how_to_bertcnn_classification_malay_10fold.ipynb

{
  "nbformat": 4,
  "nbformat_minor": 0,
  "metadata": {
    "colab": {
      "name": "Copy of bert_classification_malay_10fold.ipynb",
      "provenance": [],
      "collapsed_sections": []
    },
    "kernelspec": {
      "name": "python3",
      "display_name": "Python 3"
    },
    "accelerator": "GPU"
  },
  "cells": [
    {
      "cell_type": "code",
      "metadata": {
        "id": "ny02hYzdvSpv"
      },
      "source": [
        "# !pip uninstall tensorflow\n",
        "# !pip install tensorflow==2.1.0\n",
        "# tf.compat.v1.disable_v2_behavior()\n",
        "# !pip install --upgrade gensim --user\n",
        "!pip install -q pyyaml h5py  # Required to save models in HDF5 format\n",
        "!pip install np_utils\n",
        "!pip install --upgrade --user keras\n",
        "import nltk\n",
        "nltk.download('punkt')\n",
        "!pip install bert-tensorflow\n",
        "# !pip install bert-for-tf2\n",
        "!pip install sentencepiece\n",
        "!pip install unidecode\n",
        "%tensorflow_version 1.x\n",
        "import tensorflow as tf\n",
        "print(tf.__version__)\n",
        "tf.test.gpu_device_name()"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "YMVyz3b-QEwr",
        "outputId": "13586fcb-eec1-42fc-8597-d1d2aba1f56d",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 36
        }
      },
      "source": [
        "from google.colab import drive\n",
        "drive.mount(\"/content/gdrive\")"
      ],
      "execution_count": 1,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Mounted at /content/gdrive\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "GHd8fAfjr3P5"
      },
      "source": [
        "import pandas as pd\n",
        "import json\n",
        "import os\n",
        "import glob\n",
        "import pickle\n",
        "#from os import path\n",
        "# import malaya"
      ],
      "execution_count": 3,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "Lg4fCE3lwWOJ",
        "outputId": "80219ed3-0932-46c9-fb61-846cd84eff83",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 424
        }
      },
      "source": [
        "df_allnews_unpickled = pd.read_pickle(\"./malaya_fake_news_preprocessed_dataframe.pkl\")        #Load FakeNews Dataset from your path\n",
        "df_allnews_unpickled"
      ],
      "execution_count": 4,
      "outputs": [
        {
          "output_type": "execute_result",
          "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>news</th>\n",
              "      <th>tokens</th>\n",
              "      <th>rejoined</th>\n",
              "      <th>length</th>\n",
              "      <th>label</th>\n",
              "      <th>real</th>\n",
              "      <th>fake</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>0</th>\n",
              "      <td>demi lovato telah jujur kejam tentang pertempu...</td>\n",
              "      <td>[demi, lovato, telah, jujur, kejam, tentang, p...</td>\n",
              "      <td>demi lovato telah jujur kejam tentang pertempu...</td>\n",
              "      <td>448</td>\n",
              "      <td>1</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>selamat pagi inilah yang perlu anda ketahui se...</td>\n",
              "      <td>[selamat, pagi, inilah, yang, perlu, anda, ket...</td>\n",
              "      <td>selamat pagi inilah yang perlu anda ketahui se...</td>\n",
              "      <td>682</td>\n",
              "      <td>1</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2</th>\n",
              "      <td>tahniah untuk bintang glee lea michele pelakon...</td>\n",
              "      <td>[tahniah, untuk, bintang, glee, lea, michele, ...</td>\n",
              "      <td>tahniah untuk bintang glee lea michele pelakon...</td>\n",
              "      <td>247</td>\n",
              "      <td>1</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3</th>\n",
              "      <td>ia adalah satu kejutan yang mengejutkan untuk ...</td>\n",
              "      <td>[ia, adalah, satu, kejutan, yang, mengejutkan,...</td>\n",
              "      <td>ia adalah satu kejutan yang mengejutkan untuk ...</td>\n",
              "      <td>718</td>\n",
              "      <td>1</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>4</th>\n",
              "      <td>transkrip minggu ini dengan george stephanopou...</td>\n",
              "      <td>[transkrip, minggu, ini, dengan, george, steph...</td>\n",
              "      <td>transkrip minggu ini dengan george stephanopou...</td>\n",
              "      <td>668</td>\n",
              "      <td>1</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>...</th>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>37587</th>\n",
              "      <td>katie holmes tidak mengandung anak jamie foxx ...</td>\n",
              "      <td>[katie, holmes, tidak, mengandung, anak, jamie...</td>\n",
              "      <td>katie holmes tidak mengandung anak jamie foxx ...</td>\n",
              "      <td>336</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>37588</th>\n",
              "      <td>donald trump menegaskan bahawa serangan siber ...</td>\n",
              "      <td>[donald, trump, menegaskan, bahawa, serangan, ...</td>\n",
              "      <td>donald trump menegaskan bahawa serangan siber ...</td>\n",
              "      <td>248</td>\n",
              "      <td>1</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>37589</th>\n",
              "      <td>ia bukan cerita pertama yang bermula dengan pe...</td>\n",
              "      <td>[ia, bukan, cerita, pertama, yang, bermula, de...</td>\n",
              "      <td>ia bukan cerita pertama yang bermula dengan pe...</td>\n",
              "      <td>663</td>\n",
              "      <td>1</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>37590</th>\n",
              "      <td>presiden donald trump mengumumkan bahawa dia t...</td>\n",
              "      <td>[presiden, donald, trump, mengumumkan, bahawa,...</td>\n",
              "      <td>presiden donald trump mengumumkan bahawa dia t...</td>\n",
              "      <td>100</td>\n",
              "      <td>1</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>37591</th>\n",
              "      <td>jika hillary clinton dipilih sebagai presiden ...</td>\n",
              "      <td>[jika, hillary, clinton, dipilih, sebagai, pre...</td>\n",
              "      <td>jika hillary clinton dipilih sebagai presiden ...</td>\n",
              "      <td>688</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "<p>37592 rows × 7 columns</p>\n",
              "</div>"
            ],
            "text/plain": [
              "                                                    news  ... fake\n",
              "0      demi lovato telah jujur kejam tentang pertempu...  ...    0\n",
              "1      selamat pagi inilah yang perlu anda ketahui se...  ...    0\n",
              "2      tahniah untuk bintang glee lea michele pelakon...  ...    0\n",
              "3      ia adalah satu kejutan yang mengejutkan untuk ...  ...    0\n",
              "4      transkrip minggu ini dengan george stephanopou...  ...    0\n",
              "...                                                  ...  ...  ...\n",
              "37587  katie holmes tidak mengandung anak jamie foxx ...  ...    1\n",
              "37588  donald trump menegaskan bahawa serangan siber ...  ...    0\n",
              "37589  ia bukan cerita pertama yang bermula dengan pe...  ...    0\n",
              "37590  presiden donald trump mengumumkan bahawa dia t...  ...    0\n",
              "37591  jika hillary clinton dipilih sebagai presiden ...  ...    1\n",
              "\n",
              "[37592 rows x 7 columns]"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 4
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "lx1TuO9GcMeH"
      },
      "source": [
        "import warnings\n",
        "warnings.filterwarnings(\"ignore\", category=DeprecationWarning) \n",
        "\n",
        "import tensorflow as tf\n",
        "from __future__ import division, print_function\n",
        "import gensim\n",
        "from gensim import models\n",
        "from tensorflow.keras.callbacks import ModelCheckpoint\n",
        "from tensorflow.keras.layers import Dense, Dropout, Reshape, Flatten, concatenate, Input, Conv1D, GlobalMaxPooling1D, Embedding, Bidirectional, LSTM, GlobalAveragePooling1D\n",
        "from tensorflow.keras.regularizers import l2\n",
        "from tensorflow.keras.regularizers import L1L2\n",
        "from tensorflow.keras.initializers import RandomNormal\n",
        "import tensorflow.keras.backend as K\n",
        "# from keras.layers.recurrent import LSTM\n",
        "# from tensorflow.keras.models import Sequential\n",
        "from tensorflow.keras.preprocessing.text import Tokenizer\n",
        "from tensorflow.keras.preprocessing.sequence import pad_sequences\n",
        "from tensorflow.keras.models import Model\n",
        "from sklearn.model_selection import train_test_split\n",
        "from sklearn.model_selection import KFold\n",
        "import numpy as np\n",
        "\n",
        "# demonstration of calculating metrics for a neural network model using sklearn\n",
        "from sklearn.metrics import accuracy_score\n",
        "from sklearn.metrics import precision_score\n",
        "from sklearn.metrics import average_precision_score\n",
        "from sklearn.metrics import recall_score\n",
        "from sklearn.metrics import precision_recall_curve\n",
        "from sklearn.metrics import f1_score\n",
        "from sklearn.metrics import confusion_matrix"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "JDDrcCkdwZ9u"
      },
      "source": [
        "import json\n",
        "import re\n",
        "import sentencepiece as spm\n",
        "import time"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "mqnrqH7HwcIx",
        "outputId": "9b096a91-591c-4bb7-b714-f22786f4fdd9",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 54
        }
      },
      "source": [
        "import bert\n",
        "from bert import run_classifier\n",
        "from bert import optimization\n",
        "from bert import tokenization\n",
        "from bert import modeling\n",
        "import numpy as np\n",
        "import json\n",
        "import tensorflow_hub as hub\n",
        "import itertools\n",
        "from unidecode import unidecode\n",
        "from tqdm import tqdm\n",
        "from collections import namedtuple"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "WARNING:tensorflow:From /usr/local/lib/python3.6/dist-packages/bert/optimization.py:87: The name tf.train.Optimizer is deprecated. Please use tf.compat.v1.train.Optimizer instead.\n",
            "\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "nv7ua0J9wexJ"
      },
      "source": [
        "from bertclassify import prepro_utils\n",
        "\n",
        "sp_model = spm.SentencePieceProcessor()\n",
        "sp_model.Load('./bert-small/sp10m.cased.v4.model')\n",
        "\n",
        "with open('./bert-small/sp10m.cased.v4.vocab') as fopen:\n",
        "    v = fopen.read().split('\\n')[:-1]\n",
