text-classification-bert.ipynb

{"nbformat":4,"nbformat_minor":0,"metadata":{"colab":{"name":"text-classification-bert.ipynb","provenance":[],"collapsed_sections":[],"authorship_tag":"ABX9TyM0AxEyO4uOPzssw1zvlK0c"},"kernelspec":{"name":"python3","display_name":"Python 3"},"accelerator":"GPU"},"cells":[{"cell_type":"markdown","metadata":{"id":"WM4RKDIODek-"},"source":["# Sentiment Analysis \n","\n","Text classification and sentiment analysis using BERT and pytorch\n","\n","BERT in comparison to ELMO uses transformer architechture, and doesn't use LSTMs. BERT uses contextual word embeddings which means that a single word could have multiple meanings, and BERT does account for that in the word embeddings based on the context of the sentence in which it occurs. \n","\n","Transformers feeds the input at once when compared to sequential model of LSTMs. They prove to perform better on tasks like Q/A systems, comprehension of text etc. \n","\n","## BERT :\n","\n","GPT model does L-R understanding of the text. Sometimes Q/A systems have answers after the current text read. BERT is **bidirectional** which means that it reads from both ends and captures the details in its entirity.\n","\n","It can be easily fine-tuned for the task that is required. BERT model was trained (pre-trained) using MLM (masked language modelling) and NSP (next sentence prediction).\n","\n","15% of the words were masked and the model was to predict the masked words. NSP was given 2 sentences and it had to predict whether or not the second sentence follows the first (in a sensible manner)\n","\n","## NOTATIONS \n","\n","[CLS] \\<sentence\\> [SEP]\n","\n","[CLS] token signifies the classification token. and [SEP] signifies a separater token.\n","\n","## BERT and Huggingface\n","\n","The two kinds of models : \n","* Bert Base - 12 encoders stacked\n","* Bert large - 24 encoders stacked"]},{"cell_type":"code","metadata":{"colab":{"base_uri":"https://localhost:8080/"},"id":"rHZAZdxPEF0E","executionInfo":{"status":"ok","timestamp":1608802458527,"user_tz":-330,"elapsed":4531,"user":{"displayName":"Mallika Subramanian","photoUrl":"https://lh3.googleusercontent.com/a-/AOh14Gj3lusT2xteZqIvBVDRAXN5RCnCIyEjAhNrS9hRhv0=s64","userId":"05149503641230543668"}},"outputId":"797e3969-eb55-420e-a812-30e211270dc7"},"source":["from google.colab import drive\n","drive.mount('/content/gdrive')"],"execution_count":1,"outputs":[{"output_type":"stream","text":["Drive already mounted at /content/gdrive; to attempt to forcibly remount, call drive.mount(\"/content/gdrive\", force_remount=True).\n"],"name":"stdout"}]},{"cell_type":"code","metadata":{"id":"Q6wjpCwqkL4x","executionInfo":{"status":"ok","timestamp":1608802460388,"user_tz":-330,"elapsed":6333,"user":{"displayName":"Mallika Subramanian","photoUrl":"https://lh3.googleusercontent.com/a-/AOh14Gj3lusT2xteZqIvBVDRAXN5RCnCIyEjAhNrS9hRhv0=s64","userId":"05149503641230543668"}}},"source":["import warnings\n","warnings.filterwarnings('ignore')"],"execution_count":2,"outputs":[]},{"cell_type":"code","metadata":{"id":"ZrMjIN8BEMYr","executionInfo":{"status":"ok","timestamp":1608802460389,"user_tz":-330,"elapsed":6269,"user":{"displayName":"Mallika Subramanian","photoUrl":"https://lh3.googleusercontent.com/a-/AOh14Gj3lusT2xteZqIvBVDRAXN5RCnCIyEjAhNrS9hRhv0=s64","userId":"05149503641230543668"}}},"source":["root_path = \"/content/gdrive/MyDrive/Collab-Notebooks/Sentiment-Analysis/\""],"execution_count":3,"outputs":[]},{"cell_type":"code","metadata":{"colab":{"base_uri":"https://localhost:8080/"},"id":"UYPrwDRYEUcy","executionInfo":{"status":"ok","timestamp":1608802460389,"user_tz":-330,"elapsed":6200,"user":{"displayName":"Mallika Subramanian","photoUrl":"https://lh3.googleusercontent.com/a-/AOh14Gj3lusT2xteZqIvBVDRAXN5RCnCIyEjAhNrS9hRhv0=s64","userId":"05149503641230543668"}},"outputId":"8346e72d-26ff-49f6-d623-a48bdb9d88da"},"source":["!pip install transformers"],"execution_count":4,"outputs":[{"output_type":"stream","text":["Requirement already satisfied: transformers in /usr/local/lib/python3.6/dist-packages (2.6.0)\n","Requirement already satisfied: sacremoses in /usr/local/lib/python3.6/dist-packages (from transformers) (0.0.43)\n","Requirement already satisfied: regex!=2019.12.17 in /usr/local/lib/python3.6/dist-packages (from transformers) (2019.12.20)\n","Requirement already satisfied: requests in /usr/local/lib/python3.6/dist-packages (from transformers) (2.23.0)\n","Requirement already satisfied: filelock in /usr/local/lib/python3.6/dist-packages (from transformers) (3.0.12)\n","Requirement already satisfied: sentencepiece in /usr/local/lib/python3.6/dist-packages (from transformers) (0.1.94)\n","Requirement already satisfied: tokenizers==0.5.2 in /usr/local/lib/python3.6/dist-packages (from transformers) (0.5.2)\n","Requirement already satisfied: numpy in /usr/local/lib/python3.6/dist-packages (from transformers) (1.19.4)\n","Requirement already satisfied: tqdm>=4.27 in /usr/local/lib/python3.6/dist-packages (from transformers) (4.41.1)\n","Requirement already satisfied: boto3 in /usr/local/lib/python3.6/dist-packages (from transformers) (1.16.43)\n","Requirement already satisfied: six in /usr/local/lib/python3.6/dist-packages (from sacremoses->transformers) (1.15.0)\n","Requirement already satisfied: click in /usr/local/lib/python3.6/dist-packages (from sacremoses->transformers) (7.1.2)\n","Requirement already satisfied: joblib in /usr/local/lib/python3.6/dist-packages (from sacremoses->transformers) (1.0.0)\n","Requirement already satisfied: chardet<4,>=3.0.2 in /usr/local/lib/python3.6/dist-packages (from requests->transformers) (3.0.4)\n","Requirement already satisfied: idna<3,>=2.5 in /usr/local/lib/python3.6/dist-packages (from requests->transformers) (2.10)\n","Requirement already satisfied: urllib3!=1.25.0,!=1.25.1,<1.26,>=1.21.1 in /usr/local/lib/python3.6/dist-packages (from requests->transformers) (1.24.3)\n","Requirement already satisfied: certifi>=2017.4.17 in /usr/local/lib/python3.6/dist-packages (from requests->transformers) (2020.12.5)\n","Requirement already satisfied: jmespath<1.0.0,>=0.7.1 in /usr/local/lib/python3.6/dist-packages (from boto3->transformers) (0.10.0)\n","Requirement already satisfied: s3transfer<0.4.0,>=0.3.0 in /usr/local/lib/python3.6/dist-packages (from boto3->transformers) (0.3.3)\n","Requirement already satisfied: botocore<1.20.0,>=1.19.43 in /usr/local/lib/python3.6/dist-packages (from boto3->transformers) (1.19.43)\n","Requirement already satisfied: python-dateutil<3.0.0,>=2.1 in /usr/local/lib/python3.6/dist-packages (from botocore<1.20.0,>=1.19.43->boto3->transformers) (2.8.1)\n"],"name":"stdout"}]},{"cell_type":"code","metadata":{"id":"kr6DbD4rEY6M","executionInfo":{"status":"ok","timestamp":1608802461476,"user_tz":-330,"elapsed":7219,"user":{"displayName":"Mallika Subramanian","photoUrl":"https://lh3.googleusercontent.com/a-/AOh14Gj3lusT2xteZqIvBVDRAXN5RCnCIyEjAhNrS9hRhv0=s64","userId":"05149503641230543668"}}},"source":["import transformers\n","from transformers import BertModel, BertTokenizer, AdamW, get_linear_schedule_with_warmup\n","import torch\n","\n","import numpy as np\n","import pandas as pd\n","import seaborn as sns\n","\n","from pylab import rcParams\n","\n","import matplotlib.pyplot as plt\n","from matplotlib import rc\n","\n","from sklearn.model_selection import train_test_split\n","from sklearn.metrics import confusion_matrix, classification_report\n","\n","from collections import defaultdict\n","from textwrap import wrap\n","from torch import nn, optim\n","from torch.utils.data import Dataset, DataLoader\n","\n","%matplotlib inline\n","%config InlineBackend.figure_format='retina'\n","\n","sns.set(style='whitegrid', palette='muted', font_scale=1.2)\n","HAPPY_COLORS_PALETTE = [\"#01BEFE\", \"#FFDD00\", \"#FF7D00\", \"#FF006D\", \"#ADFF02\", \"#8F00FF\"]\n","sns.set_palette(sns.color_palette(HAPPY_COLORS_PALETTE))\n","rcParams['figure.figsize'] = 12,8\n","\n","RANDOM_SEED = 42\n","np.random.seed(RANDOM_SEED)\n","torch.manual_seed(RANDOM_SEED)\n","\n","device = torch.device(\"cuda:0\" if torch.cuda.is_available() else \"cpu\")"],"execution_count":5,"outputs":[]},{"cell_type":"code","metadata":{"colab":{"base_uri":"https://localhost:8080/","height":658},"id":"WiOgALK2E76D","executionInfo":{"status":"ok","timestamp":1608802461479,"user_tz":-330,"elapsed":7131,"user":{"displayName":"Mallika Subramanian","photoUrl":"https://lh3.googleusercontent.com/a-/AOh14Gj3lusT2xteZqIvBVDRAXN5RCnCIyEjAhNrS9hRhv0=s64","userId":"05149503641230543668"}},"outputId":"62e52f99-077f-429c-e130-09648007b412"},"source":["df = pd.read_csv(root_path+\"reviews.csv\")\n","df.head()"],"execution_count":6,"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>reviewId</th>\n","      <th>userName</th>\n","      <th>userImage</th>\n","      <th>content</th>\n","      <th>score</th>\n","      <th>thumbsUpCount</th>\n","      <th>reviewCreatedVersion</th>\n","      <th>at</th>\n","      <th>replyContent</th>\n","      <th>repliedAt</th>\n","      <th>sortOrder</th>\n","      <th>appId</th>\n","    </tr>\n","  </thead>\n","  <tbody>\n","    <tr>\n","      <th>0</th>\n","      <td>gp:AOqpTOGOU1svA5gsifHKHFSziUGg7FDs9zR0HbVyXya...</td>\n","      <td>trina croll</td>\n","      <td>https://play-lh.googleusercontent.com/a-/AOh14...</td>\n","      <td>Can't get into my account and it's been days g...</td>\n","      <td>1</td>\n","      <td>8</td>\n","      <td>5.7.0.20</td>\n","      <td>2020-12-21 01:58:55</td>\n","      <td>NaN</td>\n","      <td>NaN</td>\n","      <td>most_relevant</td>\n","      <td>com.anydo</td>\n","    </tr>\n","    <tr>\n","      <th>1</th>\n","      <td>gp:AOqpTOEJfVU5Os7_8CNOu_mpy2AlDz8kFxEFMeBMnxC...</td>\n","      <td>Yiğitcan Sümbelli</td>\n","      <td>https://play-lh.googleusercontent.com/a-/AOh14...</td>\n","      <td>I am starting to hate this app! Every time the...</td>\n","      <td>1</td>\n","      <td>63</td>\n","      <td>5.7.0.10</td>\n","      <td>2020-12-02 15:52:24</td>\n","      <td>Hi, please note that these issues usually are ...</td>\n","      <td>2020-12-03 20:47:02</td>\n","      <td>most_relevant</td>\n","      <td>com.anydo</td>\n","    </tr>\n","    <tr>\n","      <th>2</th>\n","      <td>gp:AOqpTOHhd6uZpuhbBDDxEB1SwF8zXXUh8sg6AMGlkBx...</td>\n","      <td>Chris Keyes</td>\n","      <td>https://play-lh.googleusercontent.com/a-/AOh14...</td>\n","      <td>I paid premium for years. No more! This app ha...</td>\n","      <td>1</td>\n","      <td>24</td>\n","      <td>5.7.0.10</td>\n","      <td>2020-12-15 00:58:57</td>\n","      <td>Sorry you feel this way. Any.do does not chang...</td>\n","      <td>2020-12-17 09:28:09</td>\n","      <td>most_relevant</td>\n","      <td>com.anydo</td>\n","    </tr>\n","    <tr>\n","      <th>3</th>\n","      <td>gp:AOqpTOFSPaoOVarqwKF86e7-TPUyV6BanSXKRNFGb6G...</td>\n","      <td>Jon Pember</td>\n","      <td>https://play-lh.googleusercontent.com/-97UNn5D...</td>\n","      <td>Liked Any.do at first, enough that I even sign...</td>\n","      <td>1</td>\n","      <td>102</td>\n","      <td>5.7.0.10</td>\n","      <td>2020-11-22 23:42:56</td>\n","      <td>Please note that Any.do integrates directly wi...</td>\n","      <td>2020-11-25 13:30:01</td>\n","      <td>most_relevant</td>\n","      <td>com.anydo</td>\n","    </tr>\n","    <tr>\n","      <th>4</th>\n","      <td>gp:AOqpTOEA0ubPA5n0s_6lg0BHqfQqj0gi-KIdZF_I1yZ...</td>\n","      <td>Lisa Pappas</td>\n","      <td>https://play-lh.googleusercontent.com/a-/AOh14...</td>\n","      <td>Constant crashing. After reading all the negat...</td>\n","      <td>1</td>\n","      <td>25</td>\n","      <td>5.6.0.6</td>\n","      <td>2020-10-21 17:45:24</td>\n","      <td>NaN</td>\n","      <td>NaN</td>\n","      <td>most_relevant</td>\n","      <td>com.anydo</td>\n","    </tr>\n","  </tbody>\n","</table>\n","</div>"],"text/plain":["                                            reviewId  ...      appId\n","0  gp:AOqpTOGOU1svA5gsifHKHFSziUGg7FDs9zR0HbVyXya...  ...  com.anydo\n","1  gp:AOqpTOEJfVU5Os7_8CNOu_mpy2AlDz8kFxEFMeBMnxC...  ...  com.anydo\n","2  gp:AOqpTOHhd6uZpuhbBDDxEB1SwF8zXXUh8sg6AMGlkBx...  ...  com.anydo\n","3  gp:AOqpTOFSPaoOVarqwKF86e7-TPUyV6BanSXKRNFGb6G...  ...  com.anydo\n","4  gp:AOqpTOEA0ubPA5n0s_6lg0BHqfQqj0gi-KIdZF_I1yZ...  ...  