Hugging Face's logo

MarkusHenriksson13
/
Financial_Sentiment_Classifier

Model card Files
xet
Community
MarkusHenriksson13 commited on
Commit
845fa17
·
verified ·
1 Parent(s): 66ae1c9

Delete Assignment_2.ipynb

Browse files
Files changed (1) hide show
  1. Assignment_2.ipynb +0 -1551
Assignment_2.ipynb DELETED
@@ -1,1551 +0,0 @@
1
- {
2
- "cells": [
3
- {
4
- "cell_type": "code",
5
- "execution_count": 1,
6
- "metadata": {
7
- "colab": {
8
- "base_uri": "https://localhost:8080/"
9
- },
10
- "id": "N3shQZoZPScM",
11
- "outputId": "63642e05-bd32-4fd9-f029-8f50148a1e8a"
12
- },
13
- "outputs": [],
14
- "source": [
15
- "!pip install -U sentence_transformers --q"
16
- ]
17
- },
18
- {
19
- "cell_type": "code",
20
- "execution_count": 2,
21
- "metadata": {
22
- "colab": {
23
- "base_uri": "https://localhost:8080/"
24
- },
25
- "id": "rcBH0FzwVOk6",
26
- "outputId": "f5b4b762-9b30-4474-d1d0-7ba3ab68a2ef"
27
- },
28
- "outputs": [
29
- {
30
- "name": "stdout",
31
- "output_type": "stream",
32
- "text": [
33
- "\n",
34
- "\u001b[1m[\u001b[0m\u001b[34;49mnotice\u001b[0m\u001b[1;39;49m]\u001b[0m\u001b[39;49m A new release of pip is available: \u001b[0m\u001b[31;49m24.0\u001b[0m\u001b[39;49m -> \u001b[0m\u001b[32;49m25.0.1\u001b[0m\n",
35
- "\u001b[1m[\u001b[0m\u001b[34;49mnotice\u001b[0m\u001b[1;39;49m]\u001b[0m\u001b[39;49m To update, run: \u001b[0m\u001b[32;49mpip3 install --upgrade pip\u001b[0m\n",
36
- "Note: you may need to restart the kernel to use updated packages.\n"
37
- ]
38
- }
39
- ],
40
- "source": [
41
- "pip install datasets --q"
42
- ]
43
- },
44
- {
45
- "cell_type": "code",
46
- "execution_count": 3,
47
- "metadata": {
48
- "colab": {
49
- "base_uri": "https://localhost:8080/"
50
- },
51
- "id": "y-pDMu97XyVd",
52
- "outputId": "737160a3-2c34-4293-a129-bb053cd91117"
53
- },
54
- "outputs": [
55
- {
56
- "name": "stdout",
57
- "output_type": "stream",
58
- "text": [
59
- "Collecting sentence-transformers\n",
60
- " Using cached sentence_transformers-3.4.1-py3-none-any.whl.metadata (10 kB)\n",
61
- "Requirement already satisfied: scikit-learn in /Library/Frameworks/Python.framework/Versions/3.12/lib/python3.12/site-packages (1.4.1.post1)\n",
62
- "Requirement already satisfied: pandas in /Library/Frameworks/Python.framework/Versions/3.12/lib/python3.12/site-packages (2.2.1)\n",
63
- "Collecting torch\n",
64
- " Downloading torch-2.6.0-cp312-none-macosx_11_0_arm64.whl.metadata (28 kB)\n",
65
- "Collecting transformers<5.0.0,>=4.41.0 (from sentence-transformers)\n",
66
- " Downloading transformers-4.48.3-py3-none-any.whl.metadata (44 kB)\n",
67
- "\u001b[2K \u001b[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[32m44.4/44.4 kB\u001b[0m \u001b[31m2.9 MB/s\u001b[0m eta \u001b[36m0:00:00\u001b[0m\n",
68
- "\u001b[?25hRequirement already satisfied: tqdm in /Library/Frameworks/Python.framework/Versions/3.12/lib/python3.12/site-packages (from sentence-transformers) (4.67.1)\n",
69
- "Requirement already satisfied: scipy in /Library/Frameworks/Python.framework/Versions/3.12/lib/python3.12/site-packages (from sentence-transformers) (1.12.0)\n",
70
- "Requirement already satisfied: huggingface-hub>=0.20.0 in /Library/Frameworks/Python.framework/Versions/3.12/lib/python3.12/site-packages (from sentence-transformers) (0.28.1)\n",
71
- "Requirement already satisfied: Pillow in /Library/Frameworks/Python.framework/Versions/3.12/lib/python3.12/site-packages (from sentence-transformers) (10.2.0)\n",
72