        "v = [i.split('\\t') for i in v]\n",
        "v = {i[0]: i[1] for i in v}\n",
        "\n",
        "class Tokenizer:\n",
        "    def __init__(self, v):\n",
        "        self.vocab = v\n",
        "        pass\n",
        "    \n",
        "    def tokenize(self, string):\n",
        "        return prepro_utils.encode_pieces(sp_model, string, return_unicode=False, sample=False)\n",
        "    \n",
        "    def convert_tokens_to_ids(self, tokens):\n",
        "        return [sp_model.PieceToId(piece) for piece in tokens]\n",
        "    \n",
        "    def convert_ids_to_tokens(self, ids):\n",
        "        return [sp_model.IdToPiece(i) for i in ids]\n",
        "    \n",
        "tokenizer = Tokenizer(v)"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "jJEE6qxVFBZ3"
      },
      "source": [
        "BERT_INIT_CHKPNT = './bert-small/model.ckpt'\n",
        "BERT_CONFIG = './bert-small/bert_config.json'"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "AXJPq7zdFE62",
        "outputId": "22752f58-ceea-4c61-b85a-1522ba7a46bd",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 54
        }
      },
      "source": [
        "bert_config = modeling.BertConfig.from_json_file(BERT_CONFIG)"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "WARNING:tensorflow:From /usr/local/lib/python3.6/dist-packages/bert/modeling.py:93: The name tf.gfile.GFile is deprecated. Please use tf.io.gfile.GFile instead.\n",
            "\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "5HklPwYLFMTH"
      },
      "source": [
        "MAX_SEQ_LENGTH = 128\n",
        "tokenizer.tokenize(df_allnews_unpickled.iloc[0,2])"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "ELtdj-tFwrsr"
      },
      "source": [
        "from tqdm import tqdm\n",
        "\n",
        "input_ids, input_masks, segment_ids = [], [], []\n",
        "\n",
        "for text in tqdm(df_allnews_unpickled[\"rejoined\"].tolist()):\n",
        "    tokens_a = tokenizer.tokenize(text)\n",
        "    if len(tokens_a) > MAX_SEQ_LENGTH - 2:\n",
        "        tokens_a = tokens_a[:(MAX_SEQ_LENGTH - 2)]\n",
        "    tokens = [\"<cls>\"] + tokens_a + [\"<sep>\"]\n",
        "    segment_id = [0] * len(tokens)\n",
        "    input_id = tokenizer.convert_tokens_to_ids(tokens)\n",
        "    input_mask = [1] * len(input_id)\n",
        "    padding = [0] * (MAX_SEQ_LENGTH - len(input_id))\n",
        "    input_id += padding\n",
        "    input_mask += padding\n",
        "    segment_id += padding\n",
        "    \n",
        "    input_ids.append(input_id)\n",
        "    input_masks.append(input_mask)\n",
        "    segment_ids.append(segment_id)"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "gQtjYLtMwv2l"
      },
      "source": [
        "idx_temp = 142\n",
        "\n",
        "print (len(input_ids))\n",
        "print (len(input_masks))\n",
        "print (len(segment_ids))\n",
        "\n",
        "print (input_ids[idx_temp])\n",
        "print (input_masks[idx_temp])\n",
        "print (segment_ids[idx_temp])\n",
        "\n",
        "input_ids = np.asarray(input_ids, dtype=np.int32)\n",
        "input_masks = np.asarray(input_masks, dtype=np.int32)\n",
        "segment_ids = np.asarray(segment_ids, dtype=np.int32)"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "bsuDvKk6ooAx"
      },
      "source": [
        "label_names = ['real', 'fake']\n",
        "labels_list = df_allnews_unpickled[\"label\"].values\n",
        "# labels_list = df_allnews_unpickled[label_names].values\n",
        "# labels_list = sub_df_allnews[label_names].values\n",
        "# print (labels_list[42])\n",
        "# print (sequences_all[42])\n",
        "# # Zip back the news and labels into a temporary list\n",
        "all_inputs_zipped = list(zip(input_ids, input_masks, segment_ids))\n",
        "# all_newsnlabels = list(zip(input_ids, input_masks, segment_ids, labels_list))\n",
        "# print (all_newsnlabels[42][0])\n",
        "# print (all_newsnlabels[42][1])"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "761eTWfgw61M"
      },
      "source": [
        "# print (df_allnews_unpickled.iloc[42])\n",
        "# print (len(all_newsnlabels[42][0]))\n",
        "# print (len(all_newsnlabels[42][1]))\n",
        "# print (len(all_newsnlabels[42][2]))\n",
        "# print (all_newsnlabels[42][3])\n",
        "\n",
        "print (df_allnews_unpickled.iloc[42])\n",
        "print (len(all_inputs_zipped[42][0]))\n",
        "print (len(all_inputs_zipped[42][1]))\n",
        "print (len(all_inputs_zipped[42][2]))\n",
        "print (labels_list[42])"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "NHrro5cUMlTC"
      },
      "source": [
        "num_epoch = 2\n",
        "batch_size = 64\n",
        "warmup_proportion = 0.1\n",
        "num_train_steps = int(len(df_allnews_unpickled[\"rejoined\"].tolist()) / batch_size * num_epoch)\n",
        "num_warmup_steps = int(num_train_steps * warmup_proportion)"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "CBUPN9CJT9-H"
      },
      "source": [
        "Define BERTCNN Architecture"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "QSmaNykpSG6G"
      },
      "source": [
        "class FTBERT_Model:\n",
        "    def __init__(\n",
        "        self,\n",
        "        dimension_output,\n",
        "        learning_rate = 2e-5,\n",
        "    ):\n",
        "        self.X = tf.placeholder(tf.int32, [None, None])\n",
        "        self.input_masks = tf.placeholder(tf.int32, [None, None])\n",
        "        self.segment_ids = tf.placeholder(tf.int32, [None, None])\n",
        "        self.Y = tf.placeholder(tf.int32, [None])\n",
        "        # self.X = tf.keras.layers.Input(shape=(MAX_SEQ_LENGTH,), dtype=tf.int32,\n",
        "        #                                name=\"input_word_ids\")\n",
        "        # self.masks = tf.keras.layers.Input(shape=(MAX_SEQ_LENGTH,), dtype=tf.int32,\n",
        "        #                               name=\"input_mask\")\n",
        "        # self.segments = tf.keras.layers.Input(shape=(MAX_SEQ_LENGTH,), dtype=tf.int32,\n",
        "        #                             name=\"segment_ids\")\n",
        "        # self.X = Input(shape = (None, ), dtype = \"int32\")\n",
        "        # self.masks = Input(shape = (None, ), dtype = \"int32\")\n",
        "        # self.segments = Input(shape = (None, ), dtype = \"int32\")\n",
        "\n",
        "        \n",
        "        bert_model = modeling.BertModel(\n",
        "            config=bert_config,\n",
        "            is_training=False,\n",
        "            input_ids=self.X,\n",
        "            input_mask=self.input_masks,\n",
        "            token_type_ids=self.segment_ids,\n",
        "            use_one_hot_embeddings=False)\n",
        "        \n",
        "        ####################################\n",
        "        # bert_output = bert_model.get_pooled_output()\n",
        "        bert_output = bert_model.get_sequence_output()\n",
        "        ####################################\n",
        "        \n",
        "        ####################################\n",
        "        # conv_output = Conv1D(filters=10, kernel_size=3, strides=1, activation=\"relu\")(bert_output)\n",
        "        # pool_output = Lambda(lambda x: K.max(x, axis=1), output_shape=(100, ))(conv_output)\n",
        "        \n",
        "        convs = []\n",
        "        filter_sizes = [3,4,5]\n",
        "        for filter_size in filter_sizes:\n",
        "            l_conv = Conv1D(filters=100, kernel_size=filter_size, activation='relu')(bert_output)\n",
        "            l_pool = GlobalMaxPooling1D()(l_conv)\n",
        "            convs.append(l_pool)\n",
        "        l_merge = concatenate(convs, axis=1)\n",
        "\n",
        "        # l_pool = GlobalAveragePooling1D()(bert_output)\n",
        "        # lstm_output = LSTM(512, go_backwards=True, dropout=0.2)(bert_output)\n",
        "\n",
        "        # l_pool = GlobalAveragePooling1D()(bert_output)\n",
        "        # l_dense1 = Dense(6, activation='relu')(l_pool)\n",
        "        # l_dense2 = Dense(6, activation='relu')(l_dense1)\n",
        "        # l_dense3 = Dense(6, activation='relu')(l_dense2)\n",
        "        # l_dense4 = Dense(6, activation='relu')(l_dense3)\n",
        "        # l_dense5 = Dense(6, activation='relu')(l_dense4)\n",
        "        ####################################\n",
        "\n",
        "        ####################################\n",
        "        # x3 = Dropout(0.5)(bert_output)\n",
        "        # x3 = Dropout(0.5)(pool_output)  \n",
        "        x3 = Dropout(0.5)(l_merge)  \n",
        "        # x3 = Dropout(0.5)(lstm_output)  \n",
        "        # x3 = Dropout(0.5)(l_dense5)  \n",
        "        ####################################  \n",
        "        \n",
        "        self.logits = Dense(len(list(label_names)), activation='softmax')(x3)\n",
        "\n",
        "\n",
        "        self.cost = tf.reduce_mean(\n",
        "            tf.nn.sparse_softmax_cross_entropy_with_logits(\n",
        "                logits = self.logits, labels = self.Y\n",
        "            )\n",
        "        )\n",
        "        \n",
        "        self.optimizer = optimization.create_optimizer(self.cost, learning_rate, \n",
        "                                                       num_train_steps, num_warmup_steps, False)\n",
        "        correct_pred = tf.equal(\n",
        "            tf.argmax(self.logits, 1, output_type = tf.int32), self.Y\n",
        "        )\n",
        "        self.accuracy = tf.reduce_mean(tf.cast(correct_pred, tf.float32))"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "XFCHz7UuNnWq"
      },
      "source": [
        "dimension_output = 2\n",
        "learning_rate = 2e-5\n",
        "\n",
        "tf.reset_default_graph()\n",
        "sess = tf.InteractiveSession()\n",
        "# tf.initialize_all_variables().run()\n",
        "model = FTBERT_Model(\n",
        "    dimension_output,\n",
        "    learning_rate\n",
        ")\n",
        "\n",
        "# model = FTBERT(len(list(label_names)))\n",
        "\n",
        "sess.run(tf.global_variables_initializer())\n",
        "var_lists = tf.get_collection(tf.GraphKeys.TRAINABLE_VARIABLES, scope = 'bert')\n",
        "saver = tf.train.Saver(var_list = var_lists)\n",
        "saver.restore(sess, BERT_INIT_CHKPNT)\n"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "AokXV_O313UT"
      },
      "source": [
        "# # Only run once before starting a new InteractiveSession() from above Code Section\n",
        "sess.close()"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "9d7PiusXNzTb"
      },
      "source": [
        "saver = tf.train.Saver()"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "wc35nr9dxk4J"
      },
      "source": [
        "# # model.save_weights('./bertclassify/bert_model.h5')\n",
        "save_path = saver.save(sess, \"./bertclassify/bert_training_10fold/bert_model.ckpt\")"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "WyWNU5_Ou8ga",
        "outputId": "0b2488e8-13fa-438a-cc83-e76807dfb841",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 36