com.anydo\n","\n","[5 rows x 12 columns]"]},"metadata":{"tags":[]},"execution_count":6}]},{"cell_type":"code","metadata":{"colab":{"base_uri":"https://localhost:8080/"},"id":"jP4DBmpBFGiy","executionInfo":{"status":"ok","timestamp":1608802461481,"user_tz":-330,"elapsed":7066,"user":{"displayName":"Mallika Subramanian","photoUrl":"https://lh3.googleusercontent.com/a-/AOh14Gj3lusT2xteZqIvBVDRAXN5RCnCIyEjAhNrS9hRhv0=s64","userId":"05149503641230543668"}},"outputId":"807a5bdc-83c2-4900-9ce5-2f596b6586e1"},"source":["df.shape"],"execution_count":7,"outputs":[{"output_type":"execute_result","data":{"text/plain":["(16354, 12)"]},"metadata":{"tags":[]},"execution_count":7}]},{"cell_type":"code","metadata":{"colab":{"base_uri":"https://localhost:8080/"},"id":"8oWE_OMtFJ-D","executionInfo":{"status":"ok","timestamp":1608802461482,"user_tz":-330,"elapsed":7013,"user":{"displayName":"Mallika Subramanian","photoUrl":"https://lh3.googleusercontent.com/a-/AOh14Gj3lusT2xteZqIvBVDRAXN5RCnCIyEjAhNrS9hRhv0=s64","userId":"05149503641230543668"}},"outputId":"442bf965-c87f-45c4-849e-c76caf56d2ee"},"source":["df.info()"],"execution_count":8,"outputs":[{"output_type":"stream","text":["<class 'pandas.core.frame.DataFrame'>\n","RangeIndex: 16354 entries, 0 to 16353\n","Data columns (total 12 columns):\n"," #   Column                Non-Null Count  Dtype \n","---  ------                --------------  ----- \n"," 0   reviewId              16354 non-null  object\n"," 1   userName              16354 non-null  object\n"," 2   userImage             16354 non-null  object\n"," 3   content               16354 non-null  object\n"," 4   score                 16354 non-null  int64 \n"," 5   thumbsUpCount         16354 non-null  int64 \n"," 6   reviewCreatedVersion  13988 non-null  object\n"," 7   at                    16354 non-null  object\n"," 8   replyContent          8740 non-null   object\n"," 9   repliedAt             8740 non-null   object\n"," 10  sortOrder             16354 non-null  object\n"," 11  appId                 16354 non-null  object\n","dtypes: int64(2), object(10)\n","memory usage: 1.5+ MB\n"],"name":"stdout"}]},{"cell_type":"markdown","metadata":{"id":"djzQKr-CFXZg"},"source":["To check whether or not our dataset is balanced in terms of the dataset distribution that we have. We can check this with respect to the count of the different scores.\n","\n","The distribution seems reasonable since (1,2) will be grouped and (4,5) will be grouped and 3 is double of that"]},{"cell_type":"code","metadata":{"colab":{"base_uri":"https://localhost:8080/","height":523},"id":"n1rTpSOrFM0a","executionInfo":{"status":"ok","timestamp":1608802462002,"user_tz":-330,"elapsed":7392,"user":{"displayName":"Mallika Subramanian","photoUrl":"https://lh3.googleusercontent.com/a-/AOh14Gj3lusT2xteZqIvBVDRAXN5RCnCIyEjAhNrS9hRhv0=s64","userId":"05149503641230543668"}},"outputId":"ffa13781-b182-4dd8-f458-86ca15013a6b"},"source":["sns.countplot(df.score)\n","plt.xlabel('review score')"],"execution_count":9,"outputs":[{"output_type":"execute_result","data":{"text/plain":["Text(0.5, 0, 'review score')"]},"metadata":{"tags":[]},"execution_count":9},{"output_type":"display_data","data":{"image/png":"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\n","text/plain":["<Figure size 864x576 with 1 Axes>"]},"metadata":{"tags":[],"image/png":{"width":745,"height":489}}}]},{"cell_type":"code","metadata":{"id":"oSYecX2nFqq1","executionInfo":{"status":"ok","timestamp":1608802462002,"user_tz":-330,"elapsed":7322,"user":{"displayName":"Mallika Subramanian","photoUrl":"https://lh3.googleusercontent.com/a-/AOh14Gj3lusT2xteZqIvBVDRAXN5RCnCIyEjAhNrS9hRhv0=s64","userId":"05149503641230543668"}}},"source":["def to_sentiment(rating):\n","  rating = int(rating)\n","  if rating <=2:\n","    return 0\n","  elif rating == 3:\n","    return 1\n","  else:\n","    return 2"],"execution_count":10,"outputs":[]},{"cell_type":"code","metadata":{"id":"qELIG5-0F_TA","executionInfo":{"status":"ok","timestamp":1608802462003,"user_tz":-330,"elapsed":7266,"user":{"displayName":"Mallika Subramanian","photoUrl":"https://lh3.googleusercontent.com/a-/AOh14Gj3lusT2xteZqIvBVDRAXN5RCnCIyEjAhNrS9hRhv0=s64","userId":"05149503641230543668"}}},"source":["df['sentiment'] = df.score.apply(to_sentiment)"],"execution_count":11,"outputs":[]},{"cell_type":"code","metadata":{"id":"4ZoSD9E8GKCM","executionInfo":{"status":"ok","timestamp":1608802462004,"user_tz":-330,"elapsed":7205,"user":{"displayName":"Mallika Subramanian","photoUrl":"https://lh3.googleusercontent.com/a-/AOh14Gj3lusT2xteZqIvBVDRAXN5RCnCIyEjAhNrS9hRhv0=s64","userId":"05149503641230543668"}}},"source":["class_names = ['negative','neutral','positive']"],"execution_count":12,"outputs":[]},{"cell_type":"code","metadata":{"colab":{"base_uri":"https://localhost:8080/","height":523},"id":"sH401RXpGCxk","executionInfo":{"status":"ok","timestamp":1608802462605,"user_tz":-330,"elapsed":7745,"user":{"displayName":"Mallika Subramanian","photoUrl":"https://lh3.googleusercontent.com/a-/AOh14Gj3lusT2xteZqIvBVDRAXN5RCnCIyEjAhNrS9hRhv0=s64","userId":"05149503641230543668"}},"outputId":"fe5290a3-89bb-4b36-df40-78688f93a4ed"},"source":["ax =sns.countplot(df.sentiment)\n","plt.xlabel('review sentiment')\n","ax.set_xticklabels(class_names)"],"execution_count":13,"outputs":[{"output_type":"execute_result","data":{"text/plain":["[Text(0, 0, 'negative'), Text(0, 0, 'neutral'), Text(0, 0, 'positive')]"]},"metadata":{"tags":[]},"execution_count":13},{"output_type":"display_data","data":{"image/png":"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\n","text/plain":["<Figure size 864x576 with 1 Axes>"]},"metadata":{"tags":[],"image/png":{"width":745,"height":489}}}]},{"cell_type":"markdown","metadata":{"id":"Flny3DHyGdgr"},"source":["# Data Preprocessing\n","\n","BERT transformer requires some special work on the text as preprocessing. Pretrained on v large dataset; and we are using it for sequence classification"]},{"cell_type":"code","metadata":{"id":"DJYZjI3uGQ8A","executionInfo":{"status":"ok","timestamp":1608802462606,"user_tz":-330,"elapsed":7615,"user":{"displayName":"Mallika Subramanian","photoUrl":"https://lh3.googleusercontent.com/a-/AOh14Gj3lusT2xteZqIvBVDRAXN5RCnCIyEjAhNrS9hRhv0=s64","userId":"05149503641230543668"}}},"source":["#starting with the tokenizer\n","PRE_TRAINED_MODEL_NAME = 'bert-base-cased'\n","tokenizer = BertTokenizer.from_pretrained(PRE_TRAINED_MODEL_NAME)"],"execution_count":14,"outputs":[]},{"cell_type":"markdown","metadata":{"id":"Jyp0lZa1Hr-d"},"source":["\n","BERT requires special tokens which indicate separation between sentences.\n","This comes from the fact that BERT was initially built for Q/A systems and while\n","doing classification of sequences BERT requires a tokens\n","\n","Seuqences must be of equal length, which means we'll have to do padding which will\n","append tokens of 0 meaning to the end of our sentences that we pass into the model\n","\n","We also require an attention mask - each token that has a meaning should have a value\n","of 1 in the attention mask, else if it is a padding token attention mask should be 0\n","\n","* Use a small piece of text to show how the tokenizer splits and works with the text\n","\n","* Building a dataset, that will take the reviews, use the tokenizer internally, and build a dataset that is compatible with pytorch"]},{"cell_type":"code","metadata":{"id":"tnbUvrUsHCM4","executionInfo":{"status":"ok","timestamp":1608802462607,"user_tz":-330,"elapsed":7505,"user":{"displayName":"Mallika Subramanian","photoUrl":"https://lh3.googleusercontent.com/a-/AOh14Gj3lusT2xteZqIvBVDRAXN5RCnCIyEjAhNrS9hRhv0=s64","userId":"05149503641230543668"}}},"source":["sample_txt = \"When was I last outside? I am stuck at home for 2 weeks.\""],"execution_count":15,"outputs":[]},{"cell_type":"code","metadata":{"id":"NvtpmkqTIlpF","executionInfo":{"status":"ok","timestamp":1608802462607,"user_tz":-330,"elapsed":7441,"user":{"displayName":"Mallika Subramanian","photoUrl":"https://lh3.googleusercontent.com/a-/AOh14Gj3lusT2xteZqIvBVDRAXN5RCnCIyEjAhNrS9hRhv0=s64","userId":"05149503641230543668"}}},"source":["tokens = tokenizer.tokenize(sample_txt)"],"execution_count":16,"outputs":[]},{"cell_type":"code","metadata":{"colab":{"base_uri":"https://localhost:8080/"},"id":"WYhYuvarIpCM","executionInfo":{"status":"ok","timestamp":1608802462608,"user_tz":-330,"elapsed":7387,"user":{"displayName":"Mallika Subramanian","photoUrl":"https://lh3.googleusercontent.com/a-/AOh14Gj3lusT2xteZqIvBVDRAXN5RCnCIyEjAhNrS9hRhv0=s64","userId":"05149503641230543668"}},"outputId":"d4f4118f-8729-466a-da28-5fdcea45905b"},"source":["print(tokens)\n","#keeps all the punctuation here"],"execution_count":17,"outputs":[{"output_type":"stream","text":["['When', 'was', 'I', 'last', 'outside', '?', 'I', 'am', 'stuck', 'at', 'home', 'for', '2', 'weeks', '.']\n"],"name":"stdout"}]},{"cell_type":"code","metadata":{"colab":{"base_uri":"https://localhost:8080/"},"id":"vG1U9PRmIrAM","executionInfo":{"status":"ok","timestamp":1608802462609,"user_tz":-330,"elapsed":7313,"user":{"displayName":"Mallika Subramanian","photoUrl":"https://lh3.googleusercontent.com/a-/AOh14Gj3lusT2xteZqIvBVDRAXN5RCnCIyEjAhNrS9hRhv0=s64","userId":"05149503641230543668"}},"outputId":"262340e9-fecb-4618-933a-add1427fd097"},"source":["#converting tokens to ids to pass to the bert model later\n","\n","token_ids = tokenizer.convert_tokens_to_ids(tokens)\n","print(len(token_ids))\n","print(len(tokens))\n","print(token_ids)"],"execution_count":18,"outputs":[{"output_type":"stream","text":["15\n","15\n","[1332, 1108, 146, 1314, 1796, 136, 146, 1821, 5342, 1120, 1313, 1111, 123, 2277, 119]\n"],"name":"stdout"}]},{"cell_type":"markdown","metadata":{"id":"8mX4gOrJJN0I"},"source":["## Special Tokens\n","\n","* SEP tokens\n","* Classification token\n","* Padding token\n","* Unknown token for something that is unrecognized\n","\n","We use the ecode plus method to create all the steps for tokenization and data pre processing"]},{"cell_type":"code","metadata":{"colab":{"base_uri":"https://localhost:8080/"},"id":"aFUxytNsI70-","executionInfo":{"status":"ok","timestamp":1608802462609,"user_tz":-330,"elapsed":7173,"user":{"displayName":"Mallika Subramanian","photoUrl":"https://lh3.googleusercontent.com/a-/AOh14Gj3lusT2xteZqIvBVDRAXN5RCnCIyEjAhNrS9hRhv0=s64","userId":"05149503641230543668"}},"outputId":"01ad9267-3446-4b87-9dd8-e62ff93e4ef4"},"source":["tokenizer.sep_token, tokenizer.sep_token_id"],"execution_count":19,"outputs":[{"output_type":"execute_result","data":{"text/plain":["('[SEP]', 102)"]},"metadata":{"tags":[]},"execution_count":19}]},{"cell_type":"code","metadata":{"colab":{"base_uri":"https://localhost:8080/"},"id":"1GIadKFpJctY","executionInfo":{"status":"ok","timestamp":1608802462609,"user_tz":-330,"elapsed":7108,"user":{"displayName":"Mallika Subramanian","photoUrl":"https://lh3.googleusercontent.com/a-/AOh14Gj3lusT2xteZqIvBVDRAXN5RCnCIyEjAhNrS9hRhv0=s64","userId":"05149503641230543668"}},"outputId":"29cea6d7-facc-48a7-fd7a-082d1134c706"},"source":["tokenizer.cls_token, tokenizer.cls_token_id"],"execution_count":20,"outputs":[{"output_type":"execute_result","data":{"text/plain":["('[CLS]', 101)"]},"metadata":{"tags":[]},"execution_count":20}]},{"cell_type":"code","metadata":{"colab":{"base_uri":"https://localhost:8080/"},"id":"Jb4C0sLzJfcT","executionInfo":{"status":"ok","timestamp":1608802463008,"user_tz":-330,"elapsed":7442,"user":{"displayName":"Mallika Subramanian","photoUrl":"https://lh3.googleusercontent.com/a-/AOh14Gj3lusT2xteZqIvBVDRAXN5RCnCIyEjAhNrS9hRhv0=s64","userId":"05149503641230543668"}},"outputId":"0c7f3404-9dc5-4b6e-8832-b39f324a7445"},"source":["tokenizer.pad_token, tokenizer.pad_token_id"],"execution_count":21,"outputs":[{"output_type":"execute_result","data":{"text/plain":["('[PAD]', 0)"]},"metadata":{"tags":[]},"execution_count":21}]},{"cell_type":"code","metadata":{"colab":{"base_uri":"https://localhost:8080/"},"id":"G30oy-z9JnTB","executionInfo":{"status":"ok","timestamp":1608802463009,"user_tz":-330,"elapsed":7368,"user":{"displayName":"Mallika Subramanian","photoUrl":"https://lh3.googleusercontent.com/a-/AOh14Gj3lusT2xteZqIvBVDRAXN5RCnCIyEjAhNrS9hRhv0=s64","userId":"05149503641230543668"}},"outputId":"8a51343d-ddd7-44ad-93d0-a0882bfc3240"},"source":["tokenizer.unk_token, tokenizer.unk_token_id"],"execution_count":22,"outputs":[{"output_type":"execute_result","data":{"text/plain":["('[UNK]', 100)"]},"metadata":{"tags":[]},"execution_count":22}]},{"cell_type":"code","metadata":{"colab":{"base_uri":"https://localhost:8080/"},"id":"paoYHY3vJrXN","executionInfo":{"status":"ok","timestamp":1608802463010,"user_tz":-330,"elapsed":7307,"user":{"displayName":"Mallika Subramanian","photoUrl":"https://lh3.googleusercontent.com/a-/AOh14Gj3lusT2xteZqIvBVDRAXN5RCnCIyEjAhNrS9hRhv0=s64","userId":"05149503641230543668"}},"outputId":"62cbc8e0-8b01-44d8-911c-ff1064284115"},"source":["encoding = tokenizer.encode_plus(\n","    sample_txt,\n","    max_length = 32,\n","    add_special_tokens=True,\n","    pad_to_max_length = True,\n","    return_attention_mask = True,\n","    return_token_type_ids = False,\n","    return_tensors = 'pt'\n",")\n","\n","encoding.keys()"],"execution_count":23,"outputs":[{"output_type":"execute_result","data":{"text/plain":["dict_keys(['input_ids', 'attention_mask'])"]},"metadata":{"tags":[]},"execution_count":23}]},{"cell_type":"code","metadata":{"colab":{"base_uri":"https://localhost:8080/"},"id":"_bkWdtgLKfbf","executionInfo":{"status":"ok","timestamp":1608802463010,"user_tz":-330,"elapsed":7238,"user":{"displayName":"Mallika Subramanian","photoUrl":"https://lh3.googleusercontent.com/a-/AOh14Gj3lusT2xteZqIvBVDRAXN5RCnCIyEjAhNrS9hRhv0=s64","userId":"05149503641230543668"}},"outputId":"1be2214f-af7c-4f80-a437-bd3dda7e1886"},"source":["encoding['input_ids']"],"execution_count":24,"outputs":[{"output_type":"execute_result","data":{"text/plain":["tensor([[ 101, 1332, 1108,  146, 1314, 1796,  136,  146, 1821, 5342, 1120, 1313,\n","         1111,  123, 2277,  119,  102,    0,    0,    0,    0,    0,    0,    0,\n","            0,    0,    0,    0,    0,    0,    0,    0]])"]},"metadata":{"tags":[]},"execution_count":24}]},{"cell_type":"code","metadata":{"colab":{"base_uri":"https://localhost:8080/"},"id":"0dRoFwLrKrFz","executionInfo":{"status":"ok","timestamp":1608802463011,"user_tz":-330,"elapsed":7169,"user":{"displayName":"Mallika