- "Requirement already satisfied: numpy<2.0,>=1.19.5 in /Library/Frameworks/Python.framework/Versions/3.12/lib/python3.12/site-packages (from scikit-learn) (1.26.4)\n",
73
- "Requirement already satisfied: joblib>=1.2.0 in /Library/Frameworks/Python.framework/Versions/3.12/lib/python3.12/site-packages (from scikit-learn) (1.3.2)\n",
74
- "Requirement already satisfied: threadpoolctl>=2.0.0 in /Library/Frameworks/Python.framework/Versions/3.12/lib/python3.12/site-packages (from scikit-learn) (3.3.0)\n",
75
- "Requirement already satisfied: python-dateutil>=2.8.2 in /Users/markushenriksson/Library/Python/3.12/lib/python/site-packages (from pandas) (2.9.0.post0)\n",
76
- "Requirement already satisfied: pytz>=2020.1 in /Library/Frameworks/Python.framework/Versions/3.12/lib/python3.12/site-packages (from pandas) (2024.1)\n",
77
- "Requirement already satisfied: tzdata>=2022.7 in /Library/Frameworks/Python.framework/Versions/3.12/lib/python3.12/site-packages (from pandas) (2024.1)\n",
78
- "Requirement already satisfied: filelock in /Library/Frameworks/Python.framework/Versions/3.12/lib/python3.12/site-packages (from torch) (3.16.0)\n",
79
- "Requirement already satisfied: typing-extensions>=4.10.0 in /Library/Frameworks/Python.framework/Versions/3.12/lib/python3.12/site-packages (from torch) (4.10.0)\n",
80
- "Collecting networkx (from torch)\n",
81
- " Using cached networkx-3.4.2-py3-none-any.whl.metadata (6.3 kB)\n",
82
- "Requirement already satisfied: jinja2 in /Library/Frameworks/Python.framework/Versions/3.12/lib/python3.12/site-packages (from torch) (3.1.3)\n",
83
- "Requirement already satisfied: fsspec in /Library/Frameworks/Python.framework/Versions/3.12/lib/python3.12/site-packages (from torch) (2024.2.0)\n",
84
- "Requirement already satisfied: setuptools in /Library/Frameworks/Python.framework/Versions/3.12/lib/python3.12/site-packages (from torch) (69.1.1)\n",
85
- "Collecting sympy==1.13.1 (from torch)\n",
86
- " Using cached sympy-1.13.1-py3-none-any.whl.metadata (12 kB)\n",
87
- "Collecting mpmath<1.4,>=1.1.0 (from sympy==1.13.1->torch)\n",
88
- " Using cached mpmath-1.3.0-py3-none-any.whl.metadata (8.6 kB)\n",
89
- "Requirement already satisfied: packaging>=20.9 in /Users/markushenriksson/Library/Python/3.12/lib/python/site-packages (from huggingface-hub>=0.20.0->sentence-transformers) (24.0)\n",
90
- "Requirement already satisfied: pyyaml>=5.1 in /Library/Frameworks/Python.framework/Versions/3.12/lib/python3.12/site-packages (from huggingface-hub>=0.20.0->sentence-transformers) (6.0.1)\n",
91
- "Requirement already satisfied: requests in /Library/Frameworks/Python.framework/Versions/3.12/lib/python3.12/site-packages (from huggingface-hub>=0.20.0->sentence-transformers) (2.32.3)\n",
92
- "Requirement already satisfied: six>=1.5 in /Users/markushenriksson/Library/Python/3.12/lib/python/site-packages (from python-dateutil>=2.8.2->pandas) (1.16.0)\n",
93
- "Collecting regex!=2019.12.17 (from transformers<5.0.0,>=4.41.0->sentence-transformers)\n",
94
- " Using cached regex-2024.11.6-cp312-cp312-macosx_11_0_arm64.whl.metadata (40 kB)\n",
95
- "Collecting tokenizers<0.22,>=0.21 (from transformers<5.0.0,>=4.41.0->sentence-transformers)\n",
96
- " Downloading tokenizers-0.21.0-cp39-abi3-macosx_11_0_arm64.whl.metadata (6.7 kB)\n",
97
- "Collecting safetensors>=0.4.1 (from transformers<5.0.0,>=4.41.0->sentence-transformers)\n",
98
- " Downloading safetensors-0.5.2-cp38-abi3-macosx_11_0_arm64.whl.metadata (3.8 kB)\n",
99