        }
      },
      "source": [
        "# # model.load_weights('./bertclassify/bert_model.h5')\n",
        "saver.restore(sess, \"./bertclassify/bert_training_10fold/bert_model.ckpt\")"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "INFO:tensorflow:Restoring parameters from ./bertclassify/bert_training_10fold/bert_model.ckpt\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "-kI0jPjMvkwz"
      },
      "source": [
        "# weights = []\n",
        "# for layer in model.layers:\n",
        "#     weights.append(layer.get_weights()) # list of numpy arrays\n",
        "\n",
        "# print ((weights[6][0]))\n",
        "\n",
        "vars = tf.trainable_variables()\n",
        "# print(vars)\n",
        "\n",
        "var = [v for v in tf.trainable_variables() if v.name == \"dense_19/kernel:0\"][0]\n",
        "# print(var)\n",
        "print(\"Weight matrix: \\n{0}\".format(sess.run(var))) "
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "jFEdktxoSD23"
      },
      "source": [
        "K-fold Cross Validation stars here"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "LD3nbtcKh6qb"
      },
      "source": [
        "num_folds = 10\n",
        "# num_epochs = 3    \n",
        "# batch_size = 50\n",
        "fold_no = 1\n",
        "kfold = KFold(n_splits=num_folds, shuffle=False)"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "lcPpQlkgykjn"
      },
      "source": [
        "# data_train, data_test = train_test_split(all_newsnlabels, test_size=0.10, random_state=42)\n",
        "# inputs = [input_ids, input_masks, segment_ids]\n",
        "inputs = np.asarray(all_inputs_zipped)\n",
        "targets = labels_list\n",
        "print (len(inputs))\n",
        "print (len(targets))\n",
        "\n",
        "arr_id = np.asarray(df_allnews_unpickled.index)\n",
        "print (len(arr_id))\n",
        "\n",
        "# x_train = np.asarray([row[0] for row in data_train])\n",
        "# y_train = np.asarray([row[1] for row in data_train])\n",
        "# x_test = np.asarray([row[0] for row in data_test])\n",
        "# y_test = np.asarray([row[1] for row in data_test])\n",
        "# print (x_train[42])\n",
        "# print (y_train[42])\n",
        "# print (inputs[39505])\n",
        "# print (targets[39505])"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "qBCrlwUudN-S",
        "outputId": "537dff7e-890d-4c5c-838a-5a45f76e0166",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 1000
        }
      },
      "source": [
        "fold_no = 1\n",
        "results_history = []\n",
        "#[0] list_train_acc\n",
        "#[1] list_val_acc\n",
        "#[2] list_train_loss\n",
        "#[3] list_val_loss\n",
        "\n",
        "acc_per_fold = []\n",
        "loss_per_fold = []\n",
        "\n",
        "# # Lists for metrics result\n",
        "all_accuracy = []\n",
        "all_precision = []\n",
        "all_recall = []\n",
        "all_f1 = []\n",
        "\n",
        "# # Lists for plotting Precision-Recall Curve\n",
        "all_arr_precision = []\n",
        "all_arr_recall = []\n",
        "all_thresholds = []\n",
        "all_ave_precision = []\n",
        "\n",
        "# # Lists for Confusion Matrices\n",
        "all_cf_matrices = []\n",
        "\n",
        "# # Lists for Noise Data\n",
        "all_wrong = []\n",
        "\n",
        "for train, test in kfold.split(inputs, targets):\n",
        "  saver.restore(sess, \"./bertclassify/bert_training_10fold/bert_model.ckpt\")     #Load the same initial weights for all splits\n",
        "  x_train = np.asarray(inputs[train])\n",
        "  y_train = np.asarray(targets[train])\n",
        "  x_test = np.asarray(inputs[test])\n",
        "  # y_test = np.asarray(targets[test])\n",
        "  \n",
        "  temp_arr_id = np.asarray(arr_id[test])\n",
        "\n",
        "  # data_train_val = list(zip(x_train, y_train))      #Rezip to manually split train and validation\n",
        "  # data_train_val = np.column_stack(x_train,y_train)   #Stack columns to manually split train and validation\n",
        "  data_train_val = []\n",
        "  for m in range(len(x_train)):\n",
        "    temp_list = x_train[m].tolist()\n",
        "    temp_list.append(y_train[m].tolist())\n",
        "    data_train_val.append(temp_list)\n",
        "\n",
        "  data_train, data_val = train_test_split(data_train_val, test_size=0.10, random_state=42)\n",
        "  \n",
        "  x_train_ids = np.asarray([row[0] for row in data_train_val])\n",
        "  x_train_masks = np.asarray([row[1] for row in data_train_val])\n",
        "  x_train_segments = np.asarray([row[2] for row in data_train_val])\n",
        "  y_train = np.asarray([row[3] for row in data_train_val])\n",
        "\n",
        "  # x_train_ids = np.asarray([row[0] for row in data_train])\n",
        "  # x_train_masks = np.asarray([row[1] for row in data_train])\n",
        "  # x_train_segments = np.asarray([row[2] for row in data_train])\n",
        "  # y_train = np.asarray([row[3] for row in data_train])\n",
        "\n",
        "  x_val_ids = np.asarray([row[0] for row in data_val])\n",
        "  x_val_masks = np.asarray([row[1] for row in data_val])\n",
        "  x_val_segments = np.asarray([row[2] for row in data_val])\n",
        "  y_val = np.asarray([row[3] for row in data_val])\n",
        "\n",
        "  x_test_ids = np.asarray([row[0] for row in x_test.tolist()])\n",
        "  x_test_masks = np.asarray([row[1] for row in x_test.tolist()])\n",
        "  x_test_segments = np.asarray([row[2] for row in x_test.tolist()])\n",
        "  y_test = np.asarray(targets[test])\n",
        "\n",
        "\n",
        "  # # CHECKING FOR ALL TRAIN/VALIDATE/TEST ITEMS starts\n",
        "  # print (len(x_train_ids))\n",
        "  # print (len(x_train_masks))\n",
        "  # print (len(x_train_segments))\n",
        "  # print (len(y_train))\n",
        "  # print()\n",
        "\n",
        "  # print (len(x_val_ids))\n",
        "  # print (len(x_val_masks))\n",
        "  # print (len(x_val_segments))\n",
        "  # print (len(y_val))\n",
        "  # print()\n",
        "\n",
        "  # print (len(x_test_ids))\n",
        "  # print (len(x_test_masks))\n",
        "  # print (len(x_test_segments))\n",
        "  # print (len(y_test))\n",
        "  # print()\n",
        "\n",
        "  # id_tmp = 19\n",
        "\n",
        "  # print ((x_train_ids[id_tmp]))\n",
        "  # print ((x_train_masks[id_tmp]))\n",
        "  # print ((x_train_segments[id_tmp]))\n",
        "  # print ((y_train[id_tmp]))\n",
        "  # print()\n",
        "\n",
        "  # print ((x_val_ids[id_tmp]))\n",
        "  # print ((x_val_masks[id_tmp]))\n",
        "  # print ((x_val_segments[id_tmp]))\n",
        "  # print ((y_val[id_tmp]))\n",
        "  # print()\n",
        "\n",
        "  # print ((x_test_ids[id_tmp]))\n",
        "  # print ((x_test_masks[id_tmp]))\n",
        "  # print ((x_test_segments[id_tmp]))\n",
        "  # print ((y_test[id_tmp]))\n",
        "  # print()\n",
        "  # # CHECKING FOR ALL TRAIN/VALIDATE/TEST ITEMS ends\n",
        "  \n",
        "  \n",
        "  # # Generate a print\n",
        "  print('------------------------------------------------------------------------')\n",
        "  print(f'Training for Fold {fold_no} ...')\n",
        "  \n",
        "  \n",
        "  list_train_loss = []\n",
        "  list_train_acc = []\n",
        "  list_val_loss = []\n",
        "  list_val_acc = []\n",
        "\n",
        "  #########################################\n",
        "  # # manual work for model.fit() starts here\n",
        "  # TRAIN AND VALIDATE\n",
        "  for m in range(num_epoch):\n",
        "      lasttime = time.time()\n",
        "\n",
        "      train_acc, train_loss, val_acc, val_loss = 0, 0, 0, 0\n",
        "      pbar_train = tqdm(range(0, len(x_train_ids), batch_size), desc = 'Training Minibatch Loop', position=0, leave=True)\n",
        "      for i in pbar_train:\n",
        "          index = min(i + batch_size, len(x_train_ids))\n",
        "          batch_x = x_train_ids[i: index]\n",
        "          batch_masks = x_train_masks[i: index]\n",
        "          batch_segment = x_train_segments[i: index]\n",
        "          batch_y = y_train[i: index]\n",
        "          acc, cost, _ = sess.run(\n",
        "              [model.accuracy, model.cost, model.optimizer],\n",
        "              feed_dict = {\n",
        "                  model.Y: batch_y,\n",
        "                  model.X: batch_x,\n",
        "                  model.segment_ids: batch_segment,\n",
        "                  model.input_masks: batch_masks\n",
        "              },\n",
        "          )\n",
        "          assert not np.isnan(cost)\n",
        "          train_loss += cost\n",
        "          train_acc += acc\n",
        "          pbar_train.set_postfix(cost = cost, accuracy = acc)\n",
        "      \n",
        "      pbar_val = tqdm(range(0, len(x_test_ids), batch_size), desc = 'Validation Minibatch Loop', position=0, leave=True)      #val/test\n",
        "      for i in pbar_val:\n",
        "          index = min(i + batch_size, len(x_test_ids))        #val/test \n",
        "          batch_x = x_test_ids[i: index]                      #val/test\n",
        "          batch_masks = x_test_masks[i: index]                #val/test\n",
        "          batch_segment = x_test_segments[i: index]           #val/test\n",
        "          batch_y = y_test[i: index]                          #val/test\n",
        "          acc, cost = sess.run(\n",
        "              [model.accuracy, model.cost],\n",
        "              feed_dict = {\n",
        "                  model.Y: batch_y,\n",
        "                  model.X: batch_x,\n",
        "                  model.segment_ids: batch_segment,\n",
        "                  model.input_masks: batch_masks\n",
        "              },\n",
        "          )\n",
        "          val_loss += cost\n",
        "          val_acc += acc\n",
        "          pbar_val.set_postfix(cost = cost, accuracy = acc)\n",
        "\n",
        "      \n",
        "      train_loss /= len(pbar_train)\n",
        "      train_acc /= len(pbar_train)\n",
        "      val_loss /= len(pbar_val) \n",
        "      val_acc /= len(pbar_val)\n",
        "      \n",
        "      list_train_loss.append(train_loss)\n",
        "      list_train_acc.append(train_acc)\n",
        "      list_val_loss.append(val_loss)\n",
        "      list_val_acc.append(val_acc)\n",
        "          \n",
        "      print('Time Taken:', time.time() - lasttime)\n",
        "      print('Epoch: %d, train_loss: %f, train_acc: %f, val_loss: %f, val_acc: %f\\n'\n",
        "          % (m+1, train_loss, train_acc, val_loss, val_acc)\n",
        "      )\n",
        "\n",
        "  # results_history.append(hist)\n",
        "  # results_history.append(hist.history)\n",
        "  templist = list(zip(list_train_acc, list_val_acc, list_train_loss, list_val_loss))\n",
        "  results_history.append(templist)\n",
        "  templist = []\n",