Subramanian","photoUrl":"https://lh3.googleusercontent.com/a-/AOh14Gj3lusT2xteZqIvBVDRAXN5RCnCIyEjAhNrS9hRhv0=s64","userId":"05149503641230543668"}},"outputId":"0e46c743-7416-42d1-cb58-638c68f7239f"},"source":["len(encoding['input_ids'][0])"],"execution_count":25,"outputs":[{"output_type":"execute_result","data":{"text/plain":["32"]},"metadata":{"tags":[]},"execution_count":25}]},{"cell_type":"code","metadata":{"colab":{"base_uri":"https://localhost:8080/"},"id":"diatFMHTK9Mt","executionInfo":{"status":"ok","timestamp":1608802463011,"user_tz":-330,"elapsed":7091,"user":{"displayName":"Mallika Subramanian","photoUrl":"https://lh3.googleusercontent.com/a-/AOh14Gj3lusT2xteZqIvBVDRAXN5RCnCIyEjAhNrS9hRhv0=s64","userId":"05149503641230543668"}},"outputId":"06dce060-2887-4e60-c255-059563d5ad9b"},"source":["encoding['attention_mask']"],"execution_count":26,"outputs":[{"output_type":"execute_result","data":{"text/plain":["tensor([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 0, 0, 0, 0, 0, 0, 0,\n","         0, 0, 0, 0, 0, 0, 0, 0]])"]},"metadata":{"tags":[]},"execution_count":26}]},{"cell_type":"markdown","metadata":{"id":"XSHGtQrTLJ41"},"source":["### Choosing Sequence Length\n","We can look at the distribution of the sentences in our token lengths and the choose one ideal one\n","\n","We will just use the encode method of the tokenizer here"]},{"cell_type":"code","metadata":{"id":"W0GzOR3_K_ZU","executionInfo":{"status":"ok","timestamp":1608802470828,"user_tz":-330,"elapsed":14763,"user":{"displayName":"Mallika Subramanian","photoUrl":"https://lh3.googleusercontent.com/a-/AOh14Gj3lusT2xteZqIvBVDRAXN5RCnCIyEjAhNrS9hRhv0=s64","userId":"05149503641230543668"}}},"source":["token_lens = []\n","\n","for txt in df.content:\n","  tokens = tokenizer.encode(txt, max_length=512, truncation=True)\n","  token_lens.append(len(tokens))"],"execution_count":27,"outputs":[]},{"cell_type":"code","metadata":{"colab":{"base_uri":"https://localhost:8080/","height":505},"id":"RGTsCWhWLgN8","executionInfo":{"status":"ok","timestamp":1608802471615,"user_tz":-330,"elapsed":15468,"user":{"displayName":"Mallika Subramanian","photoUrl":"https://lh3.googleusercontent.com/a-/AOh14Gj3lusT2xteZqIvBVDRAXN5RCnCIyEjAhNrS9hRhv0=s64","userId":"05149503641230543668"}},"outputId":"6369a8f9-8491-44af-8c36-f5fe1052517d"},"source":["sns.distplot(token_lens)"],"execution_count":28,"outputs":[{"output_type":"execute_result","data":{"text/plain":["<matplotlib.axes._subplots.AxesSubplot at 0x7f11d1af0dd8>"]},"metadata":{"tags":[]},"execution_count":28},{"output_type":"display_data","data":{"image/png":"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\n","text/plain":["<Figure size 864x576 with 1 Axes>"]},"metadata":{"tags":[],"image/png":{"width":749,"height":471}}}]},{"cell_type":"markdown","metadata":{"id":"Yquhe6J2Lwqy"},"source":["Here the max length of 128 seems suitable enough for our distribution of tokens and initially, we can choose a len that may clip some tokens but allows the model to train relatively faster and also helps build a faster prototype for testing"]},{"cell_type":"markdown","metadata":{"id":"8Y8WDL1VMBdB"},"source":["## Creating a DATSET"]},{"cell_type":"code","metadata":{"id":"_i9JLgcnLn3_","executionInfo":{"status":"ok","timestamp":1608802471617,"user_tz":-330,"elapsed":15240,"user":{"displayName":"Mallika Subramanian","photoUrl":"https://lh3.googleusercontent.com/a-/AOh14Gj3lusT2xteZqIvBVDRAXN5RCnCIyEjAhNrS9hRhv0=s64","userId":"05149503641230543668"}}},"source":["class GPReviewDataset(Dataset):\n","\n","\n","  def __init__(self, review, targets, tokenizer, max_len):\n","    self.review = review\n","    self.targets = targets\n","    self.tokenizer = tokenizer\n","    self.max_len = max_len\n","\n","  def __len__(self):\n","    return len(self.review)\n","\n","  #takes the index of the element from the dataset\n","  def __getitem__(self, item):\n","    review = str(self.review[item])\n","    target = self.targets[item]\n","\n","    encoding = tokenizer.encode_plus(\n","      review,\n","      max_length = self.max_len,\n","      add_special_tokens=True,\n","      pad_to_max_length = True,\n","      return_attention_mask = True,\n","      return_token_type_ids = False,\n","      return_tensors = 'pt'\n","    )\n","    return {\n","        'review_text' : review,\n","        'input_ids' : encoding['input_ids'].flatten(),\n","        'attention_mask' : encoding['attention_mask'].flatten(),\n","        'targets' : torch.tensor(target, dtype=torch.long)\n","    }\n","  "],"execution_count":29,"outputs":[]},{"cell_type":"code","metadata":{"id":"g6Acp1z2M7_S","executionInfo":{"status":"ok","timestamp":1608802471620,"user_tz":-330,"elapsed":15181,"user":{"displayName":"Mallika Subramanian","photoUrl":"https://lh3.googleusercontent.com/a-/AOh14Gj3lusT2xteZqIvBVDRAXN5RCnCIyEjAhNrS9hRhv0=s64","userId":"05149503641230543668"}}},"source":["MAX_LEN = 160\n","BATCH_SIZE = 16\n","EPOCHS = 50"],"execution_count":30,"outputs":[]},{"cell_type":"code","metadata":{"id":"YOIjJ8SsNMHG","executionInfo":{"status":"ok","timestamp":1608802471621,"user_tz":-330,"elapsed":15116,"user":{"displayName":"Mallika Subramanian","photoUrl":"https://lh3.googleusercontent.com/a-/AOh14Gj3lusT2xteZqIvBVDRAXN5RCnCIyEjAhNrS9hRhv0=s64","userId":"05149503641230543668"}}},"source":["#splitting into train, test, val\n","\n","df_train, df_test = train_test_split(df, test_size=0.1, random_state=RANDOM_SEED)\n","df_val, df_test = train_test_split(df_test, test_size = 0.5, random_state=RANDOM_SEED)"],"execution_count":31,"outputs":[]},{"cell_type":"code","metadata":{"colab":{"base_uri":"https://localhost:8080/"},"id":"6GI0qAhfNfHr","executionInfo":{"status":"ok","timestamp":1608802471622,"user_tz":-330,"elapsed":15055,"user":{"displayName":"Mallika Subramanian","photoUrl":"https://lh3.googleusercontent.com/a-/AOh14Gj3lusT2xteZqIvBVDRAXN5RCnCIyEjAhNrS9hRhv0=s64","userId":"05149503641230543668"}},"outputId":"aefb05c8-73c2-49d1-90a0-412cb75c1367"},"source":["df_train.shape, df_val.shape, df_test.shape"],"execution_count":32,"outputs":[{"output_type":"execute_result","data":{"text/plain":["((14718, 13), (818, 13), (818, 13))"]},"metadata":{"tags":[]},"execution_count":32}]},{"cell_type":"code","metadata":{"id":"PD1k6TK8NiQc","executionInfo":{"status":"ok","timestamp":1608802471623,"user_tz":-330,"elapsed":14988,"user":{"displayName":"Mallika Subramanian","photoUrl":"https://lh3.googleusercontent.com/a-/AOh14Gj3lusT2xteZqIvBVDRAXN5RCnCIyEjAhNrS9hRhv0=s64","userId":"05149503641230543668"}}},"source":["'''\n","wrapping all the dataframes into the pytorch dataloaders \n","to be able to load and use in the bert pretrained model\n","'''\n","\n","def create_data_loader(df, tokenizer, max_len, batch_size):\n","  ds = GPReviewDataset(\n","      review = df.content.to_numpy(),\n","      targets = df.sentiment.to_numpy(),\n","      tokenizer = tokenizer,\n","      max_len = max_len\n","  )\n","\n","  return DataLoader(\n","      ds,\n","      batch_size = batch_size,\n","      num_workers = 4\n","  )"],"execution_count":33,"outputs":[]},{"cell_type":"code","metadata":{"id":"QxevnPG1OI-7","executionInfo":{"status":"ok","timestamp":1608802471624,"user_tz":-330,"elapsed":14924,"user":{"displayName":"Mallika Subramanian","photoUrl":"https://lh3.googleusercontent.com/a-/AOh14Gj3lusT2xteZqIvBVDRAXN5RCnCIyEjAhNrS9hRhv0=s64","userId":"05149503641230543668"}}},"source":["train_data_loader = create_data_loader(df_train, tokenizer, MAX_LEN, BATCH_SIZE)\n","val_data_loader = create_data_loader(df_val, tokenizer, MAX_LEN, BATCH_SIZE)\n","test_data_loader = create_data_loader(df_test, tokenizer, MAX_LEN, BATCH_SIZE)"],"execution_count":34,"outputs":[]},{"cell_type":"code","metadata":{"colab":{"base_uri":"https://localhost:8080/"},"id":"GdwKE0X_OdV9","executionInfo":{"status":"ok","timestamp":1608802472875,"user_tz":-330,"elapsed":16110,"user":{"displayName":"Mallika Subramanian","photoUrl":"https://lh3.googleusercontent.com/a-/AOh14Gj3lusT2xteZqIvBVDRAXN5RCnCIyEjAhNrS9hRhv0=s64","userId":"05149503641230543668"}},"outputId":"24136549-7261-4973-ccbb-feafdc4daec6"},"source":["data = next(iter(train_data_loader))\n","data = next(iter(train_data_loader))\n","data = next(iter(train_data_loader))\n","data = next(iter(train_data_loader))\n","data = next(iter(train_data_loader))\n","data = next(iter(train_data_loader))\n","data = next(iter(train_data_loader))\n","data = next(iter(train_data_loader))\n","data = next(iter(train_data_loader))\n","data = next(iter(train_data_loader))\n","data.keys()"],"execution_count":35,"outputs":[{"output_type":"execute_result","data":{"text/plain":["dict_keys(['review_text', 'input_ids', 'attention_mask', 'targets'])"]},"metadata":{"tags":[]},"execution_count":35}]},{"cell_type":"code","metadata":{"colab":{"base_uri":"https://localhost:8080/"},"id":"mD9wxdvvOlrN","executionInfo":{"status":"ok","timestamp":1608802472877,"user_tz":-330,"elapsed":16032,"user":{"displayName":"Mallika Subramanian","photoUrl":"https://lh3.googleusercontent.com/a-/AOh14Gj3lusT2xteZqIvBVDRAXN5RCnCIyEjAhNrS9hRhv0=s64","userId":"05149503641230543668"}},"outputId":"468c314b-9a4d-4b8e-edc9-e37e8d0420c8"},"source":["print(data['input_ids'].shape)"],"execution_count":36,"outputs":[{"output_type":"stream","text":["torch.Size([16, 160])\n"],"name":"stdout"}]},{"cell_type":"code","metadata":{"id":"aQL6YBAtOvzi","executionInfo":{"status":"ok","timestamp":1608802476084,"user_tz":-330,"elapsed":19169,"user":{"displayName":"Mallika Subramanian","photoUrl":"https://lh3.googleusercontent.com/a-/AOh14Gj3lusT2xteZqIvBVDRAXN5RCnCIyEjAhNrS9hRhv0=s64","userId":"05149503641230543668"}}},"source":["#initializing a BERT model\n","\n","bert_model = BertModel.from_pretrained(PRE_TRAINED_MODEL_NAME)"],"execution_count":37,"outputs":[]},{"cell_type":"code","metadata":{"id":"YHeFIINjUZrt","executionInfo":{"status":"ok","timestamp":1608802476086,"user_tz":-330,"elapsed":19103,"user":{"displayName":"Mallika Subramanian","photoUrl":"https://lh3.googleusercontent.com/a-/AOh14Gj3lusT2xteZqIvBVDRAXN5RCnCIyEjAhNrS9hRhv0=s64","userId":"05149503641230543668"}}},"source":["#returns the last hidden state from the last encoder in the transformer\n","# also returns the pooled output\n","\n","outputs = bert_model(\n","  input_ids=encoding['input_ids'],\n","  attention_mask=encoding['attention_mask']\n",")"],"execution_count":38,"outputs":[]},{"cell_type":"code","metadata":{"id":"Jq8RDT5kUvkL","executionInfo":{"status":"ok","timestamp":1608802476089,"user_tz":-330,"elapsed":19038,"user":{"displayName":"Mallika Subramanian","photoUrl":"https://lh3.googleusercontent.com/a-/AOh14Gj3lusT2xteZqIvBVDRAXN5RCnCIyEjAhNrS9hRhv0=s64","userId":"05149503641230543668"}}},"source":["last_hidden_state = outputs[0]\n","pooled_output = outputs[1]"],"execution_count":39,"outputs":[]},{"cell_type":"code","metadata":{"colab":{"base_uri":"https://localhost:8080/"},"id":"CUoHjEmqWEDg","executionInfo":{"status":"ok","timestamp":1608802476090,"user_tz":-330,"elapsed":18963,"user":{"displayName":"Mallika Subramanian","photoUrl":"https://lh3.googleusercontent.com/a-/AOh14Gj3lusT2xteZqIvBVDRAXN5RCnCIyEjAhNrS9hRhv0=s64","userId":"05149503641230543668"}},"outputId":"5f01854f-2094-4fc9-a93d-fe174cc0c9fa"},"source":["last_hidden_state.shape"],"execution_count":40,"outputs":[{"output_type":"execute_result","data":{"text/plain":["torch.Size([1, 32, 768])"]},"metadata":{"tags":[]},"execution_count":40}]},{"cell_type":"markdown","metadata":{"id":"bRgUuq05XGnp"},"source":["768 hidden units in the last state of the topmost encoder"]},{"cell_type":"code","metadata":{"colab":{"base_uri":"https://localhost:8080/"},"id":"uwChOjXTWRkF","executionInfo":{"status":"ok","timestamp":1608802476091,"user_tz":-330,"elapsed":18801,"user":{"displayName":"Mallika Subramanian","photoUrl":"https://lh3.googleusercontent.com/a-/AOh14Gj3lusT2xteZqIvBVDRAXN5RCnCIyEjAhNrS9hRhv0=s64","userId":"05149503641230543668"}},"outputId":"9b77fe5b-e5f7-4da6-dc1b-3aafdab8f985"},"source":["pooled_output.shape"],"execution_count":41,"outputs":[{"output_type":"execute_result","data":{"text/plain":["torch.Size([1, 768])"]},"metadata":{"tags":[]},"execution_count":41}]},{"cell_type":"markdown","metadata":{"id":"HtnjQJOVXO_5"},"source":["Pooling procedure over the 32 elements into this dimension. It takes the hidden size of the 768 and applying tanh function over it : getting a summary of the hidden state"]},{"cell_type":"code","metadata":{"colab":{"base_uri":"https://localhost:8080/"},"id":"2OqJ3hEMXDOF","executionInfo":{"status":"ok","timestamp":1608802476092,"user_tz":-330,"elapsed":18648,"user":{"displayName":"Mallika Subramanian","photoUrl":"https://lh3.googleusercontent.com/a-/AOh14Gj3lusT2xteZqIvBVDRAXN5RCnCIyEjAhNrS9hRhv0=s64","userId":"05149503641230543668"}},"outputId":"1b85bc0d-f2e4-4709-b51a-b80f75fcd1d6"},"source":["bert_model.config.hidden_size"],"execution_count":42,"outputs":[{"output_type":"execute_result","data":{"text/plain":["768"]},"metadata":{"tags":[]},"execution_count":42}]},{"cell_type":"markdown","metadata":{"id":"YE8BZJcrXl0o"},"source":["## Building the Classifier\n","\n","Creating a new class - extension of the torch nn module. \n","* accepting number of classes that exist\n","* feed the data to BERT model\n","* apply drop out for regularization\n","* apply a softmax function for the final classification"]},{"cell_type":"code","metadata":{"id":"cWUsUOZFXkKN","executionInfo":{"status":"ok","timestamp":1608802476093,"user_tz":-330,"elapsed":18534,"user":{"displayName":"Mallika Subramanian","photoUrl":"https://lh3.googleusercontent.com/a-/AOh14Gj3lusT2xteZqIvBVDRAXN5RCnCIyEjAhNrS9hRhv0=s64","userId":"05149503641230543668"}}},"source":["class SentimentClassifier(nn.Module):\n","\n","  def __init__(self, n_classes):\n","    super(SentimentClassifier,self).