- "Requirement already satisfied: MarkupSafe>=2.0 in /Library/Frameworks/Python.framework/Versions/3.12/lib/python3.12/site-packages (from jinja2->torch) (2.1.5)\n",
100
- "Requirement already satisfied: charset-normalizer<4,>=2 in /Library/Frameworks/Python.framework/Versions/3.12/lib/python3.12/site-packages (from requests->huggingface-hub>=0.20.0->sentence-transformers) (3.3.2)\n",
101
- "Requirement already satisfied: idna<4,>=2.5 in /Library/Frameworks/Python.framework/Versions/3.12/lib/python3.12/site-packages (from requests->huggingface-hub>=0.20.0->sentence-transformers) (3.6)\n",
102
- "Requirement already satisfied: urllib3<3,>=1.21.1 in /Library/Frameworks/Python.framework/Versions/3.12/lib/python3.12/site-packages (from requests->huggingface-hub>=0.20.0->sentence-transformers) (2.0.7)\n",
103
- "Requirement already satisfied: certifi>=2017.4.17 in /Library/Frameworks/Python.framework/Versions/3.12/lib/python3.12/site-packages (from requests->huggingface-hub>=0.20.0->sentence-transformers) (2024.2.2)\n",
104
- "Using cached sentence_transformers-3.4.1-py3-none-any.whl (275 kB)\n",
105
- "Downloading torch-2.6.0-cp312-none-macosx_11_0_arm64.whl (66.5 MB)\n",
106
- "\u001b[2K \u001b[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[32m66.5/66.5 MB\u001b[0m \u001b[31m24.7 MB/s\u001b[0m eta \u001b[36m0:00:00\u001b[0m00:01\u001b[0m00:01\u001b[0m\n",
107
- "\u001b[?25hUsing cached sympy-1.13.1-py3-none-any.whl (6.2 MB)\n",
108
- "Downloading transformers-4.48.3-py3-none-any.whl (9.7 MB)\n",
109
- "\u001b[2K \u001b[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[32m9.7/9.7 MB\u001b[0m \u001b[31m35.3 MB/s\u001b[0m eta \u001b[36m0:00:00\u001b[0m00:01\u001b[0m00:01\u001b[0m\n",
110
- "\u001b[?25hUsing cached networkx-3.4.2-py3-none-any.whl (1.7 MB)\n",
111
- "Using cached mpmath-1.3.0-py3-none-any.whl (536 kB)\n",
112
- "Using cached regex-2024.11.6-cp312-cp312-macosx_11_0_arm64.whl (284 kB)\n",
113
- "Downloading safetensors-0.5.2-cp38-abi3-macosx_11_0_arm64.whl (408 kB)\n",
114
- "\u001b[2K \u001b[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[32m408.9/408.9 kB\u001b[0m \u001b[31m27.4 MB/s\u001b[0m eta \u001b[36m0:00:00\u001b[0m\n",
115
- "\u001b[?25hDownloading tokenizers-0.21.0-cp39-abi3-macosx_11_0_arm64.whl (2.6 MB)\n",
116
- "\u001b[2K \u001b[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[32m2.6/2.6 MB\u001b[0m \u001b[31m36.1 MB/s\u001b[0m eta \u001b[36m0:00:00\u001b[0ma \u001b[36m0:00:01\u001b[0m\n",
117
- "\u001b[?25hInstalling collected packages: mpmath, sympy, safetensors, regex, networkx, torch, tokenizers, transformers, sentence-transformers\n",
118
- "Successfully installed mpmath-1.3.0 networkx-3.4.2 regex-2024.11.6 safetensors-0.5.2 sentence-transformers-3.4.1 sympy-1.13.1 tokenizers-0.21.0 torch-2.6.0 transformers-4.48.3\n",
119
- "\n",
120
- "\u001b[1m[\u001b[0m\u001b[34;49mnotice\u001b[0m\u001b[1;39;49m]\u001b[0m\u001b[39;49m A new release of pip is available: \u001b[0m\u001b[31;49m24.0\u001b[0m\u001b[39;49m -> \u001b[0m\u001b[32;49m25.0.1\u001b[0m\n",
121
- "\u001b[1m[\u001b[0m\u001b[34;49mnotice\u001b[0m\u001b[1;39;49m]\u001b[0m\u001b[39;49m To update, run: \u001b[0m\u001b[32;49mpip3 install --upgrade pip\u001b[0m\n",
122
- "Note: you may need to restart the kernel to use updated packages.\n"
123
- ]
124
- }
125
- ],
126
- "source": [
127
- "pip install sentence-transformers scikit-learn pandas torch\n"
128
- ]
129
- },
130
- {
131
- "cell_type": "code",
132
- "execution_count": 4,
133
- "metadata": {
134
- "id": "m-tmgXuldd3C"
135
- },
136
- "outputs": [],
137
- "source": [
138