        "  # # manual work for model.fit() ends here\n",
        "  #########################################\n",
        "    \n",
        "  \n",
        "\n",
        "  #########################################\n",
        "  # # manual work for model.evaluate() starts here\n",
        "  # EVALUATE MODEL ON TEST SET\n",
        "  test_acc, test_loss = 0, 0\n",
        "\n",
        "  pbar_test = tqdm(range(0, len(x_test_ids), batch_size), desc = 'Testing Minibatch Loop', position=0, leave=True)\n",
        "  for i in pbar_test:\n",
        "      index = min(i + batch_size, len(x_test_ids))\n",
        "      batch_x = x_test_ids[i: index]\n",
        "      batch_masks = x_test_masks[i: index]\n",
        "      batch_segment = x_test_segments[i: index]\n",
        "      batch_y = y_test[i: index]\n",
        "      acc, cost = sess.run([model.accuracy, model.cost],\n",
        "            feed_dict = {\n",
        "                model.Y: batch_y,\n",
        "                model.X: batch_x,\n",
        "                model.segment_ids: batch_segment,\n",
        "                model.input_masks: batch_masks\n",
        "            },\n",
        "        )\n",
        "      test_loss += cost\n",
        "      test_acc += acc\n",
        "      pbar_test.set_postfix(cost = cost, accuracy = acc)                    \n",
        "      \n",
        "  test_loss /= len(pbar_test) \n",
        "  test_acc /= len(pbar_test)\n",
        "  print('Time Taken:', time.time() - lasttime)\n",
        "  print('test_loss: %f, test_acc: %f\\n'\n",
        "      % (test_loss, test_acc)\n",
        "  )\n",
        "  \n",
        "  print(f'Score for Fold {fold_no}: Loss of {test_loss}; Accuracy of {test_acc*100}%')\n",
        "  acc_per_fold.append(test_acc * 100)\n",
        "  loss_per_fold.append(test_loss)\n",
        "  # # manual work for model.evaluate() ends here\n",
        "  #########################################\n",
        "\n",
        "\n",
        "\n",
        "  #########################################\n",
        "  # # manual work for model.predict() starts here\n",
        "  # METRICS REQUIREMENTS CALCULATION STARTS HERE\n",
        "  yhat_probs_all = []       # yhat_probs = non-argmax/min predict_Y\n",
        "  y_probs = []              # y_probs = yhat_probs[:, 0]  #Predicted Probabilities\n",
        "  yhat_classes = []         # yhat_classes = predict_Y              #Predicted Classes\n",
        "  y_test_conv = []          # y_test_conv => y_test = real_Y        #Actual Classes\n",
        "\n",
        "  pbar_test_metrics = tqdm(range(0, len(x_test_ids), batch_size), desc = 'Testing for Metrics Minibatch Loop', position=0, leave=True)\n",
        "  for i in pbar_test_metrics:\n",
        "      index = min(i + batch_size, len(x_test_ids))\n",
        "      batch_x = x_test_ids[i: index]\n",
        "      batch_masks = x_test_masks[i: index]\n",
        "      batch_segment = x_test_segments[i: index]\n",
        "      batch_y = y_test[i: index]\n",
        "      yhat_probs = sess.run(model.logits,\n",
        "              feed_dict = {\n",
        "                  model.Y: batch_y,\n",
        "                  model.X: batch_x,\n",
        "                  model.segment_ids: batch_segment,\n",
        "                  model.input_masks: batch_masks\n",
        "              },\n",
        "            )\n",
        "      # yhat_probs_all.append(yhat_probs)\n",
        "      # y_probs = np.append(y_probs, (np.max(yhat_probs, axis=1)))\n",
        "      y_probs = np.append(y_probs, (np.asarray(yhat_probs[:, 0])))     #Consider only positive-predicted class for Precision Recall Curve\n",
        "      yhat_classes = np.append(yhat_classes, (np.argmax(yhat_probs, axis=-1)))            #formerly predict_Y\n",
        "      y_test_conv = np.append(y_test_conv, batch_y)                                       #formerly real_Y\n",
        "  \n",
        "  # # manual work for model.predict() ends here\n",
        "  #########################################\n",
        "  \n",
        "  for m in range(len(yhat_classes)):\n",
        "    if (yhat_classes[m] != y_test_conv[m]):\n",
        "      all_wrong.append(temp_arr_id[m])   \n",
        "\n",
        "\n",
        "  ## Metrics calculation\n",
        "  # accuracy: (tp + tn) / (p + n)\n",
        "  accuracy = accuracy_score(y_test_conv, yhat_classes)\n",
        "  all_accuracy.append(accuracy)\n",
        "  # precision tp / (tp + fp)\n",
        "  precision = precision_score(y_test_conv, yhat_classes)\n",
        "  all_precision.append(precision)\n",
        "  # recall: tp / (tp + fn)\n",
        "  recall = recall_score(y_test_conv, yhat_classes)\n",
        "  all_recall.append(recall)\n",
        "  # f1: 2 tp / (2 tp + fp + fn)\n",
        "  f1 = f1_score(y_test_conv, yhat_classes)\n",
        "  all_f1.append(f1)\n",
        "\n",
        "  ## Precision-Curve calculation\n",
        "  arr_precision, arr_recall, thresholds = precision_recall_curve(y_test_conv, y_probs)\n",
        "  thresholds = np.concatenate([thresholds,[thresholds[thresholds.size-1]]])\n",
        "  all_arr_precision.append(arr_precision)\n",
        "  all_arr_recall.append(arr_recall)\n",
        "  all_thresholds.append(thresholds)\n",
        "\n",
        "  ave_precision = average_precision_score(y_test_conv, y_probs)\n",
        "  all_ave_precision.append(ave_precision)\n",
        "\n",
        "  ## Confusion Matrix calculation\n",
        "  cf_matrix = confusion_matrix(y_test_conv, yhat_classes)\n",
        "  all_cf_matrices.append(cf_matrix)\n",
        "\n",
        "  # Increase fold number\n",
        "  fold_no = fold_no + 1\n",
        "####################################\n",
        "\n",
        "# == Provide average scores ==\n",
        "print('------------------------------------------------------------------------')\n",
        "print('Score per fold')\n",
        "for i in range(0, len(acc_per_fold)):\n",
        "  print('------------------------------------------------------------------------')\n",
        "  print(f'> Fold {i+1} - Loss: {loss_per_fold[i]} - Accuracy: {acc_per_fold[i]}%')\n",
        "print('------------------------------------------------------------------------')\n",
        "print('Average scores for all folds:')\n",
        "print(f'> Accuracy: {np.mean(acc_per_fold)} (+- {np.std(acc_per_fold)})')\n",
        "print(f'> Loss: {np.mean(loss_per_fold)}')\n",
        "print('------------------------------------------------------------------------')\n",
        "\n",
        "\n",
        "\n",
        "# # Create Dataframe to save all train/val history result\n",
        "df_history = pd.DataFrame(results_history)\n",
        "df_history.to_pickle(\"./bertclassify/BERTTrainHistoryDict_10fold.pkl\")\n",
        "# df_history\n",
        "\n",
        "# # Create Dataframe to save all Metrics Result\n",
        "df_allmetrics_10fold = pd.DataFrame(acc_per_fold)\n",
        "df_allmetrics_10fold.columns = ['acc_per_fold']\n",
        "df_allmetrics_10fold['loss_per_fold'] = loss_per_fold\n",
        "\n",
        "df_allmetrics_10fold['all_accuracy'] = all_accuracy\n",
        "df_allmetrics_10fold['all_precision'] = all_precision\n",
        "df_allmetrics_10fold['all_recall'] = all_recall\n",
        "df_allmetrics_10fold['all_f1'] = all_f1\n",
        "\n",
        "df_allmetrics_10fold['all_arr_precision'] = all_arr_precision\n",
        "df_allmetrics_10fold['all_arr_recall'] = all_arr_recall\n",
        "df_allmetrics_10fold['all_thresholds'] = all_thresholds\n",
        "df_allmetrics_10fold['all_ave_precision'] = all_ave_precision\n",
        "\n",
        "df_allmetrics_10fold['all_cf_matrices'] = all_cf_matrices\n",
        "\n",
        "df_allmetrics_10fold.to_pickle(\"./bertclassify/bert_all_metrics_10fold.pkl\")\n",
        "# df_allmetrics_10fold\n",
        "\n",
        "\n",
        "# # Create Dataframe to save all wrongly predicted index\n",
        "df_indices = pd.DataFrame(all_wrong)\n",
        "df_indices.columns = ['indices']\n",
        "df_indices.to_pickle(\"./bertclassify/bert_indices_wrong.pkl\")"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "INFO:tensorflow:Restoring parameters from ./bertclassify/bert_training_10fold/bert_model.ckpt\n"
          ],
          "name": "stdout"
        },
        {
          "output_type": "stream",
          "text": [
            "\rTraining Minibatch Loop:   0%|          | 0/556 [00:00<?, ?it/s]"
          ],
          "name": "stderr"
        },
        {
          "output_type": "stream",
          "text": [
            "------------------------------------------------------------------------\n",
            "Training for Fold 1 ...\n"
          ],
          "name": "stdout"
        },
        {
          "output_type": "stream",
          "text": [
            "Training Minibatch Loop: 100%|██████████| 556/556 [07:01<00:00,  1.32it/s, accuracy=0.949, cost=0.385]\n",
            "Validation Minibatch Loop: 100%|██████████| 62/62 [00:15<00:00,  3.90it/s, accuracy=0.915, cost=0.411]\n",
            "Training Minibatch Loop:   0%|          | 0/556 [00:00<?, ?it/s]"
          ],
          "name": "stderr"
        },
        {
          "output_type": "stream",
          "text": [
            "Time Taken: 437.509482383728\n",
            "Epoch: 1, train_loss: 0.528906, train_acc: 0.772671, val_loss: 0.491374, val_acc: 0.812891\n",
            "\n"
          ],
          "name": "stdout"
        },
        {
          "output_type": "stream",
          "text": [
            "Training Minibatch Loop: 100%|██████████| 556/556 [06:55<00:00,  1.34it/s, accuracy=0.949, cost=0.362]\n",
            "Validation Minibatch Loop: 100%|██████████| 62/62 [00:15<00:00,  3.94it/s, accuracy=0.915, cost=0.413]\n",
            "Testing Minibatch Loop:   0%|          | 0/62 [00:00<?, ?it/s]"
          ],
          "name": "stderr"
        },
        {
          "output_type": "stream",
          "text": [
            "Time Taken: 431.7582893371582\n",
            "Epoch: 2, train_loss: 0.443253, train_acc: 0.870692, val_loss: 0.478566, val_acc: 0.830029\n",
            "\n"
          ],
          "name": "stdout"
        },
        {
          "output_type": "stream",
          "text": [
            "Testing Minibatch Loop: 100%|██████████| 62/62 [00:15<00:00,  3.94it/s, accuracy=0.915, cost=0.413]\n",
            "Testing for Metrics Minibatch Loop:   0%|          | 0/62 [00:00<?, ?it/s]"
          ],
          "name": "stderr"
        },
        {
          "output_type": "stream",
          "text": [
            "Time Taken: 447.4979438781738\n",
            "test_loss: 0.478566, test_acc: 0.830029\n",
            "\n",
            "Score for Fold 1: Loss of 0.4785661356103036; Accuracy of 83.00285262446249%\n"
          ],
          "name": "stdout"
        },
        {
          "output_type": "stream",
          "text": [
            "Testing for Metrics Minibatch Loop: 100%|██████████| 62/62 [00:15<00:00,  3.91it/s]\n"
          ],
          "name": "stderr"
        },
        {
          "output_type": "stream",
          "text": [