__init__()\n","    self.bert = BertModel.from_pretrained(PRE_TRAINED_MODEL_NAME)\n","    self.drop = nn.Dropout(p=0.3)\n","    self.out = nn.Linear(self.bert.config.hidden_size, n_classes)\n","    self.softmax = nn.Softmax(dim = 1)\n","\n","  def forward(self, input_ids, attention_mask):\n","    outputs = self.bert(\n","        input_ids = input_ids,\n","        attention_mask = attention_mask\n","    )\n","    pooled_output = outputs[1]\n","\n","    output = self.drop(pooled_output)\n","    output = self.out(output)\n","    return self.softmax(output)"],"execution_count":43,"outputs":[]},{"cell_type":"code","metadata":{"id":"Y6PaBmm5Y9dW","executionInfo":{"status":"ok","timestamp":1608802484286,"user_tz":-330,"elapsed":26651,"user":{"displayName":"Mallika Subramanian","photoUrl":"https://lh3.googleusercontent.com/a-/AOh14Gj3lusT2xteZqIvBVDRAXN5RCnCIyEjAhNrS9hRhv0=s64","userId":"05149503641230543668"}}},"source":["model = SentimentClassifier(len(class_names))\n","model = model.to(device)"],"execution_count":44,"outputs":[]},{"cell_type":"code","metadata":{"colab":{"base_uri":"https://localhost:8080/"},"id":"lqvuzj3vZIb7","executionInfo":{"status":"ok","timestamp":1608802484291,"user_tz":-330,"elapsed":26585,"user":{"displayName":"Mallika Subramanian","photoUrl":"https://lh3.googleusercontent.com/a-/AOh14Gj3lusT2xteZqIvBVDRAXN5RCnCIyEjAhNrS9hRhv0=s64","userId":"05149503641230543668"}},"outputId":"2ef3db8a-ef67-49f1-a5c4-87ec16e01e5b"},"source":["#checking the shapes on one batch, and to ensure a sanity check\n","# before we move further ahead\n","\n","input_ids = data['input_ids'].to(device)\n","attention_mask = data['attention_mask'].to(device)\n","\n","print(input_ids.shape)\n","print(attention_mask.shape)"],"execution_count":45,"outputs":[{"output_type":"stream","text":["torch.Size([16, 160])\n","torch.Size([16, 160])\n"],"name":"stdout"}]},{"cell_type":"code","metadata":{"colab":{"base_uri":"https://localhost:8080/"},"id":"ysZLP4QVZVwQ","executionInfo":{"status":"ok","timestamp":1608802484293,"user_tz":-330,"elapsed":26511,"user":{"displayName":"Mallika Subramanian","photoUrl":"https://lh3.googleusercontent.com/a-/AOh14Gj3lusT2xteZqIvBVDRAXN5RCnCIyEjAhNrS9hRhv0=s64","userId":"05149503641230543668"}},"outputId":"08976422-935d-4045-ce0d-abccf47306db"},"source":["model(input_ids, attention_mask)"],"execution_count":46,"outputs":[{"output_type":"execute_result","data":{"text/plain":["tensor([[0.1602, 0.3840, 0.4558],\n","        [0.2467, 0.4210, 0.3323],\n","        [0.2323, 0.4162, 0.3515],\n","        [0.2433, 0.4224, 0.3344],\n","        [0.2179, 0.4080, 0.3741],\n","        [0.2323, 0.3397, 0.4280],\n","        [0.2145, 0.2355, 0.5499],\n","        [0.1596, 0.3974, 0.4431],\n","        [0.2659, 0.4495, 0.2845],\n","        [0.1907, 0.3905, 0.4189],\n","        [0.1964, 0.4000, 0.4036],\n","        [0.1375, 0.4762, 0.3863],\n","        [0.3216, 0.3366, 0.3418],\n","        [0.1708, 0.3874, 0.4418],\n","        [0.1513, 0.2914, 0.5573],\n","        [0.2821, 0.4829, 0.2350]], device='cuda:0', grad_fn=<SoftmaxBackward>)"]},"metadata":{"tags":[]},"execution_count":46}]},{"cell_type":"markdown","metadata":{"id":"0M8DHJi_ZqVo"},"source":["## Model Training\n","\n","* We use an AdamW optimizer (with weight decay fix)\n","* Linear scheduler to decay the learning rate\n","* Cross entropy loss since we're using it for classification"]},{"cell_type":"code","metadata":{"id":"SIb4fUIYZih-","executionInfo":{"status":"ok","timestamp":1608802484295,"user_tz":-330,"elapsed":26387,"user":{"displayName":"Mallika Subramanian","photoUrl":"https://lh3.googleusercontent.com/a-/AOh14Gj3lusT2xteZqIvBVDRAXN5RCnCIyEjAhNrS9hRhv0=s64","userId":"05149503641230543668"}}},"source":["optimizer = AdamW(model.parameters(), lr=2e-5, correct_bias=False)\n","\n","total_steps = len(train_data_loader)*EPOCHS #batches times epochs\n","\n","scheduler = get_linear_schedule_with_warmup(\n","    optimizer,\n","    num_warmup_steps = 0,\n","    num_training_steps = total_steps\n",")\n","\n","loss_fn = nn.CrossEntropyLoss().to(device)"],"execution_count":47,"outputs":[]},{"cell_type":"code","metadata":{"id":"oo976y4VaVSR","executionInfo":{"status":"ok","timestamp":1608802484296,"user_tz":-330,"elapsed":26341,"user":{"displayName":"Mallika Subramanian","photoUrl":"https://lh3.googleusercontent.com/a-/AOh14Gj3lusT2xteZqIvBVDRAXN5RCnCIyEjAhNrS9hRhv0=s64","userId":"05149503641230543668"}}},"source":["#helper function for training\n","\n","def train_epoch(\n","    model, \n","    data_loader,\n","    loss_fn,\n","    optimizer,\n","    device,\n","    scheduler,\n","    n_examples\n","):\n","  model = model.train()\n","  losses = []\n","  correct_predictions = 0\n","\n","  for d in data_loader :\n","    input_ids = d['input_ids'].to(device)\n","    attention_mask = d['attention_mask'].to(device)\n","    targets = d['targets'].to(device)\n","\n","    outputs = model(\n","        input_ids = input_ids,\n","        attention_mask = attention_mask\n","    )\n","\n","    _, pred = torch.max(outputs, dim=1)\n","    loss = loss_fn(outputs, targets)\n","\n","    correct_predictions += torch.sum(pred == targets)\n","    losses.append(loss.item())\n","\n","    #backpropagation steps\n","    loss.backward()\n","\n","    #gradient clipping on our model : duplicating steps that BERT original describes\n","    #way to combat the exploding gradients and preventing the instability of the training/model\n","    nn.utils.clip_grad_norm_(model.parameters(), max_norm=1.0)\n","\n","    #step of the optimizer\n","    optimizer.step()\n","    scheduler.step()\n","    optimizer.zero_grad()\n","\n","  #retain the number of correct predictions and the mean loss\n","  return correct_predictions.double() / n_examples, np.mean(losses)"],"execution_count":48,"outputs":[]},{"cell_type":"code","metadata":{"id":"51D_qIpnbrJt","executionInfo":{"status":"ok","timestamp":1608802484297,"user_tz":-330,"elapsed":26277,"user":{"displayName":"Mallika Subramanian","photoUrl":"https://lh3.googleusercontent.com/a-/AOh14Gj3lusT2xteZqIvBVDRAXN5RCnCIyEjAhNrS9hRhv0=s64","userId":"05149503641230543668"}}},"source":["'''\n","helper function to evaluate the current model\n","batch normalization and drop out arent enabled in the eval mode\n","similar to validation evaluation\n","'''\n","\n","def eval_model(model, data_loader, loss_fn, device, n_examples):\n","  model = model.eval()\n","\n","  losses = []\n","  correct_predictions = 0  \n","\n","  #disabling gradient functions so torch is faster\n","\n","  with torch.no_grad():\n","    for d in data_loader:\n","      input_ids = d['input_ids'].to(device)\n","      attention_mask = d['attention_mask'].to(device)\n","      targets = d['targets'].to(device)\n","\n","      outputs = model(\n","          input_ids = input_ids,\n","          attention_mask = attention_mask\n","      )\n","\n","      _, pred = torch.max(outputs, dim=1)\n","      loss = loss_fn(outputs, targets)\n","\n","      correct_predictions += torch.sum(pred == targets)\n","      losses.append(loss.item())\n","\n","      #no training so nothing else here. \n","  return correct_predictions.double() / n_examples, np.mean(losses)"],"execution_count":49,"outputs":[]},{"cell_type":"code","metadata":{"colab":{"base_uri":"https://localhost:8080/"},"id":"pB3_6ZAXcit3","executionInfo":{"status":"ok","timestamp":1608823638170,"user_tz":-330,"elapsed":8498655,"user":{"displayName":"Mallika Subramanian","photoUrl":"https://lh3.googleusercontent.com/a-/AOh14Gj3lusT2xteZqIvBVDRAXN5RCnCIyEjAhNrS9hRhv0=s64","userId":"05149503641230543668"}},"outputId":"e8f8a86c-9914-4099-dc13-9ad6de787b68"},"source":["#creating the training coup with the help of both of the above functions\n","\n","history = defaultdict(list)\n","best_accuracy = 0\n","\n","for epoch in range(EPOCHS):\n","\n","  print(f'Epoch {epoch+1}/{EPOCHS}')\n","  print('-'*10)\n","\n","  train_acc, train_loss = train_epoch(\n","      model,\n","      train_data_loader,\n","      loss_fn,\n","      optimizer,\n","      device, \n","      scheduler,\n","      len(df_train)\n","  )\n","\n","  print(f'Train Loss {train_loss} accuracy {train_acc}')\n","\n","  val_acc, val_loss = eval_model(\n","      model,\n","      val_data_loader,\n","      loss_fn,\n","      device,\n","      len(df_val)\n","  )\n","  print(f'Val  Loss {val_loss} accuracy {val_acc}')\n","  print()\n","\n","  history['train_acc'].append(train_acc)\n","  history['train_loss'].append(train_loss)\n","\n","  history['val_acc'].append(val_acc)\n","  history['val_loss'].append(val_loss)\n","\n","  if val_acc > best_accuracy:\n","    torch.save(model.state_dict(), root_path+'model.bin')\n","    best_accuracy = val_acc"],"execution_count":50,"outputs":[{"output_type":"stream","text":["Epoch 1/50\n","----------\n","Train Loss 0.9030266600458519 accuracy 0.6288897948090774\n","Val  Loss 0.8272772866946 accuracy 0.7127139364303179\n","\n","Epoch 2/50\n","----------\n","Train Loss 0.8061144787332286 accuracy 0.739842369887213\n","Val  Loss 0.8141079694032669 accuracy 0.7249388753056235\n","\n","Epoch 3/50\n","----------\n","Train Loss 0.7546095130236252 accuracy 0.7946731892920234\n","Val  Loss 0.8163407697127416 accuracy 0.726161369193154\n","\n","Epoch 4/50\n","----------\n","Train Loss 0.7362146293339522 accuracy 0.8132898491642886\n","Val  Loss 0.791134407887092 accuracy 0.7506112469437652\n","\n","Epoch 5/50\n","----------\n","Train Loss 0.7255059262980585 accuracy 0.8244326674819948\n","Val  Loss 0.7880381425985923 accuracy 0.7579462102689487\n","\n","Epoch 6/50\n","----------\n","Train Loss 0.7102927478111308 accuracy 0.8399239027041717\n","Val  Loss 0.797409787774086 accuracy 0.7493887530562348\n","\n","Epoch 7/50\n","----------\n","Train Loss 0.7058959507423899 accuracy 0.8440005435521131\n","Val  Loss 0.7707364650873038 accuracy 0.7750611246943765\n","\n","Epoch 8/50\n","----------\n","Train Loss 0.6956767779977425 accuracy 0.8547356977850251\n","Val  Loss 0.7737381492669766 accuracy 0.7799511002444988\n","\n","Epoch 9/50\n","----------\n","Train Loss 0.6866221665688182 accuracy 0.8641119717352901\n","Val  Loss 0.7764324431235974 accuracy 0.7713936430317848\n","\n","Epoch 10/50\n","----------\n","Train Loss 0.6843006141159845 accuracy 0.8663541242016578\n","Val  Loss 0.771144878405791 accuracy 0.773838630806846\n","\n","Epoch 11/50\n","----------\n","Train Loss 0.6791455323281496 accuracy 0.8717896453322462\n","Val  Loss 0.7712985919072077 accuracy 0.7750611246943765\n","\n","Epoch 12/50\n","----------\n","Train Loss 0.6777678804553073 accuracy 0.8731485256148933\n","Val  Loss 0.7673462617855805 accuracy 0.7799511002444988\n","\n","Epoch 13/50\n","----------\n","Train Loss 0.6723650683527408 accuracy 0.8791275988585405\n","Val  Loss 0.7738659656964816 accuracy 0.7726161369193154\n","\n","Epoch 14/50\n","----------\n","Train Loss 0.6707839937313743 accuracy 0.8800108710422612\n","Val  Loss 0.7546107448064364 accuracy 0.793398533007335\n","\n","Epoch 15/50\n","----------\n","Train Loss 0.6645104980339175 accuracy 0.886669384427232\n","Val  Loss 0.7576567221146363 accuracy 0.7897310513447433\n","\n","Epoch 16/50\n","----------\n","Train Loss 0.66553035022124 accuracy 0.8855822802011143\n","Val  Loss 0.7530606400508147 accuracy 0.793398533007335\n","\n","Epoch 17/50\n","----------\n","Train Loss 0.6619477104881536 accuracy 0.8893192009783938\n","Val  Loss 0.7525002371806365 accuracy 0.7946210268948655\n","\n","Epoch 18/50\n","----------\n","Train Loss 0.6612586679018062 accuracy 0.8898627530914527\n","Val  Loss 0.7522180000176797 accuracy 0.7946210268948655\n","\n","Epoch 19/50\n","----------\n","Train Loss 0.6602158320986706 accuracy 0.8910857453458351\n","Val  Loss 0.7407248203571026 accuracy 0.8068459657701711\n","\n","Epoch 20/50\n","----------\n","Train Loss 0.658322388905546 accuracy 0.8929202337274086\n","Val  Loss 0.7526423633098602 accuracy 0.7946210268948655\n","\n","Epoch 21/50\n","----------\n","Train Loss 0.6554771188160647 accuracy 0.8957059383068352\n","Val  Loss 0.7465881109237671 accuracy 0.8007334963325183\n","\n","Epoch 22/50\n","----------\n","Train Loss 0.6529524312071179 accuracy 0.8980160347873353\n","Val  Loss 0.7411855676999459 accuracy 0.8080684596577017\n","\n","Epoch 23/50\n","----------\n","Train Loss 0.6510127609190733 accuracy 0.900258187253703\n","Val  Loss 0.7420834807249216 accuracy 0.8056234718826406\n","\n","Epoch 24/50\n","----------\n","Train Loss 0.6480520412973736 accuracy 0.9032477238755265\n","Val  Loss 0.7458866123969738 accuracy 0.8019559902200489\n","\n","Epoch 25/50\n","----------\n","Train Loss 0.6475089467090109 accuracy 0.9037233319744531\n","Val  Loss 0.7431766723210995 accuracy 0.80440097799511\n","\n","Epoch 26/50\n","----------\n","Train Loss 0.6487344123747038 accuracy 0.9025003397200707\n","Val  Loss 0.739314554975583 accuracy 0.8080684596577017\n","\n","Epoch 27/50\n","----------\n","Train Loss 0.6460664364306823 accuracy 0.9052860442994972\n","Val  Loss 0.7253263764656507 accuracy 0.8227383863080685\n","\n","Epoch 