- "import seaborn as sns"
139
- ]
140
- },
141
- {
142
- "cell_type": "code",
143
- "execution_count": 5,
144
- "metadata": {
145
- "id": "1Z0mgYZEgjC4"
146
- },
147
- "outputs": [],
148
- "source": [
149
- "from sklearn.ensemble import RandomForestClassifier"
150
- ]
151
- },
152
- {
153
- "cell_type": "code",
154
- "execution_count": 6,
155
- "metadata": {
156
- "id": "aBmXLbZ4cc1U"
157
- },
158
- "outputs": [],
159
- "source": [
160
- "from sklearn.model_selection import train_test_split\n"
161
- ]
162
- },
163
- {
164
- "cell_type": "code",
165
- "execution_count": 7,
166
- "metadata": {
167
- "id": "LXkXdIWgUcWI"
168
- },
169
- "outputs": [],
170
- "source": [
171
- "from datasets import load_dataset, Dataset\n",
172
- "import pandas as pd"
173
- ]
174
- },
175
- {
176
- "cell_type": "code",
177
- "execution_count": 8,
178
- "metadata": {
179
- "id": "AFkI23ySgtkV"
180
- },
181
- "outputs": [],
182
- "source": [
183
- "from sklearn.metrics import accuracy_score"
184
- ]
185
- },
186
- {
187
- "cell_type": "code",
188
- "execution_count": 9,
189
- "metadata": {
190
- "colab": {
191
- "base_uri": "https://localhost:8080/"
192
- },
193
- "id": "ehvh1BJZWa_1",
194
- "outputId": "212a5f82-885d-4e61-a73f-94dcf12a3a39"
195
- },
196
- "outputs": [
197
- {
198
- "data": {
199
- "application/vnd.jupyter.widget-view+json": {
200
- "model_id": "ab0344420d964f64a16c911f17aae057",
201
- "version_major": 2,
202
- "version_minor": 0
203
- },
204
- "text/plain": [
205
- "README.md: 0%| | 0.00/515 [00:00<?, ?B/s]"
206
- ]
207
- },
208
- "metadata": {},
209
- "output_type": "display_data"
210
- },
211
- {
212
- "data": {
213
- "application/vnd.jupyter.widget-view+json": {
214
- "model_id": "9f86cbde053f4b9e91cff137a924082f",
215
- "version_major": 2,
216
- "version_minor": 0
217
- },
218
- "text/plain": [
219
- "train-00000-of-00001.parquet: 0%| | 0.00/5.89M [00:00<?, ?B/s]"
220
- ]
221
- },
222
- "metadata": {},
223
- "output_type": "display_data"
224
- },
225
- {
226
- "data": {
227
- "application/vnd.jupyter.widget-view+json": {
228
- "model_id": "2f6d00487331444299405cc97d4b18ea",
229
- "version_major": 2,
230
- "version_minor": 0
231
- },
232
- "text/plain": [
233
- "Generating train split: 0%| | 0/61199 [00:00<?, ? examples/s]"
234
- ]
235
- },
236
- "metadata": {},
237
- "output_type": "display_data"
238
- },
239
- {
240
- "name": "stdout",
241
- "output_type": "stream",
242
- "text": [
243
- " answer system_prompt \\\n",
244
- "0 neutral You are a financial sentiment analysis expert.... \n",
245
- "1 neutral You are a financial sentiment analysis expert.... \n",
246
- "2 negative You are a financial sentiment analysis expert.... \n",
247
- "3 positive You are a financial sentiment analysis expert.... \n",
248
- "4 positive You are a financial sentiment analysis expert.... \n",
249
- "\n",
250
- " user_prompt task_type \n",
251
- "0 According to Gran , the company has no plans t... sentiment_analysis \n",
252
- "1 Technopolis plans to develop in stages an area... sentiment_analysis \n",
253
- "2 The international electronic industry company ... sentiment_analysis \n",
254
- "3 With the new production plant the company woul... sentiment_analysis \n",
255
- "4 According to the company 's updated strategy f... sentiment_analysis \n"
256
- ]
257
- }
258
- ],
259
- "source": [
260
- "df = load_dataset(\"NickyNicky/Finance_sentiment_and_topic_classification_En\")\n",
261
- "\n",
262
- "# Converting 'train' split to a Pandas DataFrame\n",
263
