├── README.md └── Tweet-Semantic-Meaning.ipynb /README.md: -------------------------------------------------------------------------------- 1 | #### Semantic Meaning of Transformers in Data Science 2 | - Proper labeling and assigning semantic meaning to tweets using transformers in Data Science. 3 | - Representation on the 2D grid. 4 | -------------------------------------------------------------------------------- /Tweet-Semantic-Meaning.ipynb: -------------------------------------------------------------------------------- 1 | { 2 | "nbformat": 4, 3 | "nbformat_minor": 0, 4 | "metadata": { 5 | "colab": { 6 | "name": "Zadacha3_181225.ipynb", 7 | "provenance": [] 8 | }, 9 | "kernelspec": { 10 | "name": "python3", 11 | "display_name": "Python 3" 12 | }, 13 | "language_info": { 14 | "name": "python" 15 | } 16 | }, 17 | "cells": [ 18 | { 19 | "cell_type": "code", 20 | "execution_count": 1, 21 | "metadata": { 22 | "colab": { 23 | "base_uri": "https://localhost:8080/" 24 | }, 25 | "id": "sQoyuIuJMpWo", 26 | "outputId": "16ecf838-ff02-43ab-906e-085f911f7d47" 27 | }, 28 | "outputs": [ 29 | { 30 | "output_type": "stream", 31 | "name": "stdout", 32 | "text": [ 33 | "Mounted at /content/drive\n" 34 | ] 35 | } 36 | ], 37 | "source": [ 38 | "#add your code \n", 39 | "from google.colab import drive\n", 40 | "drive.mount('/content/drive')" 41 | ] 42 | }, 43 | { 44 | "cell_type": "code", 45 | "source": [ 46 | "import pandas as pd\n", 47 | "\n", 48 | "df = pd.read_csv('/content/train_3.csv')" 49 | ], 50 | "metadata": { 51 | "id": "0O1yfskiMze9" 52 | }, 53 | "execution_count": 2, 54 | "outputs": [] 55 | }, 56 | { 57 | "cell_type": "code", 58 | "source": [ 59 | "df.head()" 60 | ], 61 | "metadata": { 62 | "colab": { 63 | "base_uri": "https://localhost:8080/", 64 | "height": 206 65 | }, 66 | "id": "2EukyL2sM5HV", 67 | "outputId": "98fd7c7e-2c69-46bb-fa7e-55858c72c3dc" 68 | }, 69 | "execution_count": 3, 70 | "outputs": [ 71 | { 72 | "output_type": "execute_result", 73 | "data": { 74 | "text/html": [ 75 | "\n", 76 | "
\n", 77 | "
\n", 78 | "
\n", 79 | "\n", 92 | "\n", 93 | " \n", 94 | " \n", 95 | " \n", 96 | " \n", 97 | " \n", 98 | " \n", 99 | " \n", 100 | " \n", 101 | " \n", 102 | " \n", 103 | " \n", 104 | " \n", 105 | " \n", 106 | " \n", 107 | " \n", 108 | " \n", 109 | " \n", 110 | " \n", 111 | " \n", 112 | " \n", 113 | " \n", 114 | " \n", 115 | " \n", 116 | " \n", 117 | " \n", 118 | " \n", 119 | " \n", 120 | " \n", 121 | " \n", 122 | " \n", 123 | " \n", 124 | " \n", 125 | " \n", 126 | " \n", 127 | "
TweetSentiment
0TRENDING: New Yorkers encounter empty supermar...Extremely Negative
1When I couldn't find hand sanitizer at Fred Me...Positive
2Find out how you can protect yourself and love...Extremely Positive
3#Panic buying hits #NewYork City as anxious sh...Negative
4#toiletpaper #dunnypaper #coronavirus #coronav...Neutral
\n", 128 | "
\n", 129 | " \n", 139 | " \n", 140 | " \n", 177 | "\n", 178 | " \n", 202 | "
\n", 203 | "