            "INFO:tensorflow:Restoring parameters from ./bertclassify/bert_training_10fold/bert_model.ckpt\n"
          ],
          "name": "stdout"
        },
        {
          "output_type": "stream",
          "text": [
            "\rTraining Minibatch Loop:   0%|          | 0/556 [00:00<?, ?it/s]"
          ],
          "name": "stderr"
        },
        {
          "output_type": "stream",
          "text": [
            "------------------------------------------------------------------------\n",
            "Training for Fold 2 ...\n"
          ],
          "name": "stdout"
        },
        {
          "output_type": "stream",
          "text": [
            "Training Minibatch Loop: 100%|██████████| 556/556 [06:56<00:00,  1.34it/s, accuracy=0.949, cost=0.4]\n",
            "Validation Minibatch Loop: 100%|██████████| 62/62 [00:15<00:00,  3.92it/s, accuracy=0.766, cost=0.509]\n",
            "Training Minibatch Loop:   0%|          | 0/556 [00:00<?, ?it/s]"
          ],
          "name": "stderr"
        },
        {
          "output_type": "stream",
          "text": [
            "Time Taken: 432.25549960136414\n",
            "Epoch: 1, train_loss: 0.527486, train_acc: 0.774413, val_loss: 0.493656, val_acc: 0.811245\n",
            "\n"
          ],
          "name": "stdout"
        },
        {
          "output_type": "stream",
          "text": [
            "Training Minibatch Loop: 100%|██████████| 556/556 [06:56<00:00,  1.33it/s, accuracy=0.949, cost=0.367]\n",
            "Validation Minibatch Loop: 100%|██████████| 62/62 [00:15<00:00,  3.93it/s, accuracy=0.894, cost=0.44]\n",
            "Testing Minibatch Loop:   0%|          | 0/62 [00:00<?, ?it/s]"
          ],
          "name": "stderr"
        },
        {
          "output_type": "stream",
          "text": [
            "Time Taken: 432.7271065711975\n",
            "Epoch: 2, train_loss: 0.441510, train_acc: 0.872126, val_loss: 0.478023, val_acc: 0.831449\n",
            "\n"
          ],
          "name": "stdout"
        },
        {
          "output_type": "stream",
          "text": [
            "Testing Minibatch Loop: 100%|██████████| 62/62 [00:15<00:00,  3.93it/s, accuracy=0.894, cost=0.44]\n",
            "Testing for Metrics Minibatch Loop:   0%|          | 0/62 [00:00<?, ?it/s]"
          ],
          "name": "stderr"
        },
        {
          "output_type": "stream",
          "text": [
            "Time Taken: 448.49768114089966\n",
            "test_loss: 0.478023, test_acc: 0.831449\n",
            "\n",
            "Score for Fold 2: Loss of 0.4780230142416493; Accuracy of 83.14494682896522%\n"
          ],
          "name": "stdout"
        },
        {
          "output_type": "stream",
          "text": [
            "Testing for Metrics Minibatch Loop: 100%|██████████| 62/62 [00:15<00:00,  3.93it/s]\n"
          ],
          "name": "stderr"
        },
        {
          "output_type": "stream",
          "text": [
            "INFO:tensorflow:Restoring parameters from ./bertclassify/bert_training_10fold/bert_model.ckpt\n"
          ],
          "name": "stdout"
        },
        {
          "output_type": "stream",
          "text": [
            "\rTraining Minibatch Loop:   0%|          | 0/556 [00:00<?, ?it/s]"
          ],
          "name": "stderr"
        },
        {
          "output_type": "stream",
          "text": [
            "------------------------------------------------------------------------\n",
            "Training for Fold 3 ...\n"
          ],
          "name": "stdout"
        },
        {
          "output_type": "stream",
          "text": [
            "Training Minibatch Loop: 100%|██████████| 556/556 [06:57<00:00,  1.33it/s, accuracy=0.923, cost=0.398]\n",
            "Validation Minibatch Loop: 100%|██████████| 62/62 [00:15<00:00,  3.93it/s, accuracy=0.872, cost=0.454]\n",
            "Training Minibatch Loop:   0%|          | 0/556 [00:00<?, ?it/s]"
          ],
          "name": "stderr"
        },
        {
          "output_type": "stream",
          "text": [
            "Time Taken: 433.2351179122925\n",
            "Epoch: 1, train_loss: 0.533100, train_acc: 0.768690, val_loss: 0.491856, val_acc: 0.812205\n",
            "\n"
          ],
          "name": "stdout"
        },
        {
          "output_type": "stream",
          "text": [
            "Training Minibatch Loop: 100%|██████████| 556/556 [06:57<00:00,  1.33it/s, accuracy=0.949, cost=0.364]\n",
            "Validation Minibatch Loop: 100%|██████████| 62/62 [00:15<00:00,  3.93it/s, accuracy=0.872, cost=0.455]\n",
            "Testing Minibatch Loop:   0%|          | 0/62 [00:00<?, ?it/s]"
          ],
          "name": "stderr"
        },
        {
          "output_type": "stream",
          "text": [
            "Time Taken: 433.16934037208557\n",
            "Epoch: 2, train_loss: 0.445865, train_acc: 0.866814, val_loss: 0.475639, val_acc: 0.834635\n",
            "\n"
          ],
          "name": "stdout"
        },
        {
          "output_type": "stream",
          "text": [
            "Testing Minibatch Loop: 100%|██████████| 62/62 [00:15<00:00,  3.93it/s, accuracy=0.872, cost=0.455]\n",
            "Testing for Metrics Minibatch Loop:   0%|          | 0/62 [00:00<?, ?it/s]"
          ],
          "name": "stderr"
        },
        {
          "output_type": "stream",
          "text": [
            "Time Taken: 448.95533561706543\n",
            "test_loss: 0.475639, test_acc: 0.834635\n",
            "\n",
            "Score for Fold 3: Loss of 0.4756393466264971; Accuracy of 83.46345232379052%\n"
          ],
          "name": "stdout"
        },
        {
          "output_type": "stream",
          "text": [
            "Testing for Metrics Minibatch Loop: 100%|██████████| 62/62 [00:15<00:00,  3.93it/s]\n"
          ],
          "name": "stderr"
        },
        {
          "output_type": "stream",
          "text": [
            "INFO:tensorflow:Restoring parameters from ./bertclassify/bert_training_10fold/bert_model.ckpt\n"
          ],
          "name": "stdout"
        },
        {
          "output_type": "stream",
          "text": [
            "\rTraining Minibatch Loop:   0%|          | 0/556 [00:00<?, ?it/s]"
          ],
          "name": "stderr"
        },
        {
          "output_type": "stream",
          "text": [
            "------------------------------------------------------------------------\n",
            "Training for Fold 4 ...\n"
          ],
          "name": "stdout"
        },
        {
          "output_type": "stream",
          "text": [
            "Training Minibatch Loop: 100%|██████████| 556/556 [06:57<00:00,  1.33it/s, accuracy=0.923, cost=0.393]\n",
            "Validation Minibatch Loop: 100%|██████████| 62/62 [00:15<00:00,  3.92it/s, accuracy=0.872, cost=0.458]\n",
            "Training Minibatch Loop:   0%|          | 0/556 [00:00<?, ?it/s]"
          ],
          "name": "stderr"
        },
        {
          "output_type": "stream",
          "text": [
            "Time Taken: 433.2362542152405\n",
            "Epoch: 1, train_loss: 0.532530, train_acc: 0.768775, val_loss: 0.487116, val_acc: 0.819262\n",
            "\n"
          ],
          "name": "stdout"
        },
        {
          "output_type": "stream",
          "text": [
            "Training Minibatch Loop: 100%|██████████| 556/556 [06:57<00:00,  1.33it/s, accuracy=0.974, cost=0.346]\n",
            "Validation Minibatch Loop: 100%|██████████| 62/62 [00:15<00:00,  3.92it/s, accuracy=0.872, cost=0.45]\n",
            "Testing Minibatch Loop:   0%|          | 0/62 [00:00<?, ?it/s]"
          ],
          "name": "stderr"
        },
        {
          "output_type": "stream",
          "text": [
            "Time Taken: 433.19430112838745\n",
            "Epoch: 2, train_loss: 0.445254, train_acc: 0.868631, val_loss: 0.473897, val_acc: 0.829846\n",
            "\n"
          ],
          "name": "stdout"
        },
        {
          "output_type": "stream",
          "text": [
            "Testing Minibatch Loop: 100%|██████████| 62/62 [00:15<00:00,  3.93it/s, accuracy=0.872, cost=0.45]\n",
            "Testing for Metrics Minibatch Loop:   0%|          | 0/62 [00:00<?, ?it/s]"
          ],
          "name": "stderr"
        },
        {
          "output_type": "stream",
          "text": [
            "Time Taken: 448.97565174102783\n",
            "test_loss: 0.473897, test_acc: 0.829846\n",
            "\n",
            "Score for Fold 4: Loss of 0.4738973705999313; Accuracy of 82.98462167862924%\n"
          ],
          "name": "stdout"
        },
        {
          "output_type": "stream",
          "text": [
            "Testing for Metrics Minibatch Loop: 100%|██████████| 62/62 [00:15<00:00,  3.94it/s]\n"
          ],
          "name": "stderr"
        },
        {
          "output_type": "stream",
          "text": [
            "INFO:tensorflow:Restoring parameters from ./bertclassify/bert_training_10fold/bert_model.ckpt\n"
          ],
          "name": "stdout"
        },
        {
          "output_type": "stream",
          "text": [
            "\rTraining Minibatch Loop:   0%|          | 0/556 [00:00<?, ?it/s]"
          ],
          "name": "stderr"
        },
        {
          "output_type": "stream",
          "text": [
            "------------------------------------------------------------------------\n",
            "Training for Fold 5 ...\n"
          ],
          "name": "stdout"
        },
        {
          "output_type": "stream",
          "text": [
            "Training Minibatch Loop: 100%|██████████| 556/556 [06:57<00:00,  1.33it/s, accuracy=0.897, cost=0.402]\n",
            "Validation Minibatch Loop: 100%|██████████| 62/62 [00:15<00:00,  3.92it/s, accuracy=0.851, cost=0.46]\n",
            "Training Minibatch Loop:   0%|          | 0/556 [00:00<?, ?it/s]"
          ],
          "name": "stderr"
        },
        {
          "output_type": "stream",
          "text": [
            "Time Taken: 433.4512097835541\n",
            "Epoch: 1, train_loss: 0.531874, train_acc: 0.768869, val_loss: 0.492328, val_acc: 0.811862\n",
            "\n"
          ],
          "name": "stdout"
        },
        {
          "output_type": "stream",
          "text": [
            "Training Minibatch Loop: 100%|██████████| 556/556 [06:57<00:00,  1.33it/s, accuracy=0.974, cost=0.358]\n",
            "Validation Minibatch Loop: 100%|██████████| 62/62 [00:15<00:00,  3.92it/s, accuracy=0.851, cost=0.459]\n",
            "Testing Minibatch Loop:   0%|          | 0/62 [00:00<?, ?it/s]"
          ],
          "name": "stderr"
        },
        {
          "output_type": "stream",
          "text": [
            "Time Taken: 433.3758065700531\n",
            "Epoch: 2, train_loss: 0.443703, train_acc: 0.870289, val_loss: 0.476771, val_acc: 0.831267\n",
            "\n"
          ],
          "name": "stdout"
        },
        {
          "output_type": "stream",
          "text": [
            "Testing Minibatch Loop: 100%|██████████| 62/62 [00:15<00:00,  3.93it/s, accuracy=0.851, cost=0.459]\n",
            "Testing for Metrics Minibatch Loop:   0%|          | 0/62 [00:00<?, ?it/s]"
          ],
          "name": "stderr"
        },
        {