- "df = pd.DataFrame(df['train'])\n",
264
- "\n",
265
- "\n",
266
- "print(df.head())\n",
267
- "\n",
268
- "\n",
269
- "df.to_csv(\"train_data.csv\", index=False)\n"
270
- ]
271
- },
272
- {
273
- "cell_type": "code",
274
- "execution_count": 10,
275
- "metadata": {
276
- "colab": {
277
- "base_uri": "https://localhost:8080/",
278
- "height": 206
279
- },
280
- "id": "wS-PmD5WWnYC",
281
- "outputId": "36732946-2bb0-4f58-f784-5619d77698b9"
282
- },
283
- "outputs": [
284
- {
285
- "data": {
286
- "text/html": [
287
- "<div>\n",
288
- "<style scoped>\n",
289
- " .dataframe tbody tr th:only-of-type {\n",
290
- " vertical-align: middle;\n",
291
- " }\n",
292
- "\n",
293
- " .dataframe tbody tr th {\n",
294
- " vertical-align: top;\n",
295
- " }\n",
296
- "\n",
297
- " .dataframe thead th {\n",
298
- " text-align: right;\n",
299
- " }\n",
300
- "</style>\n",
301
- "<table border=\"1\" class=\"dataframe\">\n",
302
- " <thead>\n",
303
- " <tr style=\"text-align: right;\">\n",
304
- " <th></th>\n",
305
- " <th>answer</th>\n",
306
- " <th>system_prompt</th>\n",
307
- " <th>user_prompt</th>\n",
308
- " <th>task_type</th>\n",
309
- " </tr>\n",
310
- " </thead>\n",
311
- " <tbody>\n",
312
- " <tr>\n",
313
- " <th>0</th>\n",
314
- " <td>neutral</td>\n",
315
- " <td>You are a financial sentiment analysis expert....</td>\n",
316
- " <td>According to Gran , the company has no plans t...</td>\n",
317
- " <td>sentiment_analysis</td>\n",
318
- " </tr>\n",
319
- " <tr>\n",
320
- " <th>1</th>\n",
321
- " <td>neutral</td>\n",
322
- " <td>You are a financial sentiment analysis expert....</td>\n",
323
- " <td>Technopolis plans to develop in stages an area...</td>\n",
324
- " <td>sentiment_analysis</td>\n",
325
- " </tr>\n",
326
- " <tr>\n",
327
- " <th>2</th>\n",
328
- " <td>negative</td>\n",
329
- " <td>You are a financial sentiment analysis expert....</td>\n",
330
- " <td>The international electronic industry company ...</td>\n",
331
- " <td>sentiment_analysis</td>\n",
332
- " </tr>\n",
333
- " <tr>\n",
334
- " <th>3</th>\n",
335
- " <td>positive</td>\n",
336
- " <td>You are a financial sentiment analysis expert....</td>\n",
337
- " <td>With the new production plant the company woul...</td>\n",
338
- " <td>sentiment_analysis</td>\n",
339
- " </tr>\n",
340
- " <tr>\n",
341
- " <th>4</th>\n",
342
- " <td>positive</td>\n",
343
- " <td>You are a financial sentiment analysis expert....</td>\n",
344
- " <td>According to the company 's updated strategy f...</td>\n",
345
- " <td>sentiment_analysis</td>\n",
346
- " </tr>\n",
347
- " </tbody>\n",
348
- "</table>\n",
349
- "</div>"
350
- ],
351
- "text/plain": [
352
- " answer system_prompt \\\n",
353
- "0 neutral You are a financial sentiment analysis expert.... \n",
354
- "1 neutral You are a financial sentiment analysis expert.... \n",
355
- "2 negative You are a financial sentiment analysis expert.... \n",
356
- "3 positive You are a financial sentiment analysis expert.... \n",
357
- "4 positive You are a financial sentiment analysis expert.... \n",
358
- "\n",
359
- " user_prompt task_type \n",
360
- "0 According to Gran , the company has no plans t... sentiment_analysis \n",
361
- "1 Technopolis plans to develop in stages an area... sentiment_analysis \n",
362
- "2 The international electronic industry company ... sentiment_analysis \n",
363
- "3 With the new production plant the company woul... sentiment_analysis \n",
364
- "4 According to the company 's updated strategy f... sentiment_analysis "
365
- ]
366
- },
367
- "execution_count": 10,