28/50\n","----------\n","Train Loss 0.6434920560406602 accuracy 0.9077999728223943\n","Val  Loss 0.7273971014297925 accuracy 0.8202933985330073\n","\n","Epoch 29/50\n","----------\n","Train Loss 0.6431840999618821 accuracy 0.9080038048647914\n","Val  Loss 0.7327983505450762 accuracy 0.8166259168704156\n","\n","Epoch 30/50\n","----------\n","Train Loss 0.641747984290123 accuracy 0.9095665171898356\n","Val  Loss 0.7276307734159323 accuracy 0.8215158924205379\n","\n","Epoch 31/50\n","----------\n","Train Loss 0.6406881490479345 accuracy 0.9107215654300856\n","Val  Loss 0.7335351132429563 accuracy 0.812958435207824\n","\n","Epoch 32/50\n","----------\n","Train Loss 0.6390334516763687 accuracy 0.9124201657833945\n","Val  Loss 0.7355042077027835 accuracy 0.8117359413202934\n","\n","Epoch 33/50\n","----------\n","Train Loss 0.6447356087358102 accuracy 0.906576980568012\n","Val  Loss 0.7254113268393737 accuracy 0.8227383863080685\n","\n","Epoch 34/50\n","----------\n","Train Loss 0.6404314459018086 accuracy 0.910993341486615\n","Val  Loss 0.7429808676242828 accuracy 0.80440097799511\n","\n","Epoch 35/50\n","----------\n","Train Loss 0.638653676211834 accuracy 0.9127598858540562\n","Val  Loss 0.7394803579037006 accuracy 0.8092909535452323\n","\n","Epoch 36/50\n","----------\n","Train Loss 0.6377116886817891 accuracy 0.9136431580377768\n","Val  Loss 0.734404811492333 accuracy 0.812958435207824\n","\n","Epoch 37/50\n","----------\n","Train Loss 0.6379471544338309 accuracy 0.9134393259953798\n","Val  Loss 0.7345105547171372 accuracy 0.812958435207824\n","\n","Epoch 38/50\n","----------\n","Train Loss 0.6374572365828183 accuracy 0.9139828781084386\n","Val  Loss 0.7388880791572424 accuracy 0.8092909535452323\n","\n","Epoch 39/50\n","----------\n","Train Loss 0.6369914293937061 accuracy 0.9143225981791004\n","Val  Loss 0.7369897927229221 accuracy 0.8117359413202934\n","\n","Epoch 40/50\n","----------\n","Train Loss 0.6360785631381947 accuracy 0.9152738143769534\n","Val  Loss 0.7341412351681635 accuracy 0.8141809290953546\n","\n","Epoch 41/50\n","----------\n","Train Loss 0.636150844719099 accuracy 0.9152058703628211\n","Val  Loss 0.7337126663097968 accuracy 0.8141809290953546\n","\n","Epoch 42/50\n","----------\n","Train Loss 0.6355434557665949 accuracy 0.9159532545182769\n","Val  Loss 0.7318990998543226 accuracy 0.8166259168704156\n","\n","Epoch 43/50\n","----------\n","Train Loss 0.6348438216940216 accuracy 0.9164968066313358\n","Val  Loss 0.7241076735349802 accuracy 0.823960880195599\n","\n","Epoch 44/50\n","----------\n","Train Loss 0.6342207560072775 accuracy 0.9171762467726593\n","Val  Loss 0.7290943975632007 accuracy 0.8190709046454767\n","\n","Epoch 45/50\n","----------\n","Train Loss 0.633172287500423 accuracy 0.9183312950129093\n","Val  Loss 0.7330619830351609 accuracy 0.815403422982885\n","\n","Epoch 46/50\n","----------\n","Train Loss 0.6328390929362048 accuracy 0.9186030710694387\n","Val  Loss 0.7321431201237899 accuracy 0.815403422982885\n","\n","Epoch 47/50\n","----------\n","Train Loss 0.6326239429090335 accuracy 0.9188069031118359\n","Val  Loss 0.731297553731845 accuracy 0.8178484107579462\n","\n","Epoch 48/50\n","----------\n","Train Loss 0.6324577312754548 accuracy 0.9189427911401006\n","Val  Loss 0.7230592037622745 accuracy 0.8251833740831296\n","\n","Epoch 49/50\n","----------\n","Train Loss 0.6323980504403943 accuracy 0.9190107351542329\n","Val  Loss 0.7221571997954295 accuracy 0.8264058679706602\n","\n","Epoch 50/50\n","----------\n","Train Loss 0.6324191204231718 accuracy 0.9190107351542329\n","Val  Loss 0.7218134987812775 accuracy 0.8276283618581907\n","\n"],"name":"stdout"}]},{"cell_type":"code","metadata":{"colab":{"base_uri":"https://localhost:8080/","height":523},"id":"jkBZT3xPGFRq","executionInfo":{"status":"ok","timestamp":1608823639278,"user_tz":-330,"elapsed":1124,"user":{"displayName":"Mallika Subramanian","photoUrl":"https://lh3.googleusercontent.com/a-/AOh14Gj3lusT2xteZqIvBVDRAXN5RCnCIyEjAhNrS9hRhv0=s64","userId":"05149503641230543668"}},"outputId":"1feef440-8b86-4a8c-9da1-8516c10c4cf2"},"source":["plt.plot(history['train_acc'], label='train accuracy')\n","plt.plot(history['val_acc'], label='validation accuracy')\n","plt.title('Training history')\n","\n","plt.ylabel('Accuracy')\n","plt.xlabel('Epoch')\n","\n","plt.legend()\n","plt.ylim([0, 1]);"],"execution_count":51,"outputs":[{"output_type":"display_data","data":{"image/png":"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\n","text/plain":["<Figure size 864x576 with 1 Axes>"]},"metadata":{"tags":[],"image/png":{"width":732,"height":506}}}]},{"cell_type":"code","metadata":{"id":"_EuMUTr_gLRB","executionInfo":{"status":"ok","timestamp":1608823640796,"user_tz":-330,"elapsed":1527,"user":{"displayName":"Mallika Subramanian","photoUrl":"https://lh3.googleusercontent.com/a-/AOh14Gj3lusT2xteZqIvBVDRAXN5RCnCIyEjAhNrS9hRhv0=s64","userId":"05149503641230543668"}}},"source":["MODEL_PATH = root_path+'model_base_cased_state_842.bin'\n","torch.save(model.state_dict(), MODEL_PATH)"],"execution_count":52,"outputs":[]},{"cell_type":"code","metadata":{"id":"AHgyFNkF4cPS","executionInfo":{"status":"ok","timestamp":1608823645014,"user_tz":-330,"elapsed":4223,"user":{"displayName":"Mallika Subramanian","photoUrl":"https://lh3.googleusercontent.com/a-/AOh14Gj3lusT2xteZqIvBVDRAXN5RCnCIyEjAhNrS9hRhv0=s64","userId":"05149503641230543668"}}},"source":["#running the pretrained model\n","\n","# !gdown --id 1V8itWtowCYnb2Bc9KlK9SxGff9WwmogA\n","# model = SentimentClassifier(len(class_names))\n","# model.load_state_dict(torch.load('best_model_state.bin'))\n","# model = model.to(device)\n","\n","model = SentimentClassifier(len(class_names))\n","model.load_state_dict(torch.load(MODEL_PATH))\n","model = model.to(device)"],"execution_count":53,"outputs":[]},{"cell_type":"markdown","metadata":{"id":"7BUl3vcs_dWM"},"source":["# Evaluation of the Classifier\n","\n","Function to return the text reviews, predictions of our models, prediction probabilities and the real sentiment values"]},{"cell_type":"code","metadata":{"id":"s5eHPGe75a4A","executionInfo":{"status":"ok","timestamp":1608823645019,"user_tz":-330,"elapsed":26,"user":{"displayName":"Mallika Subramanian","photoUrl":"https://lh3.googleusercontent.com/a-/AOh14Gj3lusT2xteZqIvBVDRAXN5RCnCIyEjAhNrS9hRhv0=s64","userId":"05149503641230543668"}}},"source":["def get_predictions(model, data_loader):\n","  model = model.eval() #eval mode\n","\n","  review_texts = []\n","  predictions = []\n","  prediction_probs = []\n","  real_values = []\n","\n","  with torch.no_grad():\n","    for d in data_loader:\n","      \n","      texts = d['review_text']\n","      input_ids = d['input_ids'].to(device)\n","      attention_mask = d['attention_mask'].to(device)\n","      targets = d['targets'].to(device)\n","\n","      outputs = model(\n","        input_ids = input_ids,\n","        attention_mask = attention_mask\n","      )\n","\n","      _, pred = torch.max(outputs, dim=1)\n","\n","      review_texts.extend(texts)\n","      predictions.extend(pred)\n","      prediction_probs.extend(outputs)\n","      real_values.extend(targets)\n","\n","    #converting list to tensors\n","    predictions = torch.stack(predictions).cpu() #converting list of tensors to single tensor\n","    prediction_probs = torch.stack(prediction_probs).cpu()\n","    real_values = torch.stack(real_values).cpu()\n","\n","    return review_texts, predictions, prediction_probs, real_values\n"],"execution_count":54,"outputs":[]},{"cell_type":"code","metadata":{"id":"8HKD5tcrBXqq","executionInfo":{"status":"ok","timestamp":1608823653327,"user_tz":-330,"elapsed":8319,"user":{"displayName":"Mallika Subramanian","photoUrl":"https://lh3.googleusercontent.com/a-/AOh14Gj3lusT2xteZqIvBVDRAXN5RCnCIyEjAhNrS9hRhv0=s64","userId":"05149503641230543668"}}},"source":["test_acc, test_loss = eval_model(model, test_data_loader, loss_fn, device, len(df_test))"],"execution_count":55,"outputs":[]},{"cell_type":"code","metadata":{"colab":{"base_uri":"https://localhost:8080/"},"id":"7PaJGXYfBh_d","executionInfo":{"status":"ok","timestamp":1608823653337,"user_tz":-330,"elapsed":30,"user":{"displayName":"Mallika Subramanian","photoUrl":"https://lh3.googleusercontent.com/a-/AOh14Gj3lusT2xteZqIvBVDRAXN5RCnCIyEjAhNrS9hRhv0=s64","userId":"05149503641230543668"}},"outputId":"9c712fbf-8f22-4bbd-c335-b323be692d7c"},"source":["test_acc"],"execution_count":56,"outputs":[{"output_type":"execute_result","data":{"text/plain":["tensor(0.8142, device='cuda:0', dtype=torch.float64)"]},"metadata":{"tags":[]},"execution_count":56}]},{"cell_type":"code","metadata":{"id":"bevcpmwDAnnm","executionInfo":{"status":"ok","timestamp":1608823661990,"user_tz":-330,"elapsed":8669,"user":{"displayName":"Mallika Subramanian","photoUrl":"https://lh3.googleusercontent.com/a-/AOh14Gj3lusT2xteZqIvBVDRAXN5RCnCIyEjAhNrS9hRhv0=s64","userId":"05149503641230543668"}}},"source":["y_review_texts, y_pred, y_pred_probs, y_test = get_predictions(model, test_data_loader)"],"execution_count":57,"outputs":[]},{"cell_type":"code","metadata":{"colab":{"base_uri":"https://localhost:8080/"},"id":"yiaRlUHbBKIg","executionInfo":{"status":"ok","timestamp":1608823661997,"user_tz":-330,"elapsed":33,"user":{"displayName":"Mallika Subramanian","photoUrl":"https://lh3.googleusercontent.com/a-/AOh14Gj3lusT2xteZqIvBVDRAXN5RCnCIyEjAhNrS9hRhv0=s64","userId":"05149503641230543668"}},"outputId":"08b78da1-a831-4fed-e66d-b0fc34880f08"},"source":["#evaluation using classification from scikit learn\n","print(classification_report(y_test, y_pred, target_names=class_names))"],"execution_count":58,"outputs":[{"output_type":"stream","text":["              precision    recall  f1-score   support\n","\n","    negative       0.83      0.80      0.81       257\n","     neutral       0.75      0.75      0.75       253\n","    positive       0.85      0.88      0.86       308\n","\n","    accuracy                           0.81       818\n","   macro avg       0.81      0.81      0.81       818\n","weighted avg       0.81      0.81      0.81       818\n","\n"],"name":"stdout"}]},{"cell_type":"code","metadata":{"id":"tkeZqJhMB2Pv","executionInfo":{"status":"ok","timestamp":1608823661998,"user_tz":-330,"elapsed":26,"user":{"displayName":"Mallika Subramanian","photoUrl":"https://lh3.googleusercontent.com/a-/AOh14Gj3lusT2xteZqIvBVDRAXN5RCnCIyEjAhNrS9hRhv0=s64","userId":"05149503641230543668"}}},"source":["#getting confusion matrix for the corpus\n","def show_confusion_matrix(confusion_matrix):\n","\n","  hmap = sns.heatmap(confusion_matrix, annot=True, fmt=\"d\", cmap=\"Blues\")\n","  hmap.yaxis.set_ticklabels(hmap.yaxis.get_ticklabels(), rotation=0, ha='right')\n","  hmap.xaxis.set_ticklabels(hmap.xaxis.get_ticklabels(), rotation=30, ha='right')\n","  plt.ylabel('True sentiment')\n","  plt.xlabel('Predicted sentiment');"],"execution_count":59,"outputs":[]},{"cell_type":"code","metadata":{"colab":{"base_uri":"https://localhost:8080/","height":533},"id":"8uXbpmaFChlF","executionInfo":{"status":"ok","timestamp":1608823663091,"user_tz":-330,"elapsed":1108,"user":{"displayName":"Mallika Subramanian","photoUrl":"https://lh3.googleusercontent.com/a-/AOh14Gj3lusT2xteZqIvBVDRAXN5RCnCIyEjAhNrS9hRhv0=s64","userId":"05149503641230543668"}},"outputId":"8e30b100-e22a-40d3-e05e-94242b28431d"},"source":["cm = confusion_matrix(y_test, y_pred)\n","df_cm = pd.DataFrame(cm, index=class_names, columns=class_names)\n","show_confusion_matrix(df_cm)"],"execution_count":60,"outputs":[{"output_type":"display_data","data":{"image/png":"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\n","text/plain":["<Figure size 864x576 with 2 Axes>"]},"metadata":{"tags":[],"image/png":{"width":725,"height":516}}}]},{"cell_type":"code","metadata":{"id":"hytrmqahDUUk","executionInfo":{"status":"ok","timestamp":1608823663093,"user_tz":-330,"elapsed":42,"user":{"displayName":"Mallika Subramanian","photoUrl":"https://lh3.googleusercontent.com/a-/AOh14Gj3lusT2xteZqIvBVDRAXN5RCnCIyEjAhNrS9hRhv0=s64","userId":"05149503641230543668"}}},"source":["idx = 2\n","review_text = y_review_texts[idx]\n","true_sentiment = y_test[idx]\n","\n","pred_df = pd.DataFrame({\n","    'class_names':class_names,\n","    'values' : y_pred_probs[idx]\n","})"],"execution_count":61,"outputs":[]},{"cell_type":"code","metadata":{"colab":{"base_uri":"https://localhost:8080/"},"id":"s8Yk1APBDy0M","executionInfo":{"status":"ok","timestamp":1608823663094,"user_tz":-330,"elapsed":30,"user":{"displayName":"Mallika Subramanian","photoUrl":"https://lh3.googleusercontent.com/a-/AOh14Gj3lusT2xteZqIvBVDRAXN5RCnCIyEjAhNrS9hRhv0=s64","userId":"05149503641230543668"}},"outputId":"477ba643-f154-449e-9dfe-cce1c62f1b5d"},"source":["print(\"\\n\".join(wrap(review_text)))\n","print()\n","print(f'True Sentiment : {class_names[true_sentiment]}')"],"execution_count":62,"outputs":[{"output_type":"stream","text":["It has helped me so much with school, and the google calender\n","integration makes it even better. Thank you!