\n", 204 | " " 205 | ], 206 | "text/plain": [ 207 | " Tweet Sentiment\n", 208 | "0 TRENDING: New Yorkers encounter empty supermar... Extremely Negative\n", 209 | "1 When I couldn't find hand sanitizer at Fred Me... Positive\n", 210 | "2 Find out how you can protect yourself and love... Extremely Positive\n", 211 | "3 #Panic buying hits #NewYork City as anxious sh... Negative\n", 212 | "4 #toiletpaper #dunnypaper #coronavirus #coronav... Neutral" 213 | ] 214 | }, 215 | "metadata": {}, 216 | "execution_count": 3 217 | } 218 | ] 219 | }, 220 | { 221 | "cell_type": "code", 222 | "source": [ 223 | "from sklearn.preprocessing import LabelEncoder\n", 224 | "encoder = LabelEncoder()\n", 225 | "df['Sentiment'] = encoder.fit_transform(df['Sentiment'])" 226 | ], 227 | "metadata": { 228 | "id": "KVcwpi0TM_nf" 229 | }, 230 | "execution_count": 5, 231 | "outputs": [] 232 | }, 233 | { 234 | "cell_type": "code", 235 | "source": [ 236 | "pip install transformers" 237 | ], 238 | "metadata": { 239 | "colab": { 240 | "base_uri": "https://localhost:8080/" 241 | }, 242 | "id": "VRKOjwIJOIYj", 243 | "outputId": "e1338110-87b8-4967-fbc9-34236cef4a2c" 244 | }, 245 | "execution_count": 6, 246 | "outputs": [ 247 | { 248 | "output_type": "stream", 249 | "name": "stdout", 250 | "text": [ 251 | "Collecting transformers\n", 252 | " Downloading transformers-4.16.1-py3-none-any.whl (3.5 MB)\n", 253 | "\u001b[K |████████████████████████████████| 3.5 MB 5.2 MB/s \n", 254 | "\u001b[?25hCollecting huggingface-hub<1.0,>=0.1.0\n", 255 | " Downloading huggingface_hub-0.4.0-py3-none-any.whl (67 kB)\n", 256 | "\u001b[K |████████████████████████████████| 67 kB 4.0 MB/s \n", 257 | "\u001b[?25hRequirement already satisfied: tqdm>=4.27 in /usr/local/lib/python3.7/dist-packages (from transformers) (4.62.3)\n", 258 | "Collecting sacremoses\n", 259 | " Downloading sacremoses-0.0.47-py2.py3-none-any.whl (895 kB)\n", 260 | "\u001b[K |████████████████████████████████| 895 kB 53.6 MB/s \n", 261 | "\u001b[?25hRequirement already satisfied: filelock in /usr/local/lib/python3.7/dist-packages (from transformers) (3.4.2)\n", 262 | "Collecting pyyaml>=5.1\n", 263 | " Downloading PyYAML-6.0-cp37-cp37m-manylinux_2_5_x86_64.manylinux1_x86_64.manylinux_2_12_x86_64.manylinux2010_x86_64.whl (596 kB)\n", 264 | "\u001b[K |████████████████████████████████| 596 kB 55.1 MB/s \n", 265 | "\u001b[?25hRequirement already satisfied: packaging>=20.0 in /usr/local/lib/python3.7/dist-packages (from transformers) (21.3)\n", 266 | "Requirement already satisfied: importlib-metadata in /usr/local/lib/python3.7/dist-packages (from transformers) (4.10.1)\n", 267 | "Requirement already satisfied: regex!=2019.12.17 in /usr/local/lib/python3.7/dist-packages (from transformers) (2019.12.20)\n", 268 | "Requirement already satisfied: numpy>=1.17 in /usr/local/lib/python3.7/dist-packages (from transformers) (1.19.5)\n", 269 | "Requirement already satisfied: requests in /usr/local/lib/python3.7/dist-packages (from transformers) (2.23.0)\n", 270 | "Collecting tokenizers!