368
- "metadata": {},
369
- "output_type": "execute_result"
370
- }
371
- ],
372
- "source": [
373
- "df.head()"
374
- ]
375
- },
376
- {
377
- "cell_type": "code",
378
- "execution_count": 11,
379
- "metadata": {
380
- "id": "cQ3tjGFTW5kE"
381
- },
382
- "outputs": [],
383
- "source": [
384
- "df.drop(['system_prompt', 'task_type'], axis=1, inplace=True)"
385
- ]
386
- },
387
- {
388
- "cell_type": "code",
389
- "execution_count": 12,
390
- "metadata": {
391
- "colab": {
392
- "base_uri": "https://localhost:8080/",
393
- "height": 423
394
- },
395
- "id": "Va5867ATXAxD",
396
- "outputId": "c258f546-d8af-4ba5-8228-3a07f2283baf"
397
- },
398
- "outputs": [
399
- {
400
- "data": {
401
- "text/html": [
402
- "<div>\n",
403
- "<style scoped>\n",
404
- " .dataframe tbody tr th:only-of-type {\n",
405
- " vertical-align: middle;\n",
406
- " }\n",
407
- "\n",
408
- " .dataframe tbody tr th {\n",
409
- " vertical-align: top;\n",
410
- " }\n",
411
- "\n",
412
- " .dataframe thead th {\n",
413
- " text-align: right;\n",
414
- " }\n",
415
- "</style>\n",
416
- "<table border=\"1\" class=\"dataframe\">\n",
417
- " <thead>\n",
418
- " <tr style=\"text-align: right;\">\n",
419
- " <th></th>\n",
420
- " <th>answer</th>\n",
421
- " <th>user_prompt</th>\n",
422
- " </tr>\n",
423
- " </thead>\n",
424
- " <tbody>\n",
425
- " <tr>\n",
426
- " <th>0</th>\n",
427
- " <td>neutral</td>\n",
428
- " <td>According to Gran , the company has no plans t...</td>\n",
429
- " </tr>\n",
430
- " <tr>\n",
431
- " <th>1</th>\n",
432
- " <td>neutral</td>\n",
433
- " <td>Technopolis plans to develop in stages an area...</td>\n",
434
- " </tr>\n",
435
- " <tr>\n",
436
- " <th>2</th>\n",
437
- " <td>negative</td>\n",
438
- " <td>The international electronic industry company ...</td>\n",
439
- " </tr>\n",
440
- " <tr>\n",
441
- " <th>3</th>\n",
442
- " <td>positive</td>\n",
443
- " <td>With the new production plant the company woul...</td>\n",
444
- " </tr>\n",
445
- " <tr>\n",
446
- " <th>4</th>\n",
447
- " <td>positive</td>\n",
448
- " <td>According to the company 's updated strategy f...</td>\n",
449
- " </tr>\n",
450
- " <tr>\n",
451
- " <th>...</th>\n",
452
- " <td>...</td>\n",
453
- " <td>...</td>\n",
454
- " </tr>\n",
455
- " <tr>\n",
456
- " <th>61194</th>\n",
457
- " <td>Treasuries | Corporate Debt</td>\n",
458
- " <td>KfW credit line for Uniper could be raised to ...</td>\n",
459
- " </tr>\n",
460
- " <tr>\n",
461
- " <th>61195</th>\n",
462
- " <td>Treasuries | Corporate Debt</td>\n",
463
- " <td>KfW credit line for Uniper could be raised to ...</td>\n",
464
- " </tr>\n",
465
- " <tr>\n",
466
- " <th>61196</th>\n",
467
- " <td>Treasuries | Corporate Debt</td>\n",
468
- " <td>Russian https://t.co/R0iPhyo5p7 sells 1 bln r...</td>\n",
469
- " </tr>\n",
470
- " <tr>\n",
471
- " <th>61197</th>\n",
472
- " <td>Treasuries | Corporate Debt</td>\n",
473
- " <td>Global ESG bond issuance posts H1 dip as supra...</td>\n",
474
- " </tr>\n",
475
- " <tr>\n",
476
- " <th>61198</th>\n",
477
- " <td>Treasuries | Corporate Debt</td>\n",
478
- " <td>Brazil's Petrobras says it signed a $1.25 bill...</td>\n",
479
- " </tr>\n",
480
- " </tbody>\n",
481
- "</table>\n",
482
- "<p>61199 rows × 2 columns</p>\n",
483
- "</div>"
484
- ],
485
- "text/plain": [
486
- " answer \\\n",
487
- "0 neutral \n",
488
- "1 neutral \n",
489
- "2 negative \n",
490
- "3 positive \n",
491
- "4 positive \n",
492
- "... ... \n",
493
- "61194 Treasuries | Corporate Debt \n",
494
- "61195 Treasuries | Corporate Debt \n",
495