          "output_type": "stream",
          "text": [
            "Time Taken: 449.14983201026917\n",
            "test_loss: 0.476771, test_acc: 0.831267\n",
            "\n",
            "Score for Fold 5: Loss of 0.47677069133327854; Accuracy of 83.12671588313195%\n"
          ],
          "name": "stdout"
        },
        {
          "output_type": "stream",
          "text": [
            "Testing for Metrics Minibatch Loop: 100%|██████████| 62/62 [00:15<00:00,  3.94it/s]\n"
          ],
          "name": "stderr"
        },
        {
          "output_type": "stream",
          "text": [
            "INFO:tensorflow:Restoring parameters from ./bertclassify/bert_training_10fold/bert_model.ckpt\n"
          ],
          "name": "stdout"
        },
        {
          "output_type": "stream",
          "text": [
            "\rTraining Minibatch Loop:   0%|          | 0/556 [00:00<?, ?it/s]"
          ],
          "name": "stderr"
        },
        {
          "output_type": "stream",
          "text": [
            "------------------------------------------------------------------------\n",
            "Training for Fold 6 ...\n"
          ],
          "name": "stdout"
        },
        {
          "output_type": "stream",
          "text": [
            "Training Minibatch Loop: 100%|██████████| 556/556 [06:57<00:00,  1.33it/s, accuracy=0.923, cost=0.397]\n",
            "Validation Minibatch Loop: 100%|██████████| 62/62 [00:15<00:00,  3.93it/s, accuracy=0.851, cost=0.47]\n",
            "Training Minibatch Loop:   0%|          | 0/556 [00:00<?, ?it/s]"
          ],
          "name": "stderr"
        },
        {
          "output_type": "stream",
          "text": [
            "Time Taken: 433.4746427536011\n",
            "Epoch: 1, train_loss: 0.529952, train_acc: 0.771810, val_loss: 0.506601, val_acc: 0.797497\n",
            "\n"
          ],
          "name": "stdout"
        },
        {
          "output_type": "stream",
          "text": [
            "Training Minibatch Loop: 100%|██████████| 556/556 [06:57<00:00,  1.33it/s, accuracy=0.974, cost=0.357]\n",
            "Validation Minibatch Loop: 100%|██████████| 62/62 [00:15<00:00,  3.92it/s, accuracy=0.809, cost=0.479]\n",
            "Testing Minibatch Loop:   0%|          | 0/62 [00:00<?, ?it/s]"
          ],
          "name": "stderr"
        },
        {
          "output_type": "stream",
          "text": [
            "Time Taken: 433.10436272621155\n",
            "Epoch: 2, train_loss: 0.443712, train_acc: 0.869755, val_loss: 0.488163, val_acc: 0.810924\n",
            "\n"
          ],
          "name": "stdout"
        },
        {
          "output_type": "stream",
          "text": [
            "Testing Minibatch Loop: 100%|██████████| 62/62 [00:15<00:00,  3.92it/s, accuracy=0.809, cost=0.479]\n",
            "Testing for Metrics Minibatch Loop:   0%|          | 0/62 [00:00<?, ?it/s]"
          ],
          "name": "stderr"
        },
        {
          "output_type": "stream",
          "text": [
            "Time Taken: 448.9092183113098\n",
            "test_loss: 0.488163, test_acc: 0.810924\n",
            "\n",
            "Score for Fold 6: Loss of 0.4881629328573904; Accuracy of 81.09235590504062%\n"
          ],
          "name": "stdout"
        },
        {
          "output_type": "stream",
          "text": [
            "Testing for Metrics Minibatch Loop: 100%|██████████| 62/62 [00:15<00:00,  3.94it/s]\n"
          ],
          "name": "stderr"
        },
        {
          "output_type": "stream",
          "text": [
            "INFO:tensorflow:Restoring parameters from ./bertclassify/bert_training_10fold/bert_model.ckpt\n"
          ],
          "name": "stdout"
        },
        {
          "output_type": "stream",
          "text": [
            "\rTraining Minibatch Loop:   0%|          | 0/556 [00:00<?, ?it/s]"
          ],
          "name": "stderr"
        },
        {
          "output_type": "stream",
          "text": [
            "------------------------------------------------------------------------\n",
            "Training for Fold 7 ...\n"
          ],
          "name": "stdout"
        },
        {
          "output_type": "stream",
          "text": [
            "Training Minibatch Loop: 100%|██████████| 556/556 [06:57<00:00,  1.33it/s, accuracy=0.897, cost=0.399]\n",
            "Validation Minibatch Loop: 100%|██████████| 62/62 [00:15<00:00,  3.93it/s, accuracy=0.809, cost=0.474]\n",
            "Training Minibatch Loop:   0%|          | 0/556 [00:00<?, ?it/s]"
          ],
          "name": "stderr"
        },
        {
          "output_type": "stream",
          "text": [
            "Time Taken: 433.17708015441895\n",
            "Epoch: 1, train_loss: 0.526584, train_acc: 0.774827, val_loss: 0.503325, val_acc: 0.800591\n",
            "\n"
          ],
          "name": "stdout"
        },
        {
          "output_type": "stream",
          "text": [
            "Training Minibatch Loop: 100%|██████████| 556/556 [06:57<00:00,  1.33it/s, accuracy=0.949, cost=0.362]\n",
            "Validation Minibatch Loop: 100%|██████████| 62/62 [00:15<00:00,  3.93it/s, accuracy=0.83, cost=0.478]\n",
            "Testing Minibatch Loop:   0%|          | 0/62 [00:00<?, ?it/s]"
          ],
          "name": "stderr"
        },
        {
          "output_type": "stream",
          "text": [
            "Time Taken: 432.99807262420654\n",
            "Epoch: 2, train_loss: 0.441804, train_acc: 0.872154, val_loss: 0.486948, val_acc: 0.818827\n",
            "\n"
          ],
          "name": "stdout"
        },
        {
          "output_type": "stream",
          "text": [
            "Testing Minibatch Loop: 100%|██████████| 62/62 [00:15<00:00,  3.93it/s, accuracy=0.83, cost=0.478]\n",
            "Testing for Metrics Minibatch Loop:   0%|          | 0/62 [00:00<?, ?it/s]"
          ],
          "name": "stderr"
        },
        {
          "output_type": "stream",
          "text": [
            "Time Taken: 448.77702593803406\n",
            "test_loss: 0.486948, test_acc: 0.818827\n",
            "\n",
            "Score for Fold 7: Loss of 0.4869482560503867; Accuracy of 81.88272137795725%\n"
          ],
          "name": "stdout"
        },
        {
          "output_type": "stream",
          "text": [
            "Testing for Metrics Minibatch Loop: 100%|██████████| 62/62 [00:15<00:00,  3.94it/s]\n"
          ],
          "name": "stderr"
        },
        {
          "output_type": "stream",
          "text": [
            "INFO:tensorflow:Restoring parameters from ./bertclassify/bert_training_10fold/bert_model.ckpt\n"
          ],
          "name": "stdout"
        },
        {
          "output_type": "stream",
          "text": [
            "\rTraining Minibatch Loop:   0%|          | 0/556 [00:00<?, ?it/s]"
          ],
          "name": "stderr"
        },
        {
          "output_type": "stream",
          "text": [
            "------------------------------------------------------------------------\n",
            "Training for Fold 8 ...\n"
          ],
          "name": "stdout"
        },
        {
          "output_type": "stream",
          "text": [
            "Training Minibatch Loop: 100%|██████████| 556/556 [06:56<00:00,  1.33it/s, accuracy=0.923, cost=0.388]\n",
            "Validation Minibatch Loop: 100%|██████████| 62/62 [00:15<00:00,  3.94it/s, accuracy=0.83, cost=0.492]\n",
            "Training Minibatch Loop:   0%|          | 0/556 [00:00<?, ?it/s]"
          ],
          "name": "stderr"
        },
        {
          "output_type": "stream",
          "text": [
            "Time Taken: 432.55899596214294\n",
            "Epoch: 1, train_loss: 0.531124, train_acc: 0.769112, val_loss: 0.477806, val_acc: 0.828404\n",
            "\n"
          ],
          "name": "stdout"
        },
        {
          "output_type": "stream",
          "text": [
            "Training Minibatch Loop: 100%|██████████| 556/556 [06:57<00:00,  1.33it/s, accuracy=0.974, cost=0.346]\n",
            "Validation Minibatch Loop: 100%|██████████| 62/62 [00:15<00:00,  3.92it/s, accuracy=0.809, cost=0.49]\n",
            "Testing Minibatch Loop:   0%|          | 0/62 [00:00<?, ?it/s]"
          ],
          "name": "stderr"
        },
        {
          "output_type": "stream",
          "text": [
            "Time Taken: 432.82981157302856\n",
            "Epoch: 2, train_loss: 0.444860, train_acc: 0.868996, val_loss: 0.468003, val_acc: 0.841922\n",
            "\n"
          ],
          "name": "stdout"
        },
        {
          "output_type": "stream",
          "text": [
            "Testing Minibatch Loop: 100%|██████████| 62/62 [00:15<00:00,  3.93it/s, accuracy=0.809, cost=0.49]\n",
            "Testing for Metrics Minibatch Loop:   0%|          | 0/62 [00:00<?, ?it/s]"
          ],
          "name": "stderr"
        },
        {
          "output_type": "stream",
          "text": [
            "Time Taken: 448.61877703666687\n",
            "test_loss: 0.468003, test_acc: 0.841922\n",
            "\n",
            "Score for Fold 8: Loss of 0.4680031090974808; Accuracy of 84.19215429213739%\n"
          ],
          "name": "stdout"
        },
        {
          "output_type": "stream",
          "text": [
            "Testing for Metrics Minibatch Loop: 100%|██████████| 62/62 [00:15<00:00,  3.94it/s]\n"
          ],
          "name": "stderr"
        },
        {
          "output_type": "stream",
          "text": [
            "INFO:tensorflow:Restoring parameters from ./bertclassify/bert_training_10fold/bert_model.ckpt\n"
          ],
          "name": "stdout"
        },
        {
          "output_type": "stream",
          "text": [
            "\rTraining Minibatch Loop:   0%|          | 0/556 [00:00<?, ?it/s]"
          ],
          "name": "stderr"
        },
        {
          "output_type": "stream",
          "text": [
            "------------------------------------------------------------------------\n",
            "Training for Fold 9 ...\n"
          ],
          "name": "stdout"
        },
        {
          "output_type": "stream",
          "text": [
            "Training Minibatch Loop: 100%|██████████| 556/556 [06:57<00:00,  1.33it/s, accuracy=0.949, cost=0.383]\n",
            "Validation Minibatch Loop: 100%|██████████| 62/62 [00:15<00:00,  3.93it/s, accuracy=0.745, cost=0.551]\n",
            "Training Minibatch Loop:   0%|          | 0/556 [00:00<?, ?it/s]"
          ],
          "name": "stderr"
        },
        {
          "output_type": "stream",
          "text": [
            "Time Taken: 433.3423891067505\n",
            "Epoch: 1, train_loss: 0.531271, train_acc: 0.770113, val_loss: 0.490517, val_acc: 0.816698\n",
            "\n"
          ],
          "name": "stdout"
        },
        {
          "output_type": "stream",
          "text": [
            "Training Minibatch Loop: 100%|██████████| 556/556 [06:57<00:00,  1.33it/s, accuracy=0.949, cost=0.36]\n",
            "Validation Minibatch Loop: 100%|██████████| 62/62 [00:15<00:00,  3.93it/s, accuracy=0.809, cost=0.53]\n",
            "Testing Minibatch Loop:   0%|          | 0/62 [00:00<?, ?it/s]"
          ],
          "name": "stderr"
