "F1_score of The Neural Model on Test Set :  0.7173489278752437\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "start_time = timeit.default_timer()\n",
        "sentiment = model.predict(test_tw, batch_size = batch_size, verbose = 1)\n",
        "print(\"Prediction Elapsed Time: \", timeit.default_timer() - start_time)"
      ],
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "367/367 [==============================] - 789s 2s/step\n",
            "Prediction Elapsed Time:  809.3803023100008\n"
          ]
        }
      ],
      "execution_count": 26,
      "metadata": {
        "jupyter": {
          "source_hidden": false,
          "outputs_hidden": false
        },
        "nteract": {
          "transient": {
            "deleting": false
          }
        },
        "gather": {
          "logged": 1687788549040
        },
        "id": "p5vNj0WYkiEW",
        "outputId": "cb4e5be8-8dec-4770-ec8f-930eb5dbd1fe",
        "colab": {
          "base_uri": "https://localhost:8080/"
        }
      }
    },
    {
      "cell_type": "code",
      "source": [
        "predicted_sentiment = [int(i > 0.5) for i in sentiment]\n",
        "df['predicted_sentiment'] = predicted_sentiment"
      ],
      "outputs": [
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "<ipython-input-27-dd630dee222d>:1: DeprecationWarning: Conversion of an array with ndim > 0 to a scalar is deprecated, and will error in future. Ensure you extract a single element from your array before performing this operation. (Deprecated NumPy 1.25.)\n",
            "  predicted_sentiment = [int(i > 0.5) for i in sentiment]\n"
          ]
        }
      ],
      "execution_count": 27,
      "metadata": {
        "jupyter": {
          "source_hidden": false,
          "outputs_hidden": true
        },
        "nteract": {
          "transient": {
            "deleting": false
          }
        },
        "gather": {
          "logged": 1687457711523
        },
        "id": "at0yILOCkiEW",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "3ca9cf60-75a9-44e2-e1ca-d94cf2baf905"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "len(predicted_sentiment)"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "k5wUxbcT0Dce",
        "outputId": "5149f9ed-e61b-449e-8ccd-18d5280ee6c8"
      },
      "execution_count": 28,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "5999814"
            ]
          },
          "metadata": {},
          "execution_count": 28
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "## Build the index"
      ],
      "metadata": {
        "nteract": {
          "transient": {
            "deleting": false
          }
        },
        "id": "HAIQuFWokiEX"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "df.set_index('data_tweet', inplace=True)"
      ],
      "outputs": [],
      "execution_count": 29,
      "metadata": {
        "jupyter": {
          "source_hidden": false,
          "outputs_hidden": false
        },
        "nteract": {
          "transient": {
            "deleting": false
          }
        },
        "id": "lNVgUe7IkiEX"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "count_by_day = df.groupby([df.index.date, 'predicted_sentiment']).size()\n",
        "count_by_day = count_by_day.reset_index()\n",
        "count_by_day.rename(columns={0: 'count', 'level_0': 'date'}, inplace=True)\n",
        "pivoted_counts = count_by_day.pivot(index='date', columns='predicted_sentiment', values='count')"
      ],
      "outputs": [],
      "execution_count": 30,
      "metadata": {
        "jupyter": {
          "source_hidden": false,
          "outputs_hidden": false
        },
        "nteract": {
          "transient": {
            "deleting": false
          }
        },
        "gather": {
          "logged": 1687446844215
        },
        "id": "qPPQFPFJkiEX"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "index = (pivoted_counts[1] - pivoted_counts[0]) / (pivoted_counts[1] + pivoted_counts[0])"
      ],
      "outputs": [],
      "execution_count": 35,
      "metadata": {
        "jupyter": {
          "source_hidden": false,
          "outputs_hidden": false
        },
        "nteract": {
          "transient": {
            "deleting": false
          }
        },
        "gather": {
          "logged": 1687446860529
        },
        "id": "sEA0b6N8kiEY"
      }
    },
    {
      "cell_type": "markdown",
      "source": [
        "Export results to csv."