- "61196 Treasuries | Corporate Debt \n",
496
- "61197 Treasuries | Corporate Debt \n",
497
- "61198 Treasuries | Corporate Debt \n",
498
- "\n",
499
- " user_prompt \n",
500
- "0 According to Gran , the company has no plans t... \n",
501
- "1 Technopolis plans to develop in stages an area... \n",
502
- "2 The international electronic industry company ... \n",
503
- "3 With the new production plant the company woul... \n",
504
- "4 According to the company 's updated strategy f... \n",
505
- "... ... \n",
506
- "61194 KfW credit line for Uniper could be raised to ... \n",
507
- "61195 KfW credit line for Uniper could be raised to ... \n",
508
- "61196 Russian https://t.co/R0iPhyo5p7 sells 1 bln r... \n",
509
- "61197 Global ESG bond issuance posts H1 dip as supra... \n",
510
- "61198 Brazil's Petrobras says it signed a $1.25 bill... \n",
511
- "\n",
512
- "[61199 rows x 2 columns]"
513
- ]
514
- },
515
- "execution_count": 12,
516
- "metadata": {},
517
- "output_type": "execute_result"
518
- }
519
- ],
520
- "source": [
521
- "df"
522
- ]
523
- },
524
- {
525
- "cell_type": "code",
526
- "execution_count": 13,
527
- "metadata": {
528
- "colab": {
529
- "base_uri": "https://localhost:8080/"
530
- },
531
- "id": "a2PtcHIfeM5t",
532
- "outputId": "2214b201-c68d-4112-d224-855bd7103213"
533
- },
534
- "outputs": [
535
- {
536
- "name": "stdout",
537
- "output_type": "stream",
538
- "text": [
539
- "(39641, 2)\n"
540
- ]
541
- }
542
- ],
543
- "source": [
544
- "# only want to keep rows where 'answer' is 'neutral', 'positive', or 'negative'\n",
545
- "df_filtered = df[df[\"answer\"].isin([\"neutral\", \"positive\", \"negative\"])]\n",
546
- "\n",
547
- "# Showing the shape of the new DataFrame\n",
548
- "print(df_filtered.shape)\n"
549
- ]
550
- },
551
- {
552
- "cell_type": "code",
553
- "execution_count": 14,
554
- "metadata": {
555
- "colab": {
556
- "base_uri": "https://localhost:8080/"
557
- },
558
- "id": "OzmsDZ-tZuGA",
559
- "outputId": "cb743649-521b-45da-8c9a-182fff7584bd"
560
- },
561
- "outputs": [
562
- {
563
- "name": "stdout",
564
- "output_type": "stream",
565
- "text": [
566
- "(5946, 2)\n"
567
- ]
568
- }
569
- ],
570
- "source": [
571
- "df_sampled = df_filtered.sample(frac=0.15, random_state=42) # 15% sample\n",
572
- "print(df_sampled.shape) # Checking new size\n"
573
- ]
574
- },
575
- {
576
- "cell_type": "code",
577
- "execution_count": 15,
578
- "metadata": {
579
- "colab": {
580
- "base_uri": "https://localhost:8080/",
581
- "height": 466
582
- },
583
- "id": "kkaxAfyQdcgZ",
584
- "outputId": "1fa73eb8-c4ea-4b0b-8288-7dcb3c517798"
585
- },
586
- "outputs": [
587
- {
588
- "data": {
589
- "text/plain": [
590
- "<Axes: xlabel='answer', ylabel='count'>"
591
- ]
592
- },
593
- "execution_count": 15,
594
- "metadata": {},
595
- "output_type": "execute_result"
596
- },
597
- {
598
- "data": {
599
- "image/png": "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",
600
- "text/plain": [
601
- "<Figure size 640x480 with 1 Axes>"
602
- ]
603
- },
604
- "metadata": {},
605
- "output_type": "display_data"
606
- }
607
- ],
608
- "source": [
609
- "sns.countplot(x=df_sampled[\"answer\"])"
610
- ]
611
- },
612
- {
613
- "cell_type": "code",
614
- "execution_count": 16,
615
- "metadata": {
616
- "colab": {
617
- "base_uri": "https://localhost:8080/"
618
- },
619
- "id": "e2DP6ekqfNbe",
620
- "outputId": "3aacbc50-8554-40eb-9cdb-c949c30d634e"
621
- },
622
- "outputs": [
623
- {
624
- "name": "stdout",
625
- "output_type": "stream",
626
- "text": [