        },
        {
          "output_type": "stream",
          "text": [
            "Time Taken: 433.2328600883484\n",
            "Epoch: 2, train_loss: 0.441873, train_acc: 0.872997, val_loss: 0.473714, val_acc: 0.832849\n",
            "\n"
          ],
          "name": "stdout"
        },
        {
          "output_type": "stream",
          "text": [
            "Testing Minibatch Loop: 100%|██████████| 62/62 [00:15<00:00,  3.92it/s, accuracy=0.809, cost=0.53]\n",
            "Testing for Metrics Minibatch Loop:   0%|          | 0/62 [00:00<?, ?it/s]"
          ],
          "name": "stderr"
        },
        {
          "output_type": "stream",
          "text": [
            "Time Taken: 449.0365710258484\n",
            "test_loss: 0.473714, test_acc: 0.832849\n",
            "\n",
            "Score for Fold 9: Loss of 0.473714348769957; Accuracy of 83.28489622762126%\n"
          ],
          "name": "stdout"
        },
        {
          "output_type": "stream",
          "text": [
            "Testing for Metrics Minibatch Loop: 100%|██████████| 62/62 [00:15<00:00,  3.94it/s]\n"
          ],
          "name": "stderr"
        },
        {
          "output_type": "stream",
          "text": [
            "INFO:tensorflow:Restoring parameters from ./bertclassify/bert_training_10fold/bert_model.ckpt\n"
          ],
          "name": "stdout"
        },
        {
          "output_type": "stream",
          "text": [
            "\rTraining Minibatch Loop:   0%|          | 0/556 [00:00<?, ?it/s]"
          ],
          "name": "stderr"
        },
        {
          "output_type": "stream",
          "text": [
            "------------------------------------------------------------------------\n",
            "Training for Fold 10 ...\n"
          ],
          "name": "stdout"
        },
        {
          "output_type": "stream",
          "text": [
            "Training Minibatch Loop: 100%|██████████| 556/556 [06:57<00:00,  1.33it/s, accuracy=0.744, cost=0.571]\n",
            "Validation Minibatch Loop: 100%|██████████| 62/62 [00:15<00:00,  3.93it/s, accuracy=0.915, cost=0.407]\n",
            "Training Minibatch Loop:   0%|          | 0/556 [00:00<?, ?it/s]"
          ],
          "name": "stderr"
        },
        {
          "output_type": "stream",
          "text": [
            "Time Taken: 433.32864117622375\n",
            "Epoch: 1, train_loss: 0.532025, train_acc: 0.769379, val_loss: 0.492859, val_acc: 0.808607\n",
            "\n"
          ],
          "name": "stdout"
        },
        {
          "output_type": "stream",
          "text": [
            "Training Minibatch Loop: 100%|██████████| 556/556 [06:57<00:00,  1.33it/s, accuracy=0.795, cost=0.504]\n",
            "Validation Minibatch Loop: 100%|██████████| 62/62 [00:15<00:00,  3.93it/s, accuracy=0.894, cost=0.406]\n",
            "Testing Minibatch Loop:   0%|          | 0/62 [00:00<?, ?it/s]"
          ],
          "name": "stderr"
        },
        {
          "output_type": "stream",
          "text": [
            "Time Taken: 433.47540402412415\n",
            "Epoch: 2, train_loss: 0.444239, train_acc: 0.870781, val_loss: 0.477094, val_acc: 0.826409\n",
            "\n"
          ],
          "name": "stdout"
        },
        {
          "output_type": "stream",
          "text": [
            "Testing Minibatch Loop: 100%|██████████| 62/62 [00:15<00:00,  3.93it/s, accuracy=0.894, cost=0.406]\n",
            "Testing for Metrics Minibatch Loop:   0%|          | 0/62 [00:00<?, ?it/s]"
          ],
          "name": "stderr"
        },
        {
          "output_type": "stream",
          "text": [
            "Time Taken: 449.2678015232086\n",
            "test_loss: 0.477094, test_acc: 0.826409\n",
            "\n",
            "Score for Fold 10: Loss of 0.47709384320243714; Accuracy of 82.6409145709007%\n"
          ],
          "name": "stdout"
        },
        {
          "output_type": "stream",
          "text": [
            "Testing for Metrics Minibatch Loop: 100%|██████████| 62/62 [00:15<00:00,  3.93it/s]\n"
          ],
          "name": "stderr"
        },
        {
          "output_type": "stream",
          "text": [
            "------------------------------------------------------------------------\n",
            "Score per fold\n",
            "------------------------------------------------------------------------\n",
            "> Fold 1 - Loss: 0.4785661356103036 - Accuracy: 83.00285262446249%\n",
            "------------------------------------------------------------------------\n",
            "> Fold 2 - Loss: 0.4780230142416493 - Accuracy: 83.14494682896522%\n",
            "------------------------------------------------------------------------\n",
            "> Fold 3 - Loss: 0.4756393466264971 - Accuracy: 83.46345232379052%\n",
            "------------------------------------------------------------------------\n",
            "> Fold 4 - Loss: 0.4738973705999313 - Accuracy: 82.98462167862924%\n",
            "------------------------------------------------------------------------\n",
            "> Fold 5 - Loss: 0.47677069133327854 - Accuracy: 83.12671588313195%\n",
            "------------------------------------------------------------------------\n",
            "> Fold 6 - Loss: 0.4881629328573904 - Accuracy: 81.09235590504062%\n",
            "------------------------------------------------------------------------\n",
            "> Fold 7 - Loss: 0.4869482560503867 - Accuracy: 81.88272137795725%\n",
            "------------------------------------------------------------------------\n",
            "> Fold 8 - Loss: 0.4680031090974808 - Accuracy: 84.19215429213739%\n",
            "------------------------------------------------------------------------\n",
            "> Fold 9 - Loss: 0.473714348769957 - Accuracy: 83.28489622762126%\n",
            "------------------------------------------------------------------------\n",
            "> Fold 10 - Loss: 0.47709384320243714 - Accuracy: 82.6409145709007%\n",
            "------------------------------------------------------------------------\n",
            "Average scores for all folds:\n",
            "> Accuracy: 82.88156317126366 (+- 0.8142839759806219)\n",
            "> Loss: 0.4776819048389312\n",
            "------------------------------------------------------------------------\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "0R2ySYs2DKqp"
      },
      "source": [
        "#TODO: Determine history for each fold\n",
        "\n",
        "# history_dict = hist.history\n",
        "# history_dict = results_history[0]\n",
        "# # history_dict = hist\n",
        "# history_dict.keys()"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "DvbznpxfRIRl"
      },
      "source": [
        "Plot Loss and Accuracy for each Fold between Training and Validation for each epoch.\n"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "tyOe8kQrRHq9",
        "outputId": "2da225e1-3391-4f88-d81b-24937456c04b",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 1000
        }
      },
      "source": [
        "import matplotlib.pyplot as plt\n",
        "\n",
        "acc = []\n",
        "val_acc = []\n",
        "loss = []\n",
        "val_loss = []\n",
        "\n",
        "for i in range(len(results_history)):\n",
        "  tmp_acc, tmp_val_acc, tmp_loss, tmp_val_loss = [], [], [], []\n",
        "  for j in range(len(results_history[i])):\n",
        "    tmp_acc.append(np.asarray(results_history[i][j][0]))\n",
        "    tmp_val_acc.append(np.asarray(results_history[i][j][1]))\n",
        "    tmp_loss.append(np.asarray(results_history[i][j][2]))\n",
        "    tmp_val_loss.append(np.asarray(results_history[i][j][3]))\n",
        "  acc.append(tmp_acc)\n",
        "  val_acc.append(tmp_val_acc)\n",
        "  loss.append(tmp_loss)\n",
        "  val_loss.append(tmp_val_loss)\n",
        "\n",
        "  #[0] list_train_acc\n",
        "  #[1] list_val_acc\n",
        "  #[2] list_train_loss\n",
        "  #[3] list_val_loss\n",
        "\n",
        "\n",
        "# # Loss Plot\n",
        "epochs = range(1, len(acc[0]) + 1)\n",
        "fig_row = -(-(len(acc)) // 2)\n",
        "fig_col = 2\n",
        "plt.rcParams['figure.figsize'] = [12, 18]\n",
        "for i in range(len(loss)):\n",
        "  plt.subplot(fig_row, fig_col, i+1)\n",
        "  plt.plot(epochs, loss[i], 'r', label='Training loss')\n",
        "  plt.plot(epochs, val_loss[i], 'r--', label='Validation loss')\n",
        "  plt.title('NonFT-BERT+CNN Training and Validation Loss, Fold {}'.format(i+1))\n",
        "  plt.xlabel('Epochs')\n",
        "  plt.ylabel('Loss')\n",
        "  plt.legend()\n",
        "  plt.grid(True)\n",
        "  # plt.grid(which='minor', linestyle=':', linewidth='0.5', color='black')\n",
        "  # plt.minorticks_on()\n",
        "plt.tight_layout(pad=3.0)\n",
        "plt.savefig(\"./bertclassify/bert_loss_10fold.png\", format='png', dpi=150, bbox_inches='tight')\n",
        "plt.show()\n",
        "\n",
        "\n",
        "# # Accuracy Plot\n",
        "plt.clf()   # clear figure\n",
        "for i in range(len(acc)):\n",
        "  plt.subplot(fig_row, fig_col, i+1)\n",
        "  plt.plot(epochs, acc[i], 'b', label='Training acc')\n",
        "  plt.plot(epochs, val_acc[i], 'b--', label='Validation acc')\n",
        "  plt.title('NonFT-BERT+CNN Training and Validation Accuracy, Fold {}'.format(i+1))\n",
        "  plt.xlabel('Epochs')\n",
        "  plt.ylabel('Accuracy')\n",
        "  plt.legend()\n",
        "  plt.grid(True)\n",
        "plt.tight_layout(pad=3.0)\n",
        "plt.savefig(\"./bertclassify/bert_accuracy_10fold.png\", format='png', dpi=150, bbox_inches='tight')\n",
        "plt.show()"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 864x1296 with 10 Axes>"
            ]
          },
          "metadata": {
            "tags": [],
            "needs_background": "light"
          }
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 864x1296 with 10 Axes>"
            ]
          },
          "metadata": {
            "tags": [],
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "OxY3Er8-ackQ"
      },
      "source": [
        "Calculate Model Metrics\n",
        "<br> Precision, Recall, Accuracy, F1-Score"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "Hy2iGSSZxyiK",
        "outputId": "f9f96aa5-2cd3-473b-beff-0db510faf989",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 1000
        }
      },
      "source": [
        "idx_best = 0\n",
        "best_acc = 0\n",
        "\n",
        "# == Provide average scores ==\n",
        "print('------------------------------------------------------------------------')\n",
        "print('Metrics Score per fold')\n",
        "for i in range(0, len(all_accuracy)):\n",
        "  print('------------------------------------------------------------------------')\n",
        "  print(f'> Fold {i+1}')\n",
        "  print('Accuracy: %f' % all_accuracy[i])\n",
        "  print('Precision: %f' % all_precision[i])\n",
        "  print('Recall: %f' % all_recall[i])\n",
        "  print('F1 score: %f' % all_f1[i])\n",
        "  if (all_accuracy[i] >= best_acc):\n",