      ],
      "metadata": {
        "nteract": {
          "transient": {
            "deleting": false
          }
        },
        "id": "Sh5xhOQekiEY"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "index.to_csv(f'predicted_{k_model}.csv', sep = ';', index=True)"
      ],
      "outputs": [],
      "execution_count": 32,
      "metadata": {
        "jupyter": {
          "source_hidden": false,
          "outputs_hidden": false
        },
        "nteract": {
          "transient": {
            "deleting": false
          }
        },
        "gather": {
          "logged": 1687457521297
        },
        "id": "3xOPApp3kiEY"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "index.to_csv(f'predicted_{k_model}.xlsx', sep = ';', index=True)"
      ],
      "metadata": {
        "id": "ZYb0dMCCLh_N"
      },
      "execution_count": 38,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "source": [
        "## Plots"
      ],
      "metadata": {
        "nteract": {
          "transient": {
            "deleting": false
          }
        },
        "id": "d2rDL2uqkiEZ"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "fig, ax = plt.subplots(figsize=(10,6))\n",
        "\n",
        "ax.plot(pivoted_counts.index, index)\n",
        "ax.set_xlabel('Time')\n",
        "ax.set_ylabel('Index')\n",
        "ax.set_title('Sentiment Index')\n",
        "plt.xticks(rotation=45)\n",
        "plt.savefig(\"predicted_mod_orig_onestep.png\")\n",
        "\n",
        "plt.show()"
      ],
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 1000x600 with 1 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {}
        }
      ],
      "execution_count": 37,
      "metadata": {
        "jupyter": {
          "source_hidden": false,
          "outputs_hidden": false
        },
        "nteract": {
          "transient": {
            "deleting": false
          }
        },
        "gather": {
          "logged": 1687792869070
        },
        "id": "aXf57yIhkiEZ",
        "outputId": "76236622-502a-400d-d8b5-e9b7006a270b",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 601
        }
      }
    },
    {
      "cell_type": "markdown",
      "source": [
        "## predictions_2020"
      ],
      "metadata": {
        "nteract": {
          "transient": {
            "deleting": false
          }
        },
        "id": "od9emoPfkiEa"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "with open('predictions_2020.txt', 'r') as file:\n",
        "    predictions_2020 = file.read()\n",
        "\n",
        "predictions_2020 = predictions_2020.replace(' ', '').split(',')"
      ],
      "outputs": [
        {
          "output_type": "error",
          "ename": "FileNotFoundError",
          "evalue": "ignored",
          "traceback": [
            "\u001b[0;31m---------------------------------------------------------------------------\u001b[0m",
            "\u001b[0;31mFileNotFoundError\u001b[0m                         Traceback (most recent call last)",
            "\u001b[0;32m<ipython-input-34-2463a8acf09a>\u001b[0m in \u001b[0;36m<cell line: 1>\u001b[0;34m()\u001b[0m\n\u001b[0;32m----> 1\u001b[0;31m \u001b[0;32mwith\u001b[0m \u001b[0mopen\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m'predictions_2020.txt'\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0;34m'r'\u001b[0m\u001b[0;34m)\u001b[0m \u001b[0;32mas\u001b[0m \u001b[0mfile\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m      2\u001b[0m     \u001b[0mpredictions_2020\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mfile\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mread\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m      3\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m      4\u001b[0m \u001b[0mpredictions_2020\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mpredictions_2020\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mreplace\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m' '\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0;34m''\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0msplit\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m','\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
            "\u001b[0;31mFileNotFoundError\u001b[0m: [Errno 2] No such file or directory: 'predictions_2020.txt'"
          ]
        }
      ],
      "execution_count": 34,
      "metadata": {
        "jupyter": {
          "source_hidden": false,
          "outputs_hidden": false
        },
        "nteract": {
          "transient": {
            "deleting": false
          }
        },
        "gather": {