=0.11.3,>=0.10.1\n", 271 | " Downloading tokenizers-0.11.4-cp37-cp37m-manylinux_2_12_x86_64.manylinux2010_x86_64.whl (6.8 MB)\n", 272 | "\u001b[K |████████████████████████████████| 6.8 MB 36.5 MB/s \n", 273 | "\u001b[?25hRequirement already satisfied: typing-extensions>=3.7.4.3 in /usr/local/lib/python3.7/dist-packages (from huggingface-hub<1.0,>=0.1.0->transformers) (3.10.0.2)\n", 274 | "Requirement already satisfied: pyparsing!=3.0.5,>=2.0.2 in /usr/local/lib/python3.7/dist-packages (from packaging>=20.0->transformers) (3.0.7)\n", 275 | "Requirement already satisfied: zipp>=0.5 in /usr/local/lib/python3.7/dist-packages (from importlib-metadata->transformers) (3.7.0)\n", 276 | "Requirement already satisfied: urllib3!=1.25.0,!=1.25.1,<1.26,>=1.21.1 in /usr/local/lib/python3.7/dist-packages (from requests->transformers) (1.24.3)\n", 277 | "Requirement already satisfied: idna<3,>=2.5 in /usr/local/lib/python3.7/dist-packages (from requests->transformers) (2.10)\n", 278 | "Requirement already satisfied: certifi>=2017.4.17 in /usr/local/lib/python3.7/dist-packages (from requests->transformers) (2021.10.8)\n", 279 | "Requirement already satisfied: chardet<4,>=3.0.2 in /usr/local/lib/python3.7/dist-packages (from requests->transformers) (3.0.4)\n", 280 | "Requirement already satisfied: joblib in /usr/local/lib/python3.7/dist-packages (from sacremoses->transformers) (1.1.0)\n", 281 | "Requirement already satisfied: six in /usr/local/lib/python3.7/dist-packages (from sacremoses->transformers) (1.15.0)\n", 282 | "Requirement already satisfied: click in /usr/local/lib/python3.7/dist-packages (from sacremoses->transformers) (7.1.2)\n", 283 | "Installing collected packages: pyyaml, tokenizers, sacremoses, huggingface-hub, transformers\n", 284 | " Attempting uninstall: pyyaml\n", 285 | " Found existing installation: PyYAML 3.13\n", 286 | " Uninstalling PyYAML-3.13:\n", 287 | " Successfully uninstalled PyYAML-3.13\n", 288 | "Successfully installed huggingface-hub-0.4.0 pyyaml-6.0 sacremoses-0.0.47 tokenizers-0.11.4 transformers-4.16.1\n" 289 | ] 290 | } 291 | ] 292 | }, 293 | { 294 | "cell_type": "code", 295 | "source": [ 296 | "from absl import logging\n", 297 | "\n", 298 | "import tensorflow as tf\n", 299 | "\n", 300 | "import tensorflow_hub as hub\n", 301 | "import matplotlib.pyplot as plt\n", 302 | "import numpy as np\n", 303 | "import os\n", 304 | "import pandas as pd\n", 305 | "import re\n", 306 | "import seaborn as sns\n", 307 | "\n", 308 | "module_url = \"https://tfhub.dev/google/universal-sentence-encoder/4\"\n", 309 | "model = hub.load(module_url)\n", 310 | "print (\"module %s loaded\" % module_url)\n", 311 | "def embed(input):\n", 312 | " return model(input)" 313 | ], 314 | "metadata": { 315 | "colab": { 316 | "base_uri": "https://localhost:8080/" 317 | }, 318 | "id": "X7k7YbfgOWlh", 319 | "outputId": "8cc787f3-d6c9-4ccd-cf04-0749b7d93b7b" 320 | }, 321 | "execution_count": 7, 322 | "outputs": [ 323 | { 324 | "output_type": "stream", 325 | "name": "stdout", 326 | "text": [ 327 | "module https://tfhub.dev/google/universal-sentence-encoder/4 loaded\n" 328 | ] 329 | } 330 | ] 331 | }, 332 | { 333 | "cell_type": "code", 334 | "source": [ 335 | "# Reduce logging output.