\n","\n","True Sentiment : positive\n"],"name":"stdout"}]},{"cell_type":"code","metadata":{"colab":{"base_uri":"https://localhost:8080/","height":523},"id":"WG9pVLiED5hp","executionInfo":{"status":"ok","timestamp":1608823663937,"user_tz":-330,"elapsed":865,"user":{"displayName":"Mallika Subramanian","photoUrl":"https://lh3.googleusercontent.com/a-/AOh14Gj3lusT2xteZqIvBVDRAXN5RCnCIyEjAhNrS9hRhv0=s64","userId":"05149503641230543668"}},"outputId":"8b59611b-2690-4d1c-b1ee-21f71f01e4b1"},"source":["#checking with what the model thinks about the review\n","\n","sns.barplot(x='values', y='class_names', data=pred_df, orient='h')\n","plt.ylabel('sentiment')\n","plt.xlabel('probability')\n","plt.xlim([0,1])"],"execution_count":63,"outputs":[{"output_type":"execute_result","data":{"text/plain":["(0.0, 1.0)"]},"metadata":{"tags":[]},"execution_count":63},{"output_type":"display_data","data":{"image/png":"iVBORw0KGgoAAAANSUhEUgAABhcAAAPTCAYAAACzDljcAAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAAWJQAAFiUBSVIk8AAAADh0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uMy4yLjIsIGh0dHA6Ly9tYXRwbG90bGliLm9yZy+WH4yJAAAgAElEQVR4nOzde5DV9X3/8ddhYbmLoIBCMBiNW4EheJcYqaWO8Y50xtg2YuqFaNFUx1oRnYwpnQox1iRjajETNUqpMUS03qoNoqANCFHqjYATKwiiiCIroHLb/f3hb7cSLvLRPS7ax2OGcfd8v5/veZ8d5/yxzz3fT6WxsbExAAAAAAAAO6lNaw8AAAAAAAB8togLAAAAAABAEXEBAAAAAAAoIi4AAAAAAABFxAUAAAAAAKCIuAAAAAAAABQRFwAAAAAAgCLiAgAAAAAAUERcAAAAAAAAiogLAAAAAABAEXEBAAAAAAAoIi4AAAAAAABF2rb2AFBNCxYsyPr161NTU5P27du39jgAAAAAAB/L+vXrs3nz5rRv3z4DBgxo7XHEBT7f1q9fn4aGhjQ0NGTjxo2tPQ4AAAAAwCeyfv361h4hibjA51xNTU0aGhrSpk2bdOrUqbXHAXZha9euTZJ06dKllScBdnXeL4Cd4b0C2FneL4Cd1fR+UVNT08qTfEBc4HOtffv22bhxYzp16pS6urrWHgfYhT311FNJ4r0C+EjeL4Cd4b0C2FneL4Cd1fR+savc/t2GzgAAAAAAQBFxAQAAAAAAKCIuAAAAAAAARcQFAAAAAACgiLgAAAAAAAAUERcAAAAAAIAi4gIAAAAAAFBEXAAAAAAAAIqICwAAAAAAQBFxAQAAAAAAKCIuAAAAAAAARcQFAAAAAACgiLgAAAAAAAAUERcAAAAAAIAi4gIAAAAAAFBEXAAAAAAAAIqICwAAAAAAQBFxAQAAAAAAKCIuAAAAAAAARcQFAAAAAACgiLgAAAAAAAAUERcAAAAAAIAi4gIAAAAAAFBEXAAAAAAAAIqICwAAAAAAQBFxAQAAAAAAKCIuAAAAAAAARcQFAAAAAACgiLgAAAAAAAAUERcAAAAAAIAi4gIAAAAAAFBEXAAAAAAAAIqICwAAAAAAQBFxAQAAAAAAKCIuAAAAAAAARcQFAAAAAACgiLgAAAAAAAAUERcAAAAAAIAi4gIAAAAAAFBEXAAAAAAAAIqICwAAAAAAQBFxAQAAAAAAKCIuAAAAAAAARcQFAAAAAACgiLgAAAAAAAAUERcAAAAAAIAi4gIAAAAAAFBEXAAAAAAAAIqICwAAAAAAQBFxAQAAAAAAKCIuAAAAAAAARcQFAAAAAACgiLgAAAAAAAAUERcAAAAAAIAi4gIAAAAAAFBEXAAAAAAAAIqICwAAAAAAQBFxAQAAAAAAKCIuAAAAAAAARcQFAAAAAACgiLgAAAAAAAAUERcAAAAAAIAi4gIAAAAAAFBEXAAAAAAAAIqICwAAAAAAQBFxAQAAAAAAKCIuAAAAAAAARcQFAAAAAACgiLgAAAAAAAAUERcAAAAAAIAi4gIAAAAAAFBEXAAAAAAAAIqICwAAAAAAQBFxAQAAAAAAKCIuAAAAAAAARcQFAAAAAACgiLgAAAAAAAAUERcAAAAAAIAi4gIAAAAAAFBEXAAAAAAAAIqICwAAAAAAQBFxAQAAAAAAKCIuAAAAAAAARcQFAAAAAACgiLgAAAAAAAAUERcAAAAAAIAi4gIAAAAAAFBEXAAAAAAAAIqICwAAAAAAQBFxAQAAAAAAKCIuAAAAAAAARcQFAAAAAACgiLgAAAAAAAAUERcAAAAAAIAi4gIAAAAAAFBEXAAAAAAAAIqICwAAAAAAQBFxAQAAAAAAKCIuAAAAAAAARcQFAAAAAACgiLgAAAAAAAAUERcAAAAAAIAi4gIAAAAAAFBEXAAAAAAAAIqICwAAAAAAQBFxAQAAAAAAKCIuAAAAAAAARcQFAAAAAACgiLgAAAAAAAAUERcAAAAAAIAi4gIAAAAAAFBEXAAAAAAAAIqICwAAAAAAQBFxAQAAAAAAKCIuAAAAAAAARcQFAAAAAACgiLgAAAAAAAAUERcAAAAAAIAi4gIAAAAAAFBEXAAAAAAAAIqICwAAAAAAQBFxAQAAAAAAKCIuAAAAAAAARcQFAAAAAACgiLgAAAAAAAAUERcAAAAAAIAi4gIAAAAAAFBEXAAAAAAAAIqICwAAAAAAQBFxAQAAAAAAKCIuAAAAAAAARcQFAAAAAACgiLjwf9yTTz6Zurq61NXVZdmyZa09DgAAAAAAnwHiwufUFVdckbq6uowaNaq1RwEAAAAA4HNGXAAAAAAAAIq0be0BaF1HHHFEFi1a1NpjAAAAAADwGeKTCwAAAAAAQJHP5ScXrrjiitx99905/PDDM3ny5CxcuDA/+9nPMnfu3KxatSrdu3fPUUcdlTFjxmSfffbZ7nXq6+szZcqUPProo3nllVeybt269OjRI4ceemhGjRqVgw46aIdzLFy4MDfddFPmzZuX+vr69OzZM8OGDcvo0aPTt2/f1NXVJUkmTJiQP/uzP9ti7fr16zN79uzMmDEj8+fPz7Jly7Jx48Z069YtAwYMyKmnnpqTTjopbdps2YemTZuWcePGNX8/d+7c5udpMnLkyEycODHJBxs6n3XWWUmSRx55JF/4wheSJFOmTMn48ePTpk2bPPbYY+ndu/d2X+e8efNy5plnJkluueWWHHXUUVudM3v27PzqV7/K008/nTfffDO1tbXp379/vv71r+fMM89Mp06ddvizBAAAAABg1/G5jAsf9uCDD2bs2LHZsGFD82NvvPFG7r777syYMSOTJ0/e6pfvSTJnzpxcfPHFWb169RaPr1ixIg888EAeeOCBjBkzJhdffPE2n/fee+/NuHHjsmnTpubHXn311dxxxx35j//4j9x88807nPuf/umfctttt231+JtvvplZs2Zl1qxZue+++/KTn/wktbW1O7zWx3HiiSdmwoQJ2bhxY+67776cd9552z33vvvuS5L07NkzQ4cO3eLY+vXrc+WVV+b+++/f4vENGzbk+eefz/PPP59f/vKX+dnPfpb+/fu3+OsAAAAAAKDlfa5vi7RkyZKMHTs2X/nKV3LLLbdk9uzZmTlzZq688srU1tamvr4+V1999VbrXnjhhYwePTqrV6/OgAED8sMf/jCPPvpo5s6dm7vuuqv5UwY33nhjpk6dutX6hQsXNoeF3r1759prr83jjz+exx9/PNdee21qa2tzySWX7HD2rl275hvf+EZ+9KMf5a677sqsWbPyxBNPZOrUqTnnnHPSoUOHzJw5Mz/60Y+2WHfqqafm6aefzimnnJIkOeSQQ/L0009v8W/8+PEf+bPr3r17jj766CQfhJLt2bBhQx566KEkycknn7zVJyn+7u/+Lvfff3/atWuXc845J3fddVeefPLJzJw5M9///vez9957Z+nSpbngggvy7rvvfuRcAAAAAAC0vs/1JxdWrFiRo48+OpMmTUrbtv/7Ur/1rW+loaEhEydOzPz58/PSSy9lv/32az4+bty4bNiwIUOGDMnkyZO3+GRAt27dMmHChPTs2TM33XRTrr/++pxyyinp0KFD8zk/+MEPsmnTpnTp0iVTpkxJv379mo+NGDEiQ4YMyWmnnbbD2b/zne9s8/GePXtm8ODBGTp0aEaPHp077rgjY8aMSZcuXZIkbdu2bf6XJDU1NencuXPBT+1/jRgxIjNmzMiiRYvy4osv5oADDtjqnFmzZqW+vr75/A/7z//8zzz88MOpVCr58Y9/nD/90z/d4vhpp52WI488MiNHjszLL7+cO+64I+eee+7HmhUAAAAAgE/P5/qTC0ly1VVXbREWmowcObL56+eee6756zlz5mTRokVJkmuuuWa7txwaM2ZMOnXqlFWrVuWJJ55ofvyNN97If/3XfyVJRo0atUVYaPLFL34xo0aN+ngv6P8bNmxYevTokXfffTfz58//RNfanuHDh6dr165Jtv/phabHv/zlL+fAAw/c4tjtt9+eJDnhhBO2CgtN9tprr3zzm99M8r+3VwIAAAAAYNf2uY4L/fr1y7777rvNY7vvvnt69OiR5IN9DJrMnj07SdKnT5/stddeWbdu3Tb/bd68ufnazz//fPP6Z555Jo2NjUk++OX89mzvl+0ftmrVqvzLv/xL/vIv/zJHHnlkBg4cmLq6uuZ/q1atSpIsXrz4I6/1cdTW1ub4449Pktx///3Nr6vJmjVr8uijjyb54HZMH/bee+/lv//7v5MkRxxxxHZ/juvWrWv+RMSiRYu22BsDAAAAAIBd0+f6tki9evXa4fGOHTsmSd5///3mx15++eUkyfLly3PwwQfv1PM0/ZI/+WDT5iZf+tKXtrtmR8eS5Le//W0uvPDCrTaU3pY1a9bsxJQfz6mnnpqpU6fmtddey9y5c3PEEUc0H3vooYeyYcOGVCqV5j0emixdujQbN25Mklx99dXb3NviDzU0NKS+vj49e/Zs2RcBAAAAAECL+lzHhZqamp0678N/kf9xflH/4b+2//CmxE3xYls6deq03WNr1qzJRRddlNWrV2ePPfbI2WefncMPPzx77713OnXqlEqlkiQ56aST8tprr2Xz5s3FM++sww47LH379s2rr76ae++9d4u40HQbo8MOOyx77733Vq/h41i/fv3HHxYAAAAAgE/F5zoufBxNv/QfPHhwpk6d+rHXJx/cGqhpo+U/9OEI8YceeuihvP3222nTpk1uv/327L///ts8b+3atcXzlapUKjn55JNz00035eGHH87VV1+d2travP7665k3b16SrW+JlGSLTaR/+tOf5o//+I+rPisAAAAAAJ+Oz/WeCx9H0wbMS5cu3WqPgZ3Rp0+f5q+bbrG0LTs61rShdF1d3XbDwmuvvVbV2yF92IgRI5JsucfC/fffn4aGhrRv3755X4YP69u3b9q0+eB/r6VLl34qcwIAAAAA8OkQF/7AUUcdlSR5++23M2fOnOL1Q4YMab5t0YwZM7Z73iOPPLLdY023WdrR7Y6abkm0PW3btv3Ia+ys/fbbLwMHDkyS3HvvvVv895hjjknXrl23WtO1a9cMHjw4SfLggw9+4hkAAAAAANh1iAt/4Gtf+1oOOOCAJMn3vve9vPnmmzs8f9myZVvsudCrV6989atfTZJMnjw5y5Yt22rN0qVLM3ny5O1e8wtf+EKSDz7dsGTJkq2Ov/TSS5k0adIO59p9992TJG+88cYOz9tZTbc+mjlzZubNm9f86YqmTzVsy9lnn50keeqpp3Lrrbfu8PqbN2/e5msFAAAAAGDXIy78gUqlkokTJ6ZDhw5ZvHhxRowYkZtvvjkvvvhi6uvr89Zbb+V3v/tdpk6dmgsuuCDHHXfcVnsfXHbZZampqcmaNWty5pln5r777svKlSuzcuXK3HvvvTnzzDPTo0eP7c5w3HHHpU2bNtm4cWO+/e1v55FHHsnKlSuzfPny/Nu//Vu++c1vpmPHjs0BYVuaPmmwdOnSTJkyJW+99VY2bdqUTZs2paGhofjncvLJJ6empiYbN27M2LFjk3wQMIYNG7bdNccff3xOOumkJMnEiRNz4YUXZubMmVmxYkXeeeedvPrqq5k1a1Z+8IMf5Nhjj81tt91WPBcAAAAAAJ8+Gzpvw8CBA3PrrbfmkksuyYoVK3Lttdfm2muv3ea5NTU1qamp2eKxAQMG5JprrsmVV16Z1157LZdddtkWx7t165Ybbrghp59+evM1Pqx///655JJLcv3112fx4sUZM2bMFse7du2aG264IWPHjs3q1au3Odef/MmfpF+/flm6dGnGjx+f8ePHNx8bOXJkJk6cuHM/jP9vzz33zFe/+tU8/vjjefXVV5MkJ5xwQtq1a7fDdRMnTkyXLl1y5513Zvr06Zk+ffp2z/2oawEAAAAAsGsQF7bj4IMPzsMPP5y77rorM2bMyKJFi1JfX5+amprsueee+fKXv5yhQ4fm+OOPT7du3bZaf9ppp+WAAw7ITTfdlHnz5uWdd95Jz54987WvfS3f/va307179+ZzO3fuvNX6888/P/vtt19uu+22vPDCC9m0aVN69+6do446Kueee27zxtPb06FDh0yZMiU33nhjZs+enddffz3r16//RD+TESNG5PHHH2/+vulWSTtSW1ub8ePH54wzzsidd96Z3/72t82zdOnSJf369cuQIUNyzDHHNN9OCgAAAACAXVulsbGxsbWH+L9owYIFGTlyZJLkrrvuyqBBg1p5os+nRYsWZe3atenSpUvq6upaexxgF/bUU08lSQ455JBWngTY1Xm/AHaG9wpgZ3m/AHZW0/vFrvK7TnsutJIZM2Yk+eAv+5s2kAYAAAAAgM8CcaFKtrcXQpIsXrw4t956a5Jk+PDhqa2t/bTGAgAAAACAT8yeC1Vy+eWXp3PnzjnppJMycODAdO7cOStXrszjjz+eSZMmZe3atWnXrt1WmzUDAAAAAMCuTlyoks2bN+fBBx/Mgw8+uM3jtbW1+f73v79L3BsLAAAAAABKiAtV8p3vfCcHHHBA5s2blxUrVuTtt99ObW1t+vTpk6FDh+ass85Kv379WntMAAAAAAAoJi5UyZAhQzJkyJDWHgMAAAAAAFqcDZ0BAAAAAIAi4gIAAAAAAFBEXAAAAAAAAIqICwAAAAAAQBFxAQAAAAAAKCIuAAAAAAAARcQFAAAAAACgiLgAAAAAAAAUERcAAAAAAIAi4gIAAAAAAFBEXAAAAAAAAIqICwAAAAAAQBFxAQAAAAAAKCIuAAAAAAAARcQFAAAAAACgiLgAAAAAAAAUERcAAAAAAIAi4gIAAAAAAFBEXAAAAAAAAIqICwAAAAAAQBFxAQAAAAAAKCIuAAAAAAAARcQFAAAAAACgiLgAAAAAAAAUERcAAAAAAIAi4gIAAAAAAFBEXAAAAAAAAIqICwAAAAAAQBFxAQAAAAAAKCIuAAAAAAAARcQFAAAAAACgiLgAAAAAAAAUERcAAAAAAIAi4gIAAAAAAFBEXAAAAAAAAIqICwAAAAAAQBFxAQAAAAAAKCIuAAAAAAAARcQFAAAAAACgiLgAAAAAAAAUERcAAAAAAIAi4gIAAAAAAFBEXAAAAAAAAIqICwAAAAAAQBFxAQAAAAAAKCIuAAAAAAAARcQFAAAAAACgiLgAAAAAAAAUERcAAAAAAIAi4gIAAAAAAFBEXAAAAAAAAIqICwAAAAAAQBFxAQAAAAAAKCIuAAAAAAAARcQFAAAAAACgiLgAAAAAAAAUERcAAAAAAIAi4gIAAAAAAFBEXAAAAAAAAIqICwAAAAAAQBFxAQAAAAAAKCIuAAAAAAAARcQFAAAAAACgiLgAAAAAAAAUERcAAAAAAIAi4gIAAAAAAFBEXAAAAAAAAIqICwAAAAAAQBFxAQAAAAAAKCIuAAAAAAAARcQFAAAAAACgiLgAAAAAAAAUERcAAAAAAIAi4gIAAAAAAFBEXAAAAAAAAIqICwAAAAAAQBFxAQAAAAAAKCIuAAAAAAAARcQFAAAAAACgiLgAAAAAAAAUERcAAAAAAIAi4gIAAAAAAFBEXAAAAAAAAIqICwAAAAAAQBFxAQAAAAAAKCIuAAAAAAAARcQFAAAAAACgiLgAAAAAAAAUERcAAAAAAIAi4gIAAAAAAFBEXAAAAAAAAIqICwAAAAAAQBFxAQAAAAAAKCIuAAAAAAAARcQFAAAAAACgiLgAAAAAAAAUERcAAAAAAIAi4gIAAAAAAFBEXAAAAAAAAIqICwAAAAAAQBFxAQAAAAAAKCIuAAAAAAAARcQFAAAAAACgiLgAAAAAAAAUERcAAAAAAIAi4gIAAAAAAFBEXAAAAAAAAIqICwAAAAAAQBFxAQAAAAAAKCIuAAAAAAAARcQFAAAAAACgiLgAAAAAAAAUERcAAAAAAIAi4gIAAAAAAFBEXAAAAAAAAIqICwAAAAAAQBFxAQAAAAAAKCIuAAAAAAAARcQFAAAAAACgiLgAAAAAAAAUERcAAAAAAIAi4gIAAAAAAFBEXAAAAAAAAIqICwAAAAAAQBFxAQAAAAAAKCIuAAAAAAAARcQFAAAAAACgiLgAAAAAAAAUERcAAAAAAIAi4gIAAAAAAFBEXAAAAAAAAIqICwAAAAAAQBFxAQAAAAAAKCIuAAAAAAAARcQFAAAAAACgiLgAAAAAAAAUERcAAAAAAIAi4gIAAAAAAFBEXAAAAAAAAIqICwAAAAAAQBFxAQAAAAAAKCIuAAAAAAAARcQFAAAAAACgiLgAAAAAAAAUERcAAAAAAIAi4gIAAAAAAFBEXAAAAAAAAIqICwAAAAAAQBFxAQAAAAAAKCIuAAAAAAAARcQFAAAAAACgiLgAAAAAAAAUERcAAAAAAIAi4gIAAAAAAFBEXAAAAAAAAIqICwAAAAAAQBFxAQAAAAAAKCIuAAAAAAAARcQFAAAAAACgiLgAAAAAAAAUERcAAAAAAIAi4gIAAAAAAFBEXAAAAAAAAIqICwAAAAAAQBFxAQAAAAAAKCIuAAAAAAAARcQFAAAAAACgiLgAAAAAAAAUERcAAAAAAIAi4gIAAAAAAFBEXAAAAAAAAIqICwAAAAAAQBFxAQAAAAAAKCIuAAAAAAAARcQFAAAAAACgiLgAAAAAAAAUERcAAAAAAIAi4gIAAAAAAFBEXAAAAAAAAIqICwAAAAAAQBFxAQAAAAAAKCIuAAAAAAAARcQFAAAAAACgiLgAAAAAAAAUERcAAAAAAIAi4gIAAAAAAFBEXAAAAAAAAIqICwAAAAAAQBFxAQAAAAAAKCIuAAAAAAAARcQFAAAAAACgiLgAAAAAAAAUERcAAAAAAIAi4gIAAAAAAFBEXAAAAAAAAIqICwAAAAAAQBFxAQAAAAAAKCIuAAAAAAAARcQFAAAAAACgiLgAAAAAAAAUERcAAAAAAIAi4gIAAAAAAFBEXAAAAAAAAIq0rcZFly9fniTp3bt3ampqdmrN5s2bs2LFiiRJnz59qjEWAAAAAADQAqoSF4YPH542bdrk3nvvzf77779Ta5YsWZITTzwxbdq0yYIFC6oxFgAAAAAA0AKqdlukxsbGT3UdAAAAAADw6djl9lyoVCqtPQIAAAAAALADu0xcqK+vT5J06NChlScBAAAAAAB2ZJeJC/fcc0+SpG/fvq08CQAAAAAAsCMtsqHzWWedtc3Hx40bl44dO+5w7YYNG/LKK6/k7bffTqVSyZFHHtkSIwEAAAAAAFXSInFh7ty5W+2V0NjYmOeff36n1jdt4rznnnvmvPPOa4mRAAAAAACAKmmRuNCnT58tvl++fHkqlUp69uyZtm23/xSVSiUdOnRIr169csghh+Qv/uIvsscee7TESAAAAAAAQJW0SFyYMWPGFt//0R/9UZLklltuyf77798STwEAAAAAAOwiWiQu/KHDDjssST5yvwUAAAAAAOCzpypxYfLkydW4LAAAAAAAsAto09oDAAAAAAAAny3iAgAAAAAAUKQqt0VqsmTJkvziF7/I3Llzs2zZsqxduzYNDQ07XFOpVLJgwYJqjgUAAAAAAHwCVYsLd955Z/7xH/8xGzduTJI0NjZW66kAAAAAAIBPUVXiwrx58/K9730vyQdRYc8998ygQYPSrVu3tGnjTkwAAAAAAPBZVpW4cMstt6SxsTG1tbUZP358RowYkUqlUo2nAgAAAAAAPmVViQvPPPNMKpVKzjvvvJx22mnVeAoAAAAAAKCVVOUeRWvWrEmSDBs2rBqXBwAAAAAAWlFV4kLPnj2TJDU1NdW4PAAAAAAA0IqqEhcOO+ywJMmLL75YjcsDAAAAAACtqCpx4a/+6q9SU1OT2267LRs3bqzGUwAAAAAAAK2kKnHhwAMPzHe/+938/ve/z4UXXphVq1ZV42kAAAAAAIBW0LYaF/3JT36SJBk8eHBmzZqV4cOHZ+jQofnSl76Ujh07fuT6iy66qBpjAQAAAAAALaBqcaFSqSRJKpVK3n///Tz22GN57LHHdmq9uAAAAAAAALuuqsSFJGlsbNzh99vTFCUAAAAAAIBdU1XiwiOPPFKNywIAAAAAALuAqsSFvn37VuOyAAAAAADALqBNaw8AAAAAAAB8togLAAAAAABAkapt6Nzk3XffzcMPP5ynn346K1euzHvvvZdrrrlmi1snrVq1KuvWrUttbW169+5d7ZEAAAAAAIBPoKpx4Z577smECRPyzjvvJEkaGxtTqVTy3nvvbXHejBkz8t3vfjft27fPE088kS5dulRzLAAAAAAA4BOo2m2RpkyZknHjxqW+vj6NjY3p1q3bds8dMWJEdtttt6xfvz7Tp0+v1kgAAAAAAEALqEpceOWVVzJhwoQkycEHH5z7778/c+bM2e757dq1y7HHHpvGxsb85je/qcZIAAAAAABAC6lKXLj99tuzadOm7LPPPrnllluy//77f+Sagw46KEmyaNGiaowEAAAAAAC0kKrEhSeffDKVSiVnnXVWOnTosFNr+vfvnyR57bXXqjESAAAAAADQQqoSF5YvX54kGTRo0E6vadrE+d13363GSAAAAAAAQAupSlzYtGlTkqRSqez0mnXr1iVJOnbsWI2RAAAAAACAFlKVuLDHHnskSZYtW7bTaxYuXJgk6dWrVzVGAgAAAAAAWkhV4kLT7ZBmz56902umTZuWSqXSvLEzAAAAAACwa6pKXDj++OPT2NiYe+65J0uWLPnI8ydNmpQXXnghSXLSSSdVYyQAAAAAAKCFVCUunHDCCRkwYEA2btyYs88+O7NmzdrieNNeDAsWLMjf/u3f5sc//nEqlUoOP/zwDB06tBojAQAAAAAALaRtNS5aqVRyww035Iwzzsjy5ctz/vnnp2vXrs3Hzz333KxZsybvvvtukqSxsTF9+vTJddddV41xAAAAAACAFlSVTy4kSd++fTNt2rQcddRRaWxszDvvvNN87PXXX8+6devS2NiYxsbGHHnkkbnzzjvTs2fPao0DAAAAAAC0kKp8cqFJr169cvPNN+fZZ5/Nr3/96zz77LNZtWpVNm3alB49emTgwIE57u933NkAACAASURBVLjjcuihh1ZzDAAAAAAAoAVVNS40GTx4cAYPHvxpPBUAAAAAAFBlVbstEgAAAAAA8PkkLgAAAAAAAEXEBQAAAAAAoEhV91z4n//5n/zyl7/M3Llz8+qrr2bt2rVpaGjY4ZpKpZIFCxZUcywAAAAAAOATqFpc+PnPf57rrrsumzdvTpI0NjZW66kAAAAAAIBPUVXiwowZMzJx4sTm7/v06ZO6urrstttuadPGnZgAAAAAAOCzrCpx4dZbb02SdOnSJddee22GDx9ejacBAAAAAABaQVU+RvC73/0ulUolF110kbAAAAAAAACfM1WJC037Kxx22GHVuDwAAAAAANCKqhIX+vbtmyR5//33q3F5AAAAAACgFVUlLnz9619Pkjz55JPVuDwAAAAAANCKqhIXzjzzzOy99975+c9/nsWLF1fjKQAAAAAAgFZSlbjQrVu3TJo0KZ07d84ZZ5yRO+64I/X19dV4KgAAAAAA4FPWtloXPuCAA/Kv//qvOf300zN+/Pj8wz/8Q7p3754OHTrscF2lUsn06dOrNRa7mGnTpmXcuHFJkkWLFrXyNAAAAAAA7IyqxYXp06fnyiuvzJo1a5IkDQ0Neeuttz5yXaVSqdZI/+eNGjUqc+fOzciRIzNx4sTWHgcAAAAAgM+oqsSFZ599Npdcckk2b96cxsbGtG/fPl/84hez2267iQcAAAAAAPAZV5W4MGnSpGzatCnt2rXL5ZdfntNPP/0jb4cEAAAAAAB8NlQlLjz33HOpVCo5//zzM2rUqGo8BQAAAAAA0EraVOOiTfssDBs2rBqXbxVXXHFF6urqmmPJwoULc9lll2XYsGEZNGhQjj766FxxxRV55ZVXdnid+vr63HjjjTn99NNzxBFHZNCgQRk2bFguvfTSzJ8/f7vrRo0albq6ulxxxRU7vH5dXV3q6uoybdq05sduuOGG1NXVZe7cuUmSu+++u/m8pn833HDDVucPHz48SfL73/8+V111VYYPH55Bgwbl0EMPbT63sbExzzzzTH74wx/mG9/4Rg4//PAMHDgwhx9+eP78z/88P/3pT7N27dodzgwAAAAAwGdLVT650Lt374/8Jftn2YMPPpixY8dmw4YNzY+98cYbufvuuzNjxoxMnjw5dXV1W62bM2dOLr744qxevXqLx1esWJEHHnggDzzwQMaMGZOLL7646q9hZ02fPj2XXnpp1q9f3/zYh29x9cgjj+TCCy/cal19fX3mz5+f+fPn51e/+lVuvvnm9OvX71OZGQAAAACA6qpKXDj66KMzZcqUzJ8/P4MHD67GU7SaJUuWZOzYsfnKV76Sv/7rv86BBx6YDRs25OGHH851112X+vr6XH311fnFL36xxboXXngho0ePzoYNGzJgwICMHj06Q4YMSefOnbN06dJMmTIl06ZNy4033pg+ffrk9NNPb7GZzz///JxzzjkZPXp0nnrqqZxyyin5+7//+y3Oadeu3Vbr6uvrc/nll2efffbJ3/zN3+Sggw5KQ0NDnnvuueZz2rZtm+HDh2f48OHZb7/90qtXr3Tu3DlvvPFGZs+enVtvvTVLlizJpZdemqlTp7bYawIAAAAAoPVUJS6cc845+fd///fcfPPNOeWUU9KjR49qPE2rWLFiRY4++uhMmjQpbdv+74/vW9/6VhoaGjJx4sTMnz8/L730Uvbbb7/m4+PGjcuGDRsyZMiQTJ48ObW1tc3HunXrlgkTJqRnz5656aabcv311+eUU05psU2wa2trU1tbm5qamiQfBIHOnTt/5Lq1a9emf//+ueOOO9K1a9fmx3v37t389THHHJNjjjlmq7Xdu3dPXV1dTjzxxJx88sl59tlnM3v27AwdOvSTvyAAAAAAAFpVVfZc6NOnT/75n/8569evzxlnnJGZM2dW42lazVVXXbVFWGgycuTI5q8//Nf9c+bMyaJFi5Ik11xzzRZh4cPGjBmTTp06ZdWqVXniiSdaeOqP5+KLL94iLJTq1atXc1D4zW9+01JjAQAAAADQiqryyYWzzjorSbL77rtnyZIlueCCC9K1a9f079//I/8av1Kp5LbbbqvGWC2iX79+2Xfffbd5bPfdd0+PHj2yatWqvPnmm82Pz549O8kH0WWvvfbKunXrtnv9fffdNy+88EKef/75HHvssS07fKFKpbJTm3Jv3Lgx99xzT379619n4cKFWb169RZ7NDRZvHhxFaYEAAAAAODTVpW4MHfu3FQqlSQf/IK6sbEx77zzzhZ/zb8tjY2Nzet2Vb169drh8Y4dOyZJ3n///ebHXn755STJ8uXLc/DBB+/U86xatepjTthyunfvni5duuzwnJUrV+acc87Jiy+++JHXW7NmTUuNBgAAAPw/9u492uqywP/4Z3O4CSxRRCQuaoKSkFh5SbysNBnHUcPLrCaDyNWgDZmOOpmXxiYbNc0kZ2o0wzIVTTKF8TI2thQ00rxEgpMKosjFy3ARBY4geA7794eLMyCXzoNnc9Df67XWWR6+3/1897N1+V2u/fb5PgDQimoSF3r16lWLy24T1u5b8JdUq9Wm37fkS/XVq1cXj2lpa0PJ5px33nl5/vnn065du4wYMSKHH354dt1113Tp0qXp0VH/8i//knvvvTeNjY21njIAAAAAAFtBTeLCpEmTanHZD6xOnTolSQYPHpxf//rXNXufhoaGml17Y+bNm9e0j8JFF12Uk08+eaOvW7ly5dacFgAAAAAANVaTDZ1ZX9++fZMk8+fPX29FQ4kOHTokWf9xS++1cOHCLbr2lpoxY0bT78cee+wmX9ecRyYBAAAAAPDBIS5sBYccckiS5I033shjjz22RdfYeeedk/zf/g0bM2XKlM1eY+1jilrq8UTrPrppU9ecNm1a5s+f3yLvBwAAAADAtkFc2AoOPfTQ7LXXXkmSiy++OIsXL97s619++eUN9lzYd999k7y7WmDdFQNrLV68ONdcc81mr7vDDjskabkVDn369Gn6ffLkyRucf+utt/Ld7363Rd4LAAAAAIBth7iwFVQqlVxxxRXp2LFj5syZk+OPPz4///nP8/zzz2fp0qV5/fXX89xzz+XXv/51Ro8enaOOOir19fXrXePoo49O586dkySnn356HnzwwbzxxhtZsGBB7rrrrvzd3/1d06OTNmXQoEFJkqlTp+Y3v/lN3nzzzTQ0NKShoSFr1qwp/lz77LNPU2C49NJLc+utt2b+/Pl5/fXX8+CDD+bkk0/OjBkz8tGPfrT42gAAAAAAbLve14bOX/7yl5O8++X5TTfdtMHxLfHea31YDBo0KL/4xS9y9tlnZ8GCBbnyyitz5ZVXbvS1dXV1qaurW+/YDjvskIsvvjjnn39+XnnllZx++unrnd9ll10yduzYze59cPzxx2fs2LFZunRpzj777PXOnXHGGTnzzDOLPlNdXV0uu+yyfPWrX019fX3+9V//db3zbdq0yfnnn58ZM2Zs9nFOAAAAAAB8sLyvuPDEE0+kUqk0+/hfUq1Wt2jcB8WnPvWp3H///bnzzjszadKkzJw5M0uXLk1dXV26d++ePffcM0OGDMnRRx+drl27bjB+2LBh+chHPpKxY8fm6aefzooVK9KzZ88MHTo0p512Wrp167bZ9995550zfvz4XHfddXnyySezaNGivPPOO+/rMx100EG5/fbbc+211+aJJ55IfX19dtxxx3zyk5/MyJEjc8ABB+SCCy54X+8BAAAAAMC2pVKtVqtbOvizn/1s0++TJk3a6PEtse614P2YOXNm6uvr06VLlwwYMKC1pwNsw6ZOnZok2W+//Vp5JsC2zv0CaA73CqC53C+A5lp7v9hWvut8XysXNhUBxAEAAAAAAPjwsqEzAAAAAABQRFwAAAAAAACKvK/HIm3KhRdemEqlkrPPPjs9evRo1phFixblhz/8YSqVSr73ve/VYloAAAAAAEALqMnKhYkTJ2bixIlZtmxZs8csX768aRwAAAAAALDt8lgkAAAAAACgyDYTFxoaGpIkbdvW5ElNAAAAAABAC9lm4sILL7yQJOnatWsrzwQAAAAAANicFlkm8OSTT270+P/8z//kjTfe2OzY1atXZ86cOfnZz36WSqWSj33sYy0xJQAAAAAAoEZaJC6MHDkylUplvWPVajXf+ta3mn2NarWaSqWSk046qSWmBAAAAAAA1EiLbXBQrVabdWxTtttuu4waNSrHHHNMS00JAAAAAACogRaJC5dffvl6f77wwgtTqVRy1llnZZdddtnkuEqlkg4dOqRHjx4ZOHBgtttuu5aYDgAAAAAAUEMtEhdOPPHE9f584YUXJkmGDh2a/v37t8RbAAAAAAAA24gWeyzSum6++eYkSZ8+fWpxeQAAAAAAoBXVJC4ceOCBtbgsAAAAAACwDWjT2hMAAAAAAAA+WGqycmFdb775ZqZNm5b58+envr4+jY2Nf3HMGWecUetpAQAAAAAAW6hmcWHp0qW54oorcu+996ahoaForLgAAAAAAADbrprEhbfeeitf+tKX8sILL6RarRaNrVQqtZgSAAAAAADQQmoSF2644YbMmjUrSdK/f/+MGDEi++yzT7p27Zo2bWzzAAAAAAAAH2Q1iQu//e1vU6lUMnjw4Nx8883p0KFDLd4GAAAAAABoBTVZRvDyyy8nSU499VRhAQAAAAAAPmRqEhfatWuXJOnbt28tLg8AAAAAALSimsSF3XbbLUmyZMmSWlweAAAAAABoRTWJC5/73OdSrVYzadKkWlweAAAAAABoRTWJC8OHD8+gQYPyq1/9Ko899lgt3gIAAAAAAGglNYkLbdu2zfXXX5999tknp556ar7//e/n2Wefzdtvv12LtwMAAAAAALaitrW46N577930e7VazY033pgbb7yxWWMrlUqeffbZWkwLAAAAAABoATWJC9VqdbN/BgAAAAAAPrhqEhdOPPHEWlwWAAAAAADYBtQkLlx++eW1uCwAAAAAALANqMmGzgAAAAAAwIeXuAAAAAAAABSpyWORNmbBggVZtGhR3n777Xz84x9Px44dt9ZbAwAAAAAALaimceHtt9/OjTfemNtvvz2vvfZa0/F77rkn/fv3b/rzfffdl4ceeijbb799LrroolpOCQAAAAAAeJ9qFhcWLFiQ0047LbNmzUq1Wm06XqlUNnjtgAED8k//9E+pVCoZNmxYBg8eXKtpAQAAAAAA71NN9lxobGzM6aefnueffz5JctRRR+Xb3/72Jl/fr1+/7LvvvkmSyZMn12JKAAAAAABAC6lJXLjnnnvyzDPPpK6uLj/+8Y/zox/9KCNGjNjsmMMPPzzVajVPPfVULaYEAAAAAAC0kJrEhfvuuy+VSiXHH398hg4d2qwxe++9d5Jkzpw5tZgSAAAAAADQQmoSF5599tkkyV//9V83e8xOO+2UJHnzzTdrMSUAAAAAAKCF1CQurA0EPXr0aP5E2rw7lTVr1tRiSgAAAAAAQAupSVzo3LlzkuT1119v9pj//d//TZJ07dq1FlMCAAAAAABaSE3iQp8+fZIkL730UrPHPPLII0mS/v3712JKAAAAAABAC6lJXBgyZEiq1WrGjx/frNfPnz8/EydOTKVSySGHHFKLKQEAAAAAAC2kJnFh+PDhadeuXWbPnp2rr756s6998cUXc9ppp2XlypXZbrvt8vnPf74WUwIAAAAAAFpI21pctFevXjn77LPzgx/8IGPHjs3jjz+ev/mbv2k6P2nSpPzud7/Lk08+mSlTpqSxsTGVSiUXXnihPRcAAAAAAGAbV5O4kCSjRo3KihUrcu2112batGmZPn16KpVKkqy3mqFaraZSqeSss86yagEAAAAAAD4AavJYpLXOPPPM3HLLLTnssMNSV1eXarW63k+lUsmnP/3pjBs3LqNHj67lVAAAAAAAgBZSs5ULa+233365/vrrs2LFijz77LN5/fXX09jYmB133DEDBw70GCQAAAAAAPiAqXlcWKtTp07Zf//9t9bbAQAAAAAANbLV4sJ7zZkzJ/fff3+WLFmSvn375thjj82OO+7YWtMBAAAAAACaqSZxYcaMGbnyyitTqVTywx/+cINHH913330577zz0tjY2HTs2muvzXXXXZfBgwfXYkoAAAAAAEALqcmGzg8++GAeffTRrF69eoOwsGjRovzzP/9zGhoa1tvcecmSJTnjjDOyYsWKWkwJAAAAAABoITWJC4899lgqlUoOO+ywDc6NHz8+K1euTF1dXb75zW/mrrvuyje+8Y1UKpUsWrQod9xxRy2mBAAAAAAAtJCaxIUFCxYkST72sY9tcO63v/1tKpVKjjnmmIwaNSoDBgzIaaedlhNOOCHVajWTJ0+uxZQAAAAAAIAWUpO48OabbyZJunfvvsHxF154IUly3HHHrXfuyCOPTJLMmjWrFlMCAAAAAABaSE3iwtp9ExoaGtY7/uc//znVajV1dXU54IAD1ju3yy67JEmWLl1aiykBAAAAAAAtpCZxoXPnzkne3bx5XU8++WSSZK+99kqnTp02OrZ9+/a1mBIAAAAAANBCahIXPvrRjyZJHn744fWO33///alUKtl///03GLN48eIkyU477VSLKQEAAAAAAC2kbS0ueuihh2batGmZMGFC9tprrxx44IG58847M2fOnFQqlQwdOnSDMc8991ySpEePHrWYEgAAAAAA0EJqEhdGjBiRcePGZdmyZbnsssvWOzd48OAceOCBG4z53e9+l0qlkoEDB9ZiSgAAAAAAQAupyWORdtxxx/z0pz9Njx49Uq1Wm3769euXMWPGbPD62bNnZ/r06UmSgw46qBZTAgAAAAAAWkhNVi4kySc+8Yk88MADmTp1ahYvXpyePXvmU5/6VNq02bBnLFmyJF//+teTJAcffHCtpgQAAAAAALSAmsWFJGnXrl2zViLsv//+G93kGQAAAAAA2PbU5LFIAAAAAADAh5e4AAAAAAAAFBEXAAAAAACAIuICAAAAAABQRFwAAAAAAACKiAsAAAAAAEARcQEAAAAAACgiLgAAAAAAAEXEBQAAAAAAoIi4AAAAAAAAFBEXAAAAAACAIuICAAAAAABQRFwAAAAAAACKiAsAAAAAAEARcQEAAAAAACgiLgAAAAAAAEXEBQAAAAAAoIi4AAAAAAAAFBEXAAAAAACAIuICAAAAAABQRFwAAAAAAACKiAsAAAAAAEARcQEAAAAAACgiLgAAAAAAAEXEBQAAAAAAoIi4AAAAAAAAFBEXAAAAAACAIuICAAAAAABQRFwAAAAAAACKiAsAAAAAAEARcQEAAAAAACgiLgAAAAAAAEXEBQAAAAAAoIi4AAAAAAAAFBEXAAAAAACAIuICAAAAAABQRFwAAAAAAACKiAsAAAAAAEARcQEAAAAAACgiLgAAAAAAAEXEBQAAAAAAoIi4AAAAAAAAFBEXAAAAAACAIuICAAAAAABQRFwAAAAAAACKiAsAAAAAAEARcQEAAAAAACgiLgAAAAAAAEXEBQAAAAAAoIi4AAAAAAAAFBEXAAAAAACAIuICAAAAAABQRFwAAAAAAACKiAsAAAAAAEARcQEAAAAAACgiLgAAAAAAAEXEBQAAAAAAoIi4AAAAAAAAFBEXAAAAAACAIuICAAAAAABQRFwAAAAAAACKiAsAAAAAAEARcQEAAAAAACgiLgAAAAAAAEXEBQAAAAAAoIi4AAAAAAAAFBEXAAAAAACAIuICAAAAAABQRFwAAAAAAACKiAsAAAAAAEARcQEAAAAAACgiLgAAAAAAAEXEBQAAAAAAoIi4AAAAAAAAFBEXAAAAAACAIuICAAAAAABQRFwAAAAAAACKiAsAAAAAAEARcQEAAAAAACgiLgAAAAAAAEXEBQAAAAAAoIi4AAAAAAAAFBEXAAAAAACAIuICAAAAAABQRFwAAAAAAACKiAsAAAAAAEARcQEAAAAAACgiLgAAAAAAAEXEBQAAAAAAoIi4AAAAAAAAFBEXAAAAAACAIuICAAAAAABQRFwAAAAAAACKiAsAAAAAAEARcQEAAAAAACgiLgAAAAAAAEXEBQAAAAAAoIi4AAAAAAAAFKlUq9Vqa08CamXmzJmpr69Ply5dMmDAgNaeDgAAAADAFtnWvuu0cgEAAAAAACjStrUnAFvD23Mez5xfHt3a0wAAAAAA2DLD/7u1Z7AeKxcAAAAAAIAi4gIAAAAAAFBEXAAAAAAAAIqICwAAAAAAQBFxAQAAAAAAKCIuAAAAAAAARcQFAAAAAACgiLgAAAAAAAAUERcAAAAAAIAi4gIAAAAAAFBEXAAAAAAAAIqICwAAAAAAQBFxAQAAAAAAKCIuAAAAAAAARcQFAAAAAACgiLgAAAAAAAAUERcAAAAAAIAi4gIAAAAAAFBEXAAAAAAAAIqICwAAAAAAQBFxAQAAAAAAKCIuAAAAAAAARcQFAAAAAACgiLgAAAAAAAAUERcAAAAAAIAi4gIAAAAAAFBEXAAAAAAAAIqICwAAAAAAQBFxAQAAAAAAKCIuAAAAAAAARcQFAAAAAACgiLgAAAAAAAAUERcAAAAAAIAi4gIAAAAAAFBEXAAAAAAAAIqICwAAAAAAQBFxAQAAAAAAKCIuAAAAAAAARcQFAAAAAACgiLgAAAAAAAAUERcAAAAAAIAi4gIAAAAAAFBEXAAAAAAAAIqICwAAAAAAQBFxAQAAAAAAKCIuAAAAAAAARcQFAAAAAACgiLgAAAAAAAAUERcAAAAAAIAi4gIAAAAAAFBEXAAAAAAAAIqICwAAAAAAQBFxAQAAAAAAKCIuAAAAAAAARcQFAAAAAACgiLgAAAAAAAAUERcAAAAAAIAi4gIAAAAAAFBEXAAAAAAAAIqICwAAAAAAQBFxAQAAAAAAKCIuAAAAAAAARcQFAAAAAACgiLgAAAAAAAAUERcAAAAAAIAi4gIAAAAAAFBEXAAAAAAAAIqICwAAAAAAQBFxAQAAAAAAKCIuAAAAAAAARcQFAAAAAACgiLgAAAAAAAAUERcAAAAAAIAi4gIAAAAAAFBEXAAAAAAAAIqICwAAAAAAQBFxAQAAAAAAKCIuAAAAAAAARcQFAAAAAACgiLgAAAAAAAAUERcAAAAAAIAi4gIAAAAAAFBEXAAAAAAAAIqICwAAAAAAQBFxAQAAAAAAKCIuAAAAAAAARcQFAAAAAACgiLgAAAAAAAAUERcAAAAAAIAi4gIAAAAAAFBEXAAAAAAAAIqICwAAAAAAQBFxAQAAAAAAKCIuAAAAAAAARcQFAAAAAACgiLgAAAAAAAAUERcAAAAAAIAi4gIAAAAAAFBEXAAAAAAAAIqICwAAAAAAQBFxAQAAAAAAKCIuAAAAAAAARcQFAAAAAACgiLgAAAAAAAAUERc+wEaOHJkBAwbkggsueF/XGTBgQAYMGJAJEya00MwAAAAAAPgwExc+pCZMmNAUDQAAAAAAoCWJCwAAAAAAQJG2rT0Btty4ceNa5DozZ85skesAAAAAAPD/BysXAAAAAACAIlYuJLngggsyceLEHHjggRk3blyefPLJ/OIXv8j06dOzbNmy9OzZM0OHDs0//MM/ZIcddtjkdWbOnJmbb745jz/+eBYuXJi2bdumb9++Ofzww3PKKaekW7dumxz7pz/9Kb/85S/z1FNPZdGiRalUKunWrVt69OiRAw44IEcddVQGDx683piRI0fmiSeeyIknnpgrrrgiSfLyyy/nyCOPXO917913oXfv3pk0adIG5y+//PKcdNJJSZIXXnghxx57bJJkzJgxOe644zY595UrV+bggw/OihUrMnr06JxzzjkbvOall17KLbfckj/84Q957bXXsmbNmvTs2TOHHXZY/v7v/z69evXa5PUBAAAAANi2iAvvMX78+Hz3u9/NmjVrmo7NmzcvN9xwQ+69997cdNNN2WOPPTYY9/Of/zxXXXXVeuNWrVqVGTNmZMaMGbnttttyzTXX5IADDtjo2CuvvHKD46+++mpeffXVTJs2LbNmzcpPf/rTFvqUf1n//v0zaNCgPPPMM7n77rs3GxcefPDBrFixIkkybNiwDc7fcMMNGTNmTBoaGtY7PmfOnMyZMyd33HFHrr766hxxxBEt+yEAAAAAAKgJcWEdc+fOzaWXXppBgwblnHPOyd57753ly5fn3nvvzU9+8pMsXLgwX/va13L33XenQ4cOTePuueeepjiw11575Zxzzsm+++6bVatWZfLkyfn3f//3LF26NF/96ldz9913p2/fvk1jX3rppYwZMyZJMmTIkIwaNSr9+vVLly5dsmzZsrz44ouZMmVKli9f3qzP0Lt37/zpT3/KPffck+985ztJ3l0Vsa42bZr3NKxhw4blmWeeySOPPJIlS5ZscuXF3XffnSQZNGhQ+vXrt965W2+9Nd///veTJEcddVSGDx+ePffcM23atMmzzz6b//iP/8hTTz2Vs846K3fccUf22muvZs0NAAAAAIDWY8+FdSxYsCD9+vXLuHHjcsghh6Rbt27Zbbfd8vWvfz3f+973krz7f9vfeuutTWNWr16dyy+/PEmyxx575LbbbstnP/vZ7LTTTunVq1dGjBiRG2+8Me3bt8+KFSuavmhf6/e//30aGxuz0047ZezYsTnssMPSq1evbL/99unTp08+85nP5KKLLtpg3KZUKpV07tw57du3bzrWuXPn9X622267Zl3r2GOPTV1dXRoaGvJf//VfG33NkiVL8sgjjyTZcNXCwoULmx7X9JWvfCU//vGPM2TIkHTv3j3dunXLoYcemnHjxuWAAw7IqlWrmiILAAAAAADbNnHhPb7xjW9s9Mv3YcOGNe15MGHChKbjkyZNyuuvv54kOffcc9OlS5cNxg4cODBf+MIXml6/ZMmSpnONjY1Jkm7duq0XBLYFO++8c4YMGZLk3dUZG3PfffeloaEhdXV1Gzw6afz4zhRGsAAAIABJREFU8Vm9enV69uyZc889d6Pj27Vrl7POOitJ8vDDD2fZsmUt+AkAAAAAAKgFcWEdnTp1yiGHHLLJ83/1V3+V5N3Njtd+CT516tQkyXbbbZfPfOYzmxx79NFHJ3k3Jqz7mKK99947STJr1qxcddVVeeONN97fh2hhxx9/fJJk+vTpmTt37gbn10aHtSsS1vXoo48mSdPKhLfeemujP2sfpVStVvPMM8/U8uMAAAAAANAC7Lmwjt122y11dXWbPL92I+dqtZpXX30122+/fV599dUkye677562bTf9t3PPPfds+n3tmCT59Kc/naFDh+aBBx7I9ddfnxtuuCEf//jHs99++2X//ffPkCFD0qlTp/f70bbY0KFD06lTp6xYsSJ33313zjzzzKZz8+bNy7Rp05JsfCPnl156Kcm7AWJTKx/ea91VHQAAAAAAbJusXFjHX/oSf93zb7311np//UtjO3fuvMHYtf7t3/4t5513Xvr27ZvGxsZMnz49N9xwQ04//fQcfPDBueSSS1JfX1/0WVpKp06dmlZsvDcQrN3Ied3XrGtL5rxq1aotmCUAAAAAAFuTlQvrWLFiRbPPr40Fa/+6JWPXateuXUaNGpVRo0Zl7ty5eeqpp/LHP/4xDz30UBYtWpRbbrkl06ZNy69+9avNro6olWHDhuWuu+7K3LlzM23atHziE59I8n+xYe3qhvfq1KlTli1bllNPPTXf/OY3t+qcAQAAAACoHSsX1jF37tymDZY3Zvbs2UmSSqWSXr16JUl69+6dJJkzZ04aGho2OXbWrFlNv68dszG77bZbTjjhhFx66aV56KGHMnLkyCTJn//85zz00EPN/iwtaciQIdl5552T/F9QePrppzNnzpwkG38kUpL07ds3STJ//vzaTxIAAAAAgK1GXFjHihUr8sgjj2zy/AMPPJAk6d+/f7bffvskyX777ZckWblyZaZMmbLJsffff3+SpK6uLp/85CebNZ+2bduut8fBiy++2Kxxa8eutblg0hx1dXU57rjjkiT33XdfGhoamh6JtPPOO+fggw/e6Li1m2P//ve/b9oAGwAAAACADz5x4T3GjBmTlStXbnD8nnvuyfTp05MkJ510UtPxI444IjvttFOS5KqrrtroPgMzZszIbbfdliQ58sgj061bt6Zzc+bMyZo1azY5n3nz5jX9vsMOOzT7c6z72oULFzZ73KYcf/zxSd7dcPnhhx/Ob37zmyTJscceu8lNsEeMGJH27dvnrbfeykUXXZR33nlns++xdmUIAAAAAADbNnFhHT169MiLL76YkSNH5tFHH80bb7yRefPm5ZprrsmFF16YJNl9990zYsSIpjHt27dvOvfCCy9k+PDhmTx5cpYsWZLXXnstt912W0455ZSsXr06nTp12mDvgeuuuy5Dhw7NmDFj8sgjj+S1117LsmXLMm/evNx5551NKxc6deqUI444otmfZeDAgWnT5t1/vD/60Y/yyiuvZPXq1WloaNiilQx777139txzzyTJZZddlsWLFyfZ9CORkqRnz5751re+leTdlRuf//zn85//+Z+ZP39+li9fngULFuSPf/xjfvazn+Vv//Zv84//+I/F8wIAAAAAYOuzofM6dt9993zta1/LJZdckq985SsbnO/Ro0d+8pOfpEOHDusd/9znPpeFCxfmqquuysyZMzN69OgNxnbt2jXXXHNNdt111w3OvfLKKxk7dmzGjh270Xl17NgxP/jBD9KjR49mf5bu3bvnmGOOyb333psJEyZkwoQJTed69+6dSZMmNftaaw0bNixjxozJK6+8kiTp169fBg0atNkxX/ziF9OmTZtceumlee6553L++edv8rUDBw4snhMAAAAAAFufuPAew4cPzx577JEbb7wxTz/9dJYvX56ePXvmyCOPzOjRozf5aKJRo0blkEMOyc0335zHH388ixYtSl1dXfr27Zsjjjgip5xyynqPQ1rr3HPPzZAhQ/LYY4/lueeey6JFi/Lmm2+mQ4cO2W233TJkyJB86UtfatpAusTll1+e/v375/7778/cuXOzcuXKVKvV4uusNWzYsFx99dVNj3Ha3KqFdX3hC1/I4Ycfnl/+8pd59NFHM2/evCxfvjwdO3bMRz7ykQwcODCHHXZYhg4dusVzAwAAAABg66lU38+3zR8SF1xwQSZOnJgDDzww48aNa+3p0IJmzpyZ+vr6tF38TLo++p3Wng4AAAAAwBZZNfy/U19fny5dumTAgAGtPR17LgAAAAAAAGXEBQAAAAAAoIi4AAAAAAAAFBEXAAAAAACAIuICAAAAAABQpG1rT2BbcMUVV+SKK65o7WkAAAAAAMAHgpULAAAAAABAEXEBAAAAAAAoIi4AAAAAAABFxAUAAAAAAKCIuAAAAAAAABQRFwAAAAAAgCLiAgAAAAAAUERcAAAAAAAAiogLAAAAAABAEXEBAAAAAAAoIi4AAAAAAABFxAUAAAAAAKCIuAAAAAAAABQRFwAAAAAAgCLiAgAAAAAAUERcAAAAAAAAiogLAAAAAABAEXEBAAAAAAAoIi4AAAAAAABFxAUAAAAAAKCIuAAAAAAAABQRFwAAAAAAgCLiAgAAAAAAUERcAAAAAAAAiogLAAAAAABAEXEBAAAAAAAoIi4AAAAAAABFxAUAAAAAAKCIuAAAAAAAABQRFwAAAAAAgCLiAgAAAAAAUERcAAAAAAAAiogLAAAAAABAEXEBAAAAAAAoIi4AAAAAAABFxAUAAAAAAKCIuAAAAAAAABQRFwAAAAAAgCLiAgAAAAAAUERcAAAAAAAAiogLAAAAAABAEXEBAAAAAAAoIi4AAAAAAABFxAUAAAAAAKCIuAAAAAAAABQRFwAAAAAAgCLiAgAAAAAAUERcAAAAAAAAiogLAAAAAABAEXEBAAAAAAAoIi4AAAAAAABFxAUAAAAAAKCIuAAAAAAAABQRFwAAAAAAgCLiAgAAAAAAUERcAAAAAAAAiogLAAAAAABAEXEBAAAAAAAoIi4AAAAAAABFxAUAAAAAAKCIuAAAAAAAABQRFwAAAAAAgCLiAgAAAAAAUERcAAAAAAAAiogLAAAAAABAEXEBAAAAAAAoIi4AAAAAAABFxAUAAAAAAKCIuAAAAAAAABQRFwAAAAAAgCLiAgAAAAAAUERcAAAAAAAAiogLAAAAAABAEXEBAAAAAAAoIi4AAAAAAABFxAUAAAAAAKCIuAAAAAAAABQRFwAAAAAAgCLiAgAAAAAAUERcAAAAAAAAiogLAAAAAABAEXEBAAAAAAAoIi4AAAAAAABFxAUAAAAAAKCIuAAAAAAAABQRFwAAAAAAgCLiAgAAAAAAUERcAAAAAAAAiogLAAAAAABAEXEBAAAAAAAoIi4AAAAAAABFxAUAAAAAAKCIuAAAAAAAABQRFwAAAAAAgCLiAgAAAAAAUERcAAAAAAAAiogLAAAAAABAEXEBAAAAAAAo0ra1JwBbQ8fdP53d//ql1p4GAAAAAMAWmTlzZmtPYT1WLgAAAAAAAEXEBQBIMnXq1EydOrW1pwF8ALhfAM3hXgE0l/sF0Fz19fWtPYX1iAsAAAAAAEARcQEAAAAAACgiLgAAAAAAAEXEBQAAAAAAoIi4AAAAAAAAFBEXAAAAAACAIuICAAAAAABQRFwAAAAAAACKiAsAAAAAAEARcQEAAAAAACgiLgAAAAAAAEXEBQAAAAAAoIi4AAAAAAAAFBEXAAAAAACAIuICAAAAAABQRFwAAAAAAACKiAsAAAAAAEARcQEAAAAAACgiLgAAAAAAAEXEBQAAAAAAoIi4AAAAAAAAFBEXAAAAAACAIuICAAAAAABQRFwAAAAAAACKiAsAAAAAAEARcQEAAAAAACjStrUnALW0atWqJMmKFSsyc+bMVp4N8EHgXgE0l/sF0BzuFUBzuV8AzbX2O8/WJi7wodbY2JgkWbNmTerr61t5NsAHgXsF0FzuF0BzuFcAzeV+ATTX2u88W5u4wIdahw4dsmrVqtTV1aVDhw6tPR0AAAAAgC2yatWqNDY2bjPfc1aq1Wq1tScBAAAAAAB8cNjQGQAAAAAAKCIuAAAAAAAARcQFAAAAAACgiLgAAAAAAAAUERcAAAAAAIAi4gIAAAAAAFBEXAAAAAAAAIqICwAAAAAAQBFxAQAAAAAAKCIuAAAAAAAARcQFAAAAAACgiLgAAAAAAAAUERcAAAAAAIAi4gIAAAAAAFBEXAAAAAAAAIqICwAAAAAAQJG2rT0BaI7Jkydn/PjxeeaZZ7J06dJ07949Q4YMySmnnJIBAwa87+vPnDkzN910U/7whz9k8eLF6dq1awYNGpSTTz45RxxxRAt8AmBrqMW9olqtZurUqZkyZUqmTp2a2bNnZ9myZenQoUP69OmTgw8+OMOHD0/fvn1b+NMAtVTr/7ZYV7VazZe//OU88cQTSZLevXtn0qRJLfoeQO1sjfvFSy+9lNtvvz1TpkzJa6+9lsbGxnTv3j39+/fPQQcdlJNPPjkdO3ZskfcCaqeW94vly5fntttuy+TJkzN79uzU19enY8eO2XXXXTNkyJCMGDEivXv3bqFPAtRCtVrN7Nmz8/TTTzf9zJw5M++8806S5MEHH0yfPn1a5L221nedlWq1Wm2xq0ENfOc738n48eM3eq59+/a55JJLcsIJJ2zx9SdOnJhvf/vbTf8iv9cXv/jFXHzxxVt8fWDrqNW9YvTo0Zk8efJmX9OxY8f/1969R0Vd538cfyEXhVIBL2heVtIVFS/ryqEoLAPaMl0Ly1tlkpe1PLplmpdSf4FlbZvbmuZpV931gHiJAtxUWu+SplKWCiGwsqbCJnnBFIYEh/n94Zk5IAMywMwgPR/nePbrfD7fz+fznXP27fR5fz+fjxYuXKinnnrK5vYBOJ69f1vcLCEhQQsWLLD8neQCcPtwRLxYtWqVPvjgA5WWllZbpyEnGwDYhz3jRWZmpqZOnaoff/yx2jqenp5asmSJHnvssTr1AcD+8vLyFB4eXm15Q/1778i5TrZFQqO2atUqyz/OERERSkxM1MGDB7VmzRr17NlTpaWlev3113XkyJE6tX/kyBEtWLBAZWVl6tmzp9asWaODBw8qMTFRERERkqQNGzZo1apVDfZMABqePWNFcXGxJCkoKEiLFi3Sv/71Lx06dEi7du1SdHS0fHx89PPPP2vBggXau3dvQz4WADuw92+Lm124cEF//vOf5ebmpg4dOjRImwAcwxHx4sMPP9R7772n0tJShYeHa9WqVUpNTdWhQ4eUlJSkV199VT179myoRwJgJ/aMF0VFRZbEgru7uyZOnKjk5GQdPHhQn332mV566SV5eXmppKREc+bM0cmTJxv68QDYQYcOHfTwww8rKCioQdt19FwnKxfQaF26dEnh4eEyGAwKDQ3V6tWr5eLiYikvLCzU8OHDdeHCBQ0YMEAff/yxzX2MGjVKx48fV9u2bbVlyxb5+PhYykwmkyZNmqQDBw7Iy8tLu3btkq+vb4M8G4CGY+9YsXLlSg0ZMkR9+vSxWn7q1CmNHDlSBoNB3bt317Zt2+r1PADsxxG/LW42c+ZMbdu2TZMnT9bx48eVlpbGygXgNuCIePHNN9/o6aeflslk0uzZszVlypSGfAQADmLveFFxBeScOXM0adKkKnW2bdummTNnSpImTJig1157rR5PBMBeioqKdOjQIQ0YMEDt2rWTJC1fvlwrVqyQ1DArFxw918nKBTRaSUlJMhgMkqRXXnml0j/OkuTj46PJkydLko4dO6bvvvvOpvbT09N1/PhxSdLkyZMr/Z9NklxcXDRr1ixJksFg0ObNm+v0HADsy96xYtq0adUmFiTJ399fTz75pCQpNzdX+fn5NrUPwHHsHS9utm/fPm3btk2dOnXS9OnT69UWAMdyRLz405/+JJPJpJCQEBILwG3M3vHixIkTlusRI0ZYrfPII49YzmX573//a1P7ABznzjvvVEREhCWx0NCcMddJcgGNlnmP865duyowMNBqnaFDh1qubX0DsOIe6hXbqSgwMFBdu3atU/sAHMPesaI2evToYbmuaR9UAM7lyHhhMBgUHR0tSVqwYIE8PT3r3BYAx7N3vMjOztbRo0clSVFRUXUbJIBGwd7xonnz5pbrmxMXFT83l7Vp08am9gE0Hc6Y6yS5gEbLnM0fMGBAtXU6dOggPz+/SvVtbd/Pz6/GPZDN/df37UUA9mHvWFEbFy5csFy3bNmywdsH0DAcGS+WLVum/Px8RUREKCwsrM7tAHAOe8eLffv2SZJcXV0VEhJSqez69es2tQXAuewdLyquov7888+t1tmzZ49KSkokSQ8++KBN7QNoOpwx10lyAY1SQUGBZVlhly5daqxr3ovs1KlTNvVhrl/b9ouLi1VQUGBTHwDsyxGxojZ27NghSfL29pa/v3+Dtw+g/hwZLzIyMhQXFycvLy/LHskAbh+OiBcZGRmW+5s3b66UlBSNGzdO/fv3V2BgoEJCQvTyyy9btjYA0Dg5Il4MHTrUslL63Xff1cqVK3XmzBldu3ZN+fn5io2N1bx58yTd2B7pscces/UxADQRzpjrdKvX3YCdFBYWWq5vtaTPXH758uU69VHb9s19mN82AOB8jogVt5KcnKysrCxJ0ujRo+Xq6tqg7QNoGI6KF0ajUQsXLpTRaNSMGTPUsWNHm9sA4FyOiBc//PCDJKl169aKiYlRfHx8pfJLly4pJSVF//73vzVnzhw9//zzNrUPwDEcES/c3Ny0du1avfzyy/r666+1bNkyLVu2rFKdnj17aubMmRo3bpxNbQNoWpwx18nKBTRK5sy/VHl/QWvM5cXFxTb1YV4y6OHhUWM986FIN48LgPM5IlbUJDc3VzExMZKkjh07chgj0Ig5Kl6sXbtWmZmZCggI0HPPPWfz/QCczxHx4urVq5JuHNQaHx+vX//611q9erWOHj1qmTzs2LGjysvL9c4772jv3r22PQQAh3DU74t27drp/fff16OPPmq1/OLFi8rPz2fOAviFc8ZcJ8kFAADqoLCwUNOmTVNxcbHc3d313nvvqVWrVs4eFgAnysvL0/Lly+Xi4qLo6Gi5ubFIGIB1JpNJklRWViY/Pz+tW7dOgwcPlqenp1q2bKlHH31UsbGx8vLykiQtXbrUmcMF4GRbt25VeHi4duzYoYkTJ2rz5s1KS0vTzp07tWjRIhmNRq1evVrPPPOMLl686OzhAvgFIbmARsn8I1qSrl27VmNdc/kdd9xhUx+enp6SpNLS0hrr/fzzz1bHBcD5HBErrDEYDHrhhRf0/fffq1mzZnrnnXcUFBRU73YB2I8j4sUbb7yhkpISjR49WgMHDrR9kAAaBUfEi4p9PPfcc/L29q5Sp2vXrho5cqQkKScnR2fPnrWpDwD254h4cfDgQc2aNUulpaWKjo7W3Llz1atXL7Vu3VpdunTRM888o7i4ODVv3lwnTpzQW2+9ZfuDAGgSnDHXSXIBjZKPj4/l+lZZd3O5tR/ktemjtu3XpQ8A9uWIWHGz0tJSTZ8+XUePHpUkLVq0SMOHD69XmwDsz97xYufOnfriiy/Upk0bzZo1q26DBNAoOPK/RSTV+IJCxbKTJ0/a1AcA+3NEvFi9erVMJpO6du2qp556ymqdnj17atiwYZKkzz//3LL1GoBfFmfMdZJcQKPUvn17S+bsVm/o5OXlSZL8/f1t6sNcv7bt33HHHRzmDDQyjogVFRmNRr3yyis6cOCAJGn27NkcmgbcJuwdL8z3XLx4UcHBwQoICKjyJy0tTZKUn59v+Yy3C4HGxxG/L+6++27LdU3bKrZu3dpyXVRUZFMfAOzPEfHC/FJTYGCgXFxcqq3Xr18/STf+m+XUqVM29QGgaXDGXCfJBTRKLi4uCgwMlCQdP3682nrnzp1TQUGBJFnq15a5fkFBgaUNa44dO1an9gHYnyNihZnJZNL8+fO1Y8cOSdILL7zAAc7AbcSR8QLA7c0R8aJv376W68uXL1dbr2JZy5YtbeoDgP05Il6Yt1Myn9VSnVuVA2j6nDHXSXIBjdZDDz0kSTp9+rROnDhhtc7nn39uuQ4LC6tT+5KUkpJitU5mZqbOnDlTp/YBOIa9Y4VZTEyMNm/eLEl69tlnNXPmzDq1A8B57BkvRowYoeTk5Br/mH+8t2vXzvLZ5MmT6/FEAOzF3r8vhgwZYjn0/auvvqq23uHDhy3XvXv3tqkPAI5h73jRvn17STfmJ2pKIGRkZFiu77rrLpv6ANA0OGOuk+QCGq3IyEjL8sKlS5dW+Uf08uXLWr16tSRpwIABNmfb+vXrp/79+0u6sYfhzW8MmUwmLV26VNKNw00ef/zxOj0HAPuyd6yQpL/85S9av369JOmJJ57QggUL6jlqAM5gz3jh6+ur3r171/jHfICjh4eH5TO2XAQaJ3v/vvD29rac2RQbG2t1b+Tc3FwlJydLunH2AvECaJzsHS9CQkIkSWfOnFFiYqLVOjk5Odq6daskqU+fPmrbtq1NfQBoGpwx1+n6xhtvvFHvVgA78PT0lKurq7788kudOXNGOTk58vf3l6urq7755hvNmjVLZ8+elZubm5YuXVolM5+YmKgnnnhCK1asUKdOnay+6dO9e3dt3rxZRUVFSk1N1a9+9Svdeeed+v777xUTE6M9e/ZIkl566SWFhoY65LkB2MbesWLNmjX64IMPJEkPPPCA3nrrLV2/fl1lZWVW/zRr1kzNmpG7BxojR/y2qElSUpLy8/PVqlUrTZgwoSEfDUADc0S86Nu3r5KTk3Xp0iXt3LlT7du3l4+Pj4qKirR9+3a9+uqrunLlitzd3fX++++TXAAaKXvHi7vvvlsJCQkyGo3au3evDAaD2rZtKw8PD/3444/asmWL5s2bp+LiYkk3VlzX55w5APZ18uRJnTlzRufOndO5c+eUlpamzMxMSVJwcLCuXr1qKfPw8JCnp6fl3sY41+lW7xYAO5oyZYry8vK0ceNGbd++Xdu3b69U7u7urjfffFODBg2qU/uDBg3Sm2++qYULFyonJ0cTJ06sUmfs2LHsqw40cvaMFfHx8Zbr1NRUBQUF1Vj/7bff1siRI23uB4Bj2Pu3BYCmw97xomPHjvroo480bdo0nT59Wn/84x+r1PHy8tK7775reQsRQONkz3jh7++v5cuXa9asWbp69apWr15tWQlRkZubm+bOncuWzkAjFx0drbS0NKtl06dPr/T3uswvOHquk+QCGr3o6GgNGTJEGzZs0HfffaeffvpJ7dq107333quoqCgFBATUq/3IyEj16dNHa9eu1aFDh3T+/Hm1bt1agYGBGjduXKX9ygA0XvaOFQCaDuIFgNqyd7wYOHCgtm7dqrVr12r37t3Kz89XeXm5OnXqpNDQUEVFRbF3OnCbsGe8ePDBB5WSkqKNGzdq//79OnXqlIqKitS8eXN17txZ99xzj8aNG6fu3bs34BMBuF05cq7TxcRx8gAAAAAAAAAAwAZsCg0AAAAAAAAAAGxCcgEAAAAAAAAAANiE5AIAAAAAAAAAALAJyQUAAAAAAAAAAGATkgsAAAAAAAAAAMAmJBcAAAAAAAAAAIBNSC4AAAAAAAAAAACbkFwAAAAAAAAAAAA2IbkAAAAAAAAAAABsQnIBAAAAAAAAAADYhOQCAAAAAAAAAACwCckFAAAAAAAAAABgE5ILAAAAAAAAAADAJiQXAAAAADQ5hw8fVkBAgAICApSYmOjs4SgxMdEynsOHD9erLXM78+bNs1oeFhamgIAAjR8/3mr5+PHjFRAQoLCwsHqNAwAAAL9sJBcAAAAAAAAAAIBNSC4AAAAAACxY2QAAAIDacHP2AAAAAAAAtZednV2v++Pi4hpoJAAAAPglY+UCAAAAAAAAAACwCckFAAAAAAAAAABgE7ZFAgAAAFCtxMREzZ8/X5IUGxur4OBgffrpp0pOTlZubq6Ki4vVoUMHPfDAA5oyZYr8/PyqtJGXl6fw8HBJ0vTp0zVjxgwdOXJEGzdu1JEjR3T+/HmVlpbqq6++UqtWrSz3Xbx4UevWrdO+ffuUl5cng8Egb29v9e3bV8OHD9ewYcPk4uJS62fZvn27Pv74Y2VlZeny5ctq166dQkJCNGnSJHXv3r3a+65du6Z9+/Zp//79ysjI0NmzZ2UwGOTl5aVOnTrpnnvu0bPPPqsuXbrUeiwmk8nm79EsICBAkhQZGal33nmn1n2ajR8/XmlpaerUqZN2795t+XzevHlKSkqy/D0/P9/SV0WxsbEaMGCABg8erCtXrmjgwIHauHHjLfuNjIxUZmamfH19tW/fPnl4eNg8dgAAADQeJBcAAAAA1Mr169c1depU7du3r9Lnp0+fVlxcnJKSkrRixQqFhITU2M6HH36o5cuXy2QyVVtn165devXVV1VcXFzp8/Pnz2vPnj3as2eP1q1bp5UrV8rX1/eWY1+4cKE+/vjjSp/973//06effqrPPvtMS5Ys0e9//3ur986cOVO7du2q8vmVK1d05coVnThxQvHx8Vq8eLEiIyNvOZaG+h6dqUWLFnr88ccVFxenb7/9VidPnlSPHj2qrZ+RkaHMzExJ0hNPPEFiAQAAoAkguQAAAACgVt5//32lp6crODhY48aNU9euXVVYWKitW7cqOTlZRUVFevHFF5WUlCR/f3+rbezcuVNZWVm6++67NWHCBPXu3VtGo1FHjx6Vu7u7JCktLU0zZsyQ0WiUq6urRo8erd/97ndq1aqVTp06pbi4OB07dkzffvutnn/+eSUkJNQ4Wb1+/Xqlp6erV69eioqKUo8ePVRUVKTdu3drw4YNKi0t1dy5c+Xn56fg4OAq9xuNRnXr1k1hYWHq16+f7rrrLrm5uencuXP6+uuvtWnTJhkMBr3++uvq0qWLgoKC7P492sPMmTM1ceKVPAInAAAJzUlEQVREzZ8/XxkZGWrfvr3WrFlTpV7nzp0lSWPHjrUcDp2QkGBZ4WJNQkKC5Xr06NENPHIAAAA4A8kFAAAAALWSnp6ukSNHasmSJZW2Ixo8eLAGDRqkBQsWqKSkRDExMfrnP/9ptY2srCwFBwdr1apVatGiheXz3/72t5JuTOTPmzdPRqNRzZo108qVKzVkyBBLvb59+2rYsGF65ZVXlJKSoqysLP3tb3/TjBkzahz3/fffr48++qhSEiIkJEQPPPCApk6dKqPRqIULFyolJUXNmlU+mm7+/Pnq1q1blXb79u2riIgITZgwQWPGjFFBQYGWLVtmmXC35/doD35+fvLz85OXl5ckyd3dXT179qy2fo8ePRQUFKSvv/5aycnJmjVrltUkT0lJibZs2SJJCg4OdmjCBAAAAPbDgc4AAAAAaqVNmzZauHCh1XMORo0apdDQUEnSl19+qdzcXKttNGvWTEuWLKmUWKho165dys/Pl3TjDfeKiYWKbSxevFg+Pj6SpPj4eJWVlVU7bnd3d7399ttWJ74HDx6sp556SpL0/fffa//+/VXqWEssVNSxY0dNnjxZkvTVV1/p8uXLNdZviO+xsRg7dqwk6fLly9q5c6fVOikpKSoqKpJ04/kAAADQNJBcAAAAAFArQ4cOtbzVbo15kl6SDhw4YLXOwIEDazz4uOLkvnni2pqWLVtq+PDhkqTCwkKdOHGi2rqhoaE1HpBcm3FX9NNPP+ns2bP6z3/+o5ycHOXk5FiSJSaTqcaxSA3zPTYWjzzyiCXJc/OZFmbmz729vfXoo486bGwAAACwL7ZFAgAAAFAr/fr1q7F8wIABluvs7GyrdXr16lVjGzk5OZIkLy8vBQQE1Fh34MCBli2IsrOz1b9/f6v1bjXuPn36yN3dXWVlZdWOOz09XXFxcTpw4IAuXLhQY3uFhYU1ljfE99hYeHh4KDIyUv/4xz906NAhnT17tlLyKDc3V99++60kacSIERzkDAAA0ISwcgEAAABArbRt27bW5dVtDdSqVasa2zDf5+PjU+Xsg5r6q2lC/1bjdnNzk7e3d6X+K/r73/+uUaNGafPmzbdMLEjStWvXaixviO+xMRkzZoxcXFxkMpn0ySefVCqruJphzJgxjh4aAAAA7IjkAgAAAACHcXV1dfYQbJKWlqalS5fKZDLJ19dXs2fP1ieffKJDhw4pPT1d2dnZys7O1tq1ay33mEwm5w3YCbp166Z7771XkpSYmCij0ShJKi0t1ebNmyXdWGXSo0cPp40RAAAADY9tkQAAAADUyq3e2q9Ybl4JYCvzfYWFhSovL69x9ULF/sz7/t+qnjXXr1+3rBC4edybNm2SdCMpsm7dOnXv3t1qG1euXKmxD1vG0xDfo6ONHTtWBw8e1I8//qi9e/cqPDxcO3futKwoYdUCAABA08PKBQAAAAC1kp6eXmP5sWPHLNe3Oi+hOub7DAaD5fyF6pj38r9Vf7cad2ZmpsrKyqy2Yx5DQEBAtYkFScrIyKixD1vG0xDfo6OFh4erXbt2kqSEhIRK/9uyZUsNHTrUaWMDAACAfZBcAAAAAFArKSkpMhgM1ZZX3G///vvvr1MfoaGhlmvzqgFrioqKtGXLFkmSr6+v+vTpU23d/fv3q6CgoNrymsZ9/fp1SVJJSUm19xsMBiUlJVVbfjNHfI/10aJFC0k3tjWqLXd3dz355JOSpNTUVB05ckQHDx6UdOMgZ3ObAAAAaDpILgAAAAColYsXL2rx4sVWzxRISEjQ/v37JUn33XdfjW/51yQsLEydO3eWdCO5kJqaWqVOeXm5/u///s+y5c4zzzwjN7fqd3wtKyvT/PnzLasTKvriiy8sk/ndunWrlNwwfyZJp0+f1jfffFPl/uvXr+u1117T+fPna/eAcsz3WB/t27eXdGOcV69erfV9o0aNUrNmzWQ0GvXSSy9Znm/06NF2GScAAACcizMXAAAAANRK//79lZiYqLy8PD399NPq2rWrCgsLtXXrVsub+56enlq0aFGd+3B1ddXbb7+tqKgoGY1GvfjiixozZowiIiLUqlUrnT59WnFxcZYtkXr16qU//OEPtxz3gQMHNGrUKEVFRalHjx4qLi7Wrl27tH79ehmNRrm6umrx4sVVzngYOXKkdu/erfLyck2dOlUTJ07UoEGD1KJFC2VnZ2vdunXKysrSoEGDdOTIkVo9oyO+x/oICgrSJ598ovLycr322msaP3682rRpYym/66675OnpWeW+zp07KzQ0VKmpqZZkS//+/dWrVy+HjR0AAACOQ3IBAAAAQK28/PLLio2N1d69e5WWllal/M4779SKFSvk7+9fr36Cg4P1wQcfaM6cOSouLlZ8fLzi4+Or1Bs4cKBWrlwpDw+PGtsbN26cevfurU2bNmnu3LlVyt3d3bVkyRIFBwdXKXv44Yc1duxYbdy4UVeuXNFf//rXKnVGjBihkSNHKioqqlbP56jvsa6GDh2qVatWKTc3V9u3b9f27dsrlcfGxuqee+6xeu+YMWMqrTZh1QIAAEDTRXIBAAAAQK24ubnpo48+0qeffqrk5GSdPHlSxcXF8vPz05AhQzRlyhT5+fk1SF8RERHasWOH4uLilJqaqrNnz6qkpETe3t7q27evhg0bpmHDhlVZaVCdmJgYhYaGatOmTcrKytJPP/2ktm3b6r777tOkSZNq3H4oOjpa9957rzZu3KjMzEyVlJTI19dXgYGBevLJJxUREaHDhw/X+tkc+T3WRYsWLbRhwwatWbNGX3zxhc6cOSODwaDy8vJb3vvQQw/Jx8dHhYWFuuOOOzRs2DAHjBgAAADO4GKyttEnAAAAAEhKTEzU/PnzJdX8xjogST/88IPCwsJUXl6uMWPGKCYmxtlDAgAAgJ1woDMAAAAAoEGYz2qQbmyRBAAAgKaL5AIAAAAAoN6Kioq0fv16SdJvfvMbBQYGOnlEAAAAsCfOXAAAAAAA1ElBQYF+/vlnnTt3Th9++KEuXbokSZo2bZqTRwYAAAB7I7kAAAAAAKiT2bNnKy0trdJnw4cP14MPPuikEQEAAMBRSC4AAAAAAOqlefPm6tKliyIjI/Xcc885ezgAAABwABeTyWRy9iAAAAAAAAAAAMDtgwOdAQAAAAAAAACATUguAAAAAAAAAAAAm5BcAAAAAAAAAAAANiG5AAAAAAAAAAAAbEJyAQAAAAAAAAAA2ITkAgAAAAAAAAAAsAnJBQAAAAAAAAAAYBOSCwAAAAAAAAAAwCYkFwAAAAAAAAAAgE1ILgAAAAAAAAAAAJuQXAAAAAAAAAAAADYhuQAAAAAAAAAAAGxCcgEAAAAAAAAAANjk/wHM1gBJtk7PowAAAABJRU5ErkJggg==\n","text/plain":["<Figure size 864x576 with 1 Axes>"]},"metadata":{"tags":[],"image/png":{"width":779,"height":489}}}]},{"cell_type":"markdown","metadata":{"id":"gwisteWdEg0L"},"source":["## Predictions on Raw Text"]},{"cell_type":"code","metadata":{"id":"U8A7NUESEYyZ","executionInfo":{"status":"ok","timestamp":1608824545918,"user_tz":-330,"elapsed":1158,"user":{"displayName":"Mallika Subramanian","photoUrl":"https://lh3.googleusercontent.com/a-/AOh14Gj3lusT2xteZqIvBVDRAXN5RCnCIyEjAhNrS9hRhv0=s64","userId":"05149503641230543668"}}},"source":["review_text = \"I love completing my todos! Best app EVER!!!\""],"execution_count":74,"outputs":[]},{"cell_type":"code","metadata":{"id":"9RWOAoM1EqG5","executionInfo":{"status":"ok","timestamp":1608824545919,"user_tz":-330,"elapsed":993,"user":{"displayName":"Mallika Subramanian","photoUrl":"https://lh3.googleusercontent.com/a-/AOh14Gj3lusT2xteZqIvBVDRAXN5RCnCIyEjAhNrS9hRhv0=s64","userId":"05149503641230543668"}}},"source":["encoded_review = tokenizer.encode_plus(\n","    review_text,\n","    max_length=MAX_LEN,\n","    add_special_tokens = True,\n","    return_token_type_ids = False,\n","    pad_to_max_length = True,\n","    return_attention_mask = True,\n","    return_tensors = 'pt'\n",")"],"execution_count":75,"outputs":[]},{"cell_type":"code","metadata":{"id":"md3WrelVE5ju","executionInfo":{"status":"ok","timestamp":1608824545920,"user_tz":-330,"elapsed":811,"user":{"displayName":"Mallika Subramanian","photoUrl":"https://lh3.googleusercontent.com/a-/AOh14Gj3lusT2xteZqIvBVDRAXN5RCnCIyEjAhNrS9hRhv0=s64","userId":"05149503641230543668"}}},"source":["input_ids = encoded_review['input_ids'].to(device)\n","attention_mask = encoded_review['attention_mask'].to(device)"],"execution_count":76,"outputs":[]},{"cell_type":"code","metadata":{"id":"ua_A4ow9E_tf","executionInfo":{"status":"ok","timestamp":1608824546367,"user_tz":-330,"elapsed":803,"user":{"displayName":"Mallika Subramanian","photoUrl":"https://lh3.googleusercontent.com/a-/AOh14Gj3lusT2xteZqIvBVDRAXN5RCnCIyEjAhNrS9hRhv0=s64","userId":"05149503641230543668"}}},"source":["output = model(input_ids, attention_mask)\n","_,prediction = torch.max(output, dim=1)"],"execution_count":77,"outputs":[]},{"cell_type":"code","metadata":{"colab":{"base_uri":"https://localhost:8080/"},"id":"N6Skc7FWFF_e","executionInfo":{"status":"ok","timestamp":1608824546767,"user_tz":-330,"elapsed":948,"user":{"displayName":"Mallika Subramanian","photoUrl":"https://lh3.googleusercontent.com/a-/AOh14Gj3lusT2xteZqIvBVDRAXN5RCnCIyEjAhNrS9hRhv0=s64","userId":"05149503641230543668"}},"outputId":"a1a6b161-22f9-4fa0-c833-aec30edbc493"},"source":["print(f'Review Text:{review_text}')\n","print(f'Sentiment : {class_names[prediction]}')"],"execution_count":78,"outputs":[{"output_type":"stream","text":["Review Text:I love completing my todos! Best app EVER!!!\n","Sentiment : positive\n"],"name":"stdout"}]},{"cell_type":"code","metadata":{"id":"FgmW-9kWFOQP","executionInfo":{"status":"ok","timestamp":1608823663943,"user_tz":-330,"elapsed":72,"user":{"displayName":"Mallika Subramanian","photoUrl":"https://lh3.googleusercontent.com/a-/AOh14Gj3lusT2xteZqIvBVDRAXN5RCnCIyEjAhNrS9hRhv0=s64","userId":"05149503641230543668"}}},"source":[""],"execution_count":68,"outputs":[]}]}