        "    idx_best = i\n",
        "    best_acc = all_accuracy[i]\n",
        "print('------------------------------------------------------------------------')\n",
        "print('------------------------------------------------------------------------')\n",
        "print('Average scores for all folds:')\n",
        "print(f'> Average Accuracy: {np.mean(np.asarray(all_accuracy))}')\n",
        "print(f'> Average Precision: {np.mean(np.asarray(all_precision))}')\n",
        "print(f'> Average Recall: {np.mean(np.asarray(all_recall))}')\n",
        "print(f'> Average F1 Score: {np.mean(np.asarray(all_f1))}')\n",
        "print('------------------------------------------------------------------------')\n",
        "print('------------------------------------------------------------------------')\n",
        "\n"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "------------------------------------------------------------------------\n",
            "Metrics Score per fold\n",
            "------------------------------------------------------------------------\n",
            "> Fold 1\n",
            "Accuracy: 0.829663\n",
            "Precision: 0.831234\n",
            "Recall: 0.918854\n",
            "F1 score: 0.872851\n",
            "------------------------------------------------------------------------\n",
            "> Fold 2\n",
            "Accuracy: 0.831182\n",
            "Precision: 0.831921\n",
            "Recall: 0.925010\n",
            "F1 score: 0.875999\n",
            "------------------------------------------------------------------------\n",
            "> Fold 3\n",
            "Accuracy: 0.834472\n",
            "Precision: 0.829908\n",
            "Recall: 0.930843\n",
            "F1 score: 0.877482\n",
            "------------------------------------------------------------------------\n",
            "> Fold 4\n",
            "Accuracy: 0.829663\n",
            "Precision: 0.829191\n",
            "Recall: 0.927565\n",
            "F1 score: 0.875624\n",
            "------------------------------------------------------------------------\n",
            "> Fold 5\n",
            "Accuracy: 0.831182\n",
            "Precision: 0.835048\n",
            "Recall: 0.920970\n",
            "F1 score: 0.875907\n",
            "------------------------------------------------------------------------\n",
            "> Fold 6\n",
            "Accuracy: 0.810934\n",
            "Precision: 0.809123\n",
            "Recall: 0.919091\n",
            "F1 score: 0.860608\n",
            "------------------------------------------------------------------------\n",
            "> Fold 7\n",
            "Accuracy: 0.818780\n",
            "Precision: 0.816104\n",
            "Recall: 0.923230\n",
            "F1 score: 0.866368\n",
            "------------------------------------------------------------------------\n",
            "> Fold 8\n",
            "Accuracy: 0.842065\n",
            "Precision: 0.841085\n",
            "Recall: 0.932422\n",
            "F1 score: 0.884402\n",
            "------------------------------------------------------------------------\n",
            "> Fold 9\n",
            "Accuracy: 0.832954\n",
            "Precision: 0.829181\n",
            "Recall: 0.928287\n",
            "F1 score: 0.875940\n",
            "------------------------------------------------------------------------\n",
            "> Fold 10\n",
            "Accuracy: 0.826120\n",
            "Precision: 0.828223\n",
            "Recall: 0.924543\n",
            "F1 score: 0.873736\n",
            "------------------------------------------------------------------------\n",
            "------------------------------------------------------------------------\n",
            "Average scores for all folds:\n",
            "> Average Accuracy: 0.8287015945330296\n",
            "> Average Precision: 0.8281019024813631\n",
            "> Average Recall: 0.925081459952745\n",
            "> Average F1 Score: 0.8738917396412372\n",
            "------------------------------------------------------------------------\n",
            "------------------------------------------------------------------------\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "omMUUs8W9pPr"
      },
      "source": [
        "Plot Precision-Recall Curve"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "rEoLXKNQ0a1N",
        "outputId": "3f4dac97-f683-408f-d8c3-da8309749271",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 730
        }
      },
      "source": [
        "import matplotlib.cm as cm\n",
        "\n",
        "colors = cm.tab10(np.linspace(0, 1, len(all_arr_recall)))\n",
        "\n",
        "\n",
        "plt.clf()   # clear figure\n",
        "plt.rcParams['figure.figsize'] = [20, 12]\n",
        "\n",
        "for i in range(len(all_arr_recall)):\n",
        "  plt.plot(all_arr_recall[i], all_arr_precision[i], color=colors[i], label='NonFT-BERT+CNN, Fold {}, AP={:0.2f}'.format(i+1, all_ave_precision[i]))\n",
        "  # plt.plot(np.linspace(0, 1, thresholds[i].size), thresholds[i], linestyle='--', label='Threshold')\n",
        "plt.title('NonFT-BERT+CNN Precision-Recall')\n",
        "plt.xlabel('Recall')\n",
        "plt.ylabel('Precision')\n",
        "plt.legend()\n",
        "plt.grid(True)\n",
        "# plt.grid(which='minor', linestyle=':', linewidth='0.5', color='black')\n",
        "# plt.minorticks_on()\n",
        "plt.savefig(\"./bertclassify/bert_10fold_precision_recall.png\", format='png', dpi=150, bbox_inches='tight')\n",
        "plt.show()\n"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 1440x864 with 1 Axes>"
            ]
          },
          "metadata": {
            "tags": [],
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "cH7lzjNr97zb"
      },
      "source": [
        "Draw Confusion Matrix"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "Z3zGbGIn-Q3L",
        "outputId": "5b0913ea-9139-457d-93de-a24b325bcce9",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 852
        }
      },
      "source": [
        "import seaborn as sns\n",
        "\n",
        "plt.rcParams['figure.figsize'] = [12, 9]\n",
        "for j in range(len(all_cf_matrices)):\n",
        "  cf_matrix_best = all_cf_matrices[j]\n",
        "\n",
        "  stats_text = \"\\n\\nAccuracy={:0.3f}\\nPrecision={:0.3f}\\nRecall={:0.3f}\\nF1 Score={:0.3f}\".format(\n",
        "                  all_accuracy[j],all_precision[j],all_recall[j],all_f1[j])\n",
        "\n",
        "  group_names = ['True Neg','False Pos','False Neg','True Pos']\n",
        "\n",
        "  group_counts = [\"{0:0.0f}\".format(value) for value in\n",
        "                  cf_matrix_best.flatten()]\n",
        "\n",
        "  group_percentages = [\"{0:.2%}\".format(value) for value in\n",
        "                      cf_matrix_best.flatten()/np.sum(cf_matrix_best)]\n",
        "\n",
        "  labels = [f'{v1}\\n{v2}\\n{v3}' for v1, v2, v3 in\n",
        "            zip(group_names,group_counts,group_percentages)]\n",
        "            \n",
        "  labels = np.asarray(labels).reshape(2,2)\n",
        "  categories = ['Fake', 'Real']\n",
        "  # cmap = sns.dark_palette(\"red\")\n",
        "  # cmap = sns.cubehelix_palette(8, start=.5, rot=-.75, as_cmap=True)\n",
        "  # cmap = sns.color_palette(\"GnBu_d\")\n",
        "\n",
        "  plt.clf() \n",
        "  plt.figure()\n",
        "  svm = sns.heatmap(cf_matrix_best, annot=labels, fmt='', cmap='PuBu',xticklabels=categories,yticklabels=categories)\n",
        "\n",
        "  plt.ylabel('True Label')\n",
        "  plt.xlabel('Predicted Label' + stats_text)\n",
        "\n",
        "  # figure = svm.get_figure()    \n",
        "  # figure.savefig('./bertclassify/bert_conf_matrix.png', dpi=150)\n",
        "\n",
        "  plt.savefig(\"./bertclassify/cfpr/bert_best_10fold_conf_matrix_fold{}.png\".format(j+1), format='png', dpi=150, bbox_inches='tight')"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "/usr/local/lib/python3.6/dist-packages/statsmodels/tools/_testing.py:19: FutureWarning: pandas.util.testing is deprecated. Use the functions in the public API at pandas.testing instead.\n",
            "  import pandas.util.testing as tm\n"
          ],
          "name": "stderr"
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 864x648 with 0 Axes>"
            ]
          },
          "metadata": {
            "tags": []
          }
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 864x648 with 0 Axes>"
            ]
          },
          "metadata": {
            "tags": []
          }
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 864x648 with 0 Axes>"
            ]
          },
          "metadata": {
            "tags": []
          }
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 864x648 with 0 Axes>"
            ]
          },
          "metadata": {
            "tags": []
          }
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 864x648 with 0 Axes>"
            ]
          },
          "metadata": {
            "tags": []
          }
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 864x648 with 0 Axes>"
            ]
          },
          "metadata": {
            "tags": []
          }
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 864x648 with 0 Axes>"
            ]
          },
          "metadata": {
            "tags": []
          }
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 864x648 with 0 Axes>"
            ]
          },
          "metadata": {
            "tags": []
          }
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 864x648 with 0 Axes>"
            ]
          },
          "metadata": {
            "tags": []
          }
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 864x648 with 0 Axes>"
            ]
          },
          "metadata": {
            "tags": []
          }
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 864x648 with 2 Axes>"
            ]
          },
          "metadata": {
            "tags": [],
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "b4fEIvKdX26j"
      },
      "source": [
        "for m in range(len(all_arr_precision)):\n",
        "  df_prerec = pd.DataFrame(all_arr_precision[m])\n",
        "  df_prerec.columns = ['precision']\n",
        "  df_prerec['recall'] = all_arr_recall[m]\n",
        "  df_prerec['threshold'] = all_thresholds[m]\n",
        "  df_prerec\n",
        "  df_prerec.to_pickle(\"./bertclassify/cfpr/bert_best_10fold_precison_recall_curve_fold{}.pkl\".format(m+1))"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "NZtuwyjDF9Kz",
        "outputId": "34234b37-9ed7-4fed-a888-1b600a673c5f",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 54
        }
      },
      "source": [
        "!jupyter nbconvert --to html ./bertclassify/bert_classification_malay_10fold.ipynb"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "[NbConvertApp] Converting notebook ./bertclassify/bert_classification_malay_10fold.ipynb to html\n",
            "[NbConvertApp] Writing 997552 bytes to ./bertclassify/bert_classification_malay_10fold.html\n"
          ],
          "name": "stdout"
        }
      ]
    }
  ]
}