          "logged": 1688553166748
        },
        "id": "pfi7q_vukiEa",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 219
        },
        "outputId": "5a3f03a5-3000-44dc-8c52-91ff6c567167"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "df['predicted_sentiment'] = predictions_2020[:-1]\n",
        "df['predicted_sentiment'] = df['predicted_sentiment'].astype(int)"
      ],
      "outputs": [],
      "execution_count": null,
      "metadata": {
        "jupyter": {
          "source_hidden": false,
          "outputs_hidden": false
        },
        "nteract": {
          "transient": {
            "deleting": false
          }
        },
        "gather": {
          "logged": 1688553169730
        },
        "id": "cQyr-85ikiEb"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "df.set_index('data_tweet', inplace=True)"
      ],
      "outputs": [],
      "execution_count": null,
      "metadata": {
        "jupyter": {
          "source_hidden": false,
          "outputs_hidden": false
        },
        "nteract": {
          "transient": {
            "deleting": false
          }
        },
        "gather": {
          "logged": 1688553170951
        },
        "id": "6oTfNKyHkiEc"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "count_by_day = df.groupby([df.index.date, 'predicted_sentiment']).size()\n",
        "count_by_day = count_by_day.reset_index()\n",
        "count_by_day.rename(columns={0: 'count', 'level_0': 'date'}, inplace=True)\n",
        "pivoted_counts = count_by_day.pivot(index='date', columns='predicted_sentiment', values='count')"
      ],
      "outputs": [],
      "execution_count": null,
      "metadata": {
        "jupyter": {
          "source_hidden": false,
          "outputs_hidden": false
        },
        "nteract": {
          "transient": {
            "deleting": false
          }
        },
        "gather": {
          "logged": 1688553176591
        },
        "id": "k7ttellEkiEc"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "index = (pivoted_counts[1] - pivoted_counts[0]) / (pivoted_counts[1] + pivoted_counts[0])"
      ],
      "outputs": [],
      "execution_count": null,
      "metadata": {
        "jupyter": {
          "source_hidden": false,
          "outputs_hidden": false
        },
        "nteract": {
          "transient": {
            "deleting": false
          }
        },
        "gather": {
          "logged": 1688553178379
        },
        "id": "azFWYmShkiEd"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "fig, ax = plt.subplots(figsize=(10,6))\n",
        "\n",
        "ax.plot(pivoted_counts.index, index)\n",
        "ax.set_xlabel('Time')\n",
        "ax.set_ylabel('Index')\n",
        "ax.set_title('Sentiment Index')\n",
        "plt.xticks(rotation=45)\n",
        "\n",
        "plt.show()"
      ],
      "outputs": [],
      "execution_count": null,
      "metadata": {
        "jupyter": {
          "source_hidden": false,
          "outputs_hidden": false
        },
        "nteract": {
          "transient": {
            "deleting": false
          }
        },
        "gather": {
          "logged": 1688553180999
        },
        "id": "tK5kZC5lkiEd"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "index.to_csv(f'predicted_2020.csv', sep = ';', index=True)"
      ],
      "outputs": [],
      "execution_count": null,
      "metadata": {
        "jupyter": {
          "source_hidden": false,
          "outputs_hidden": false
        },
        "nteract": {
          "transient": {
            "deleting": false
          }
        },
        "gather": {
          "logged": 1688553238330
        },
        "id": "tevq4R4ekiEe"
      }
    }
  ],
  "metadata": {
    "kernelspec": {
      "name": "python310-sdkv2",
      "language": "python",
      "display_name": "Python 3.10 - SDK v2"
    },
    "language_info": {
      "name": "python",
      "version": "3.10.11",
      "mimetype": "text/x-python",
      "codemirror_mode": {
        "name": "ipython",
        "version": 3
      },
      "pygments_lexer": "ipython3",
      "nbconvert_exporter": "python",
      "file_extension": ".py"
    },
    "microsoft": {
      "ms_spell_check": {
        "ms_spell_check_language": "en"
      },
      "host": {
        "AzureML": {
          "notebookHasBeenCompleted": true
        }
      }
    },
    "kernel_info": {
      "name": "python310-sdkv2"
    },
    "nteract": {
      "version": "nteract-front-end@1.0.0"
    },
    "colab": {
      "provenance": [],
      "gpuType": "T4"
    },
    "accelerator": "TPU"
  },
  "nbformat": 4,
  "nbformat_minor": 0
}