\n", 336 | "logging.set_verbosity(logging.ERROR)\n", 337 | "\n", 338 | "message_embeddings = embed(df.Tweet.values)" 339 | ], 340 | "metadata": { 341 | "id": "DdJj1Ww9OgRi" 342 | }, 343 | "execution_count": 9, 344 | "outputs": [] 345 | }, 346 | { 347 | "cell_type": "code", 348 | "source": [ 349 | " message_embeddings" 350 | ], 351 | "metadata": { 352 | "colab": { 353 | "base_uri": "https://localhost:8080/" 354 | }, 355 | "id": "faO2SlyZOm1O", 356 | "outputId": "833d5ae8-9c44-4817-ffcc-090385cdd48f" 357 | }, 358 | "execution_count": 11, 359 | "outputs": [ 360 | { 361 | "output_type": "execute_result", 362 | "data": { 363 | "text/plain": [ 364 | "" 378 | ] 379 | }, 380 | "metadata": {}, 381 | "execution_count": 11 382 | } 383 | ] 384 | }, 385 | { 386 | "cell_type": "code", 387 | "source": [ 388 | "from sklearn.cluster import KMeans\n", 389 | "\n", 390 | "km = KMeans(n_clusters=3)\n", 391 | "km.fit(message_embeddings)\n", 392 | "clusters = km.labels_.tolist()" 393 | ], 394 | "metadata": { 395 | "id": "AGP2xK12Oxhq" 396 | }, 397 | "execution_count": 12, 398 | "outputs": [] 399 | }, 400 | { 401 | "cell_type": "code", 402 | "source": [ 403 | "from sklearn.decomposition import PCA\n", 404 | "import numpy as np\n", 405 | "\n", 406 | "data = message_embeddings\n", 407 | "\n", 408 | "pca = PCA(3)\n", 409 | " \n", 410 | "#Transform the received data\n", 411 | "df_new = pca.fit_transform(data)" 412 | ], 413 | "metadata": { 414 | "id": "m4YiZ-pcO7kY" 415 | }, 416 | "execution_count": 13, 417 | "outputs": [] 418 | }, 419 | { 420 | "cell_type": "code", 421 | "source": [ 422 | "#Initialize the class object\n", 423 | "kmeans = KMeans(n_clusters=5)\n", 424 | " \n", 425 | "#predict the labels of clusters.\n", 426 | "label = kmeans.fit_predict(df_new)\n", 427 | " \n", 428 | "#Getting unique labels\n", 429 | "u_labels = np.unique(label)\n", 430 | " \n", 431 | "labels = kmeans.labels_" 432 | ], 433 | "metadata": { 434 | "id": "dvwQ38M5PBm4" 435 | }, 436 | "execution_count": 14, 437 | "outputs": [] 438 | }, 439 | { 440 | "cell_type": "code", 441 | "source": [ 442 | "import matplotlib.pyplot as plt\n", 443 | "import seaborn as sns\n", 444 | "%matplotlib inline\n", 445 | "#plotting the results:\n", 446 | "for i in u_labels:\n", 447 | " plt.scatter(df_new[label == i , 0] , df_new[label == i , 1] , label = i)\n", 448 | "plt.legend()\n", 449 | "plt.show()" 450 | ], 451 | "metadata": { 452 | "colab": { 453 | "base_uri": "https://localhost:8080/", 454 | "height": 265 455 | }, 456 | "id": "qRG5DnVCPIKl", 457 | "outputId": "62cf8e81-397e-4b76-fc06-2b83f52e4bca" 458 | }, 459 | "execution_count": 15, 460 | "outputs": [ 461 | { 462 | "output_type": "display_data", 463 | "data": { 464 | "image/png": "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\n", 465 | "text/plain": [ 466 | "
" 467 | ] 468 | }, 469 | "metadata": { 470 | "needs_background": "light" 471 | } 472 | } 473 | ] 474 | } 475 | ] 476 | } --------------------------------------------------------------------------------