627
- "answer\n",
628
- "negative 1236\n",
629
- "neutral 1236\n",
630
- "positive 1236\n",
631
- "Name: count, dtype: int64\n",
632
- "(3708, 2)\n"
633
- ]
634
- },
635
- {
636
- "name": "stderr",
637
- "output_type": "stream",
638
- "text": [
639
- "/var/folders/xc/v1l81vkx6fjc9wpqc0tsnl400000gn/T/ipykernel_11468/1830774783.py:5: DeprecationWarning: DataFrameGroupBy.apply operated on the grouping columns. This behavior is deprecated, and in a future version of pandas the grouping columns will be excluded from the operation. Either pass `include_groups=False` to exclude the groupings or explicitly select the grouping columns after groupby to silence this warning.\n",
640
- " df_balanced = df_sampled.groupby(\"answer\").apply(lambda x: x.sample(min_class_count, random_state=42)).reset_index(drop=True)\n"
641
- ]
642
- }
643
- ],
644
- "source": [
645
- "# Undersampling each class to match the class with the smallest number of samples\n",
646
- "min_class_count = df_sampled[\"answer\"].value_counts().min()\n",
647
- "\n",
648
- "# Sampling an equal number of rows from each class\n",
649
- "df_balanced = df_sampled.groupby(\"answer\").apply(lambda x: x.sample(min_class_count, random_state=42)).reset_index(drop=True)\n",
650
- "\n",
651
- "# Showing the new class distribution\n",
652
- "print(df_balanced[\"answer\"].value_counts())\n",
653
- "print(df_balanced.shape)\n"
654
- ]
655
- },
656
- {
657
- "cell_type": "code",
658
- "execution_count": 17,
659
- "metadata": {
660
- "id": "dJosNJACYDCc"
661
- },
662
- "outputs": [
663
- {
664
- "data": {
665
- "application/vnd.jupyter.widget-view+json": {
666
- "model_id": "593c0e8a5f6b4b9495ff422cc2382975",
667
- "version_major": 2,
668
- "version_minor": 0
669
- },
670
- "text/plain": [
671
- "modules.json: 0%| | 0.00/349 [00:00<?, ?B/s]"
672
- ]
673
- },
674
- "metadata": {},
675
- "output_type": "display_data"
676
- },
677
- {
678
- "data": {
679
- "application/vnd.jupyter.widget-view+json": {
680
- "model_id": "850ce1db88c64b81802f2a60f45801d4",
681
- "version_major": 2,
682
- "version_minor": 0
683
- },
684
- "text/plain": [
685
- "config_sentence_transformers.json: 0%| | 0.00/116 [00:00<?, ?B/s]"
686
- ]
687
- },
688
- "metadata": {},
689
- "output_type": "display_data"
690
- },
691
- {
692
- "data": {
693
- "application/vnd.jupyter.widget-view+json": {
694
- "model_id": "8cfd7d7217a24485818919eebbca3cb2",
695
- "version_major": 2,
696
- "version_minor": 0
697
- },
698
- "text/plain": [
699
- "README.md: 0%| | 0.00/10.7k [00:00<?, ?B/s]"
700
- ]
701
- },
702
- "metadata": {},
703
- "output_type": "display_data"
704
- },
705
- {
706
- "data": {
707
- "application/vnd.jupyter.widget-view+json": {
708
- "model_id": "6aa8a1f649ec471ebe07ee374e80de62",
709
- "version_major": 2,
710
- "version_minor": 0
711
- },
712
- "text/plain": [
713
- "sentence_bert_config.json: 0%| | 0.00/53.0 [00:00<?, ?B/s]"
714
- ]
715
- },
716
- "metadata": {},
717
- "output_type": "display_data"
718
- },
719
- {
720
- "data": {
721
- "application/vnd.jupyter.widget-view+json": {
722
- "model_id": "cb49e7e953654af8af5fd5d22f78ce59",
723
- "version_major": 2,
724
- "version_minor": 0
725
- },
726
- "text/plain": [
727
- "config.json: 0%| | 0.00/612 [00:00<?, ?B/s]"
728
- ]
729
- },
730
- "metadata": {},
731
- "output_type": "display_data"
732
- },
733
- {
734
- "data": {
735
- "application/vnd.jupyter.widget-view+json": {
736
- "model_id": "12bb5d1510b1422582f091a49fa617a0",
737
- "version_major": 2,
738
- "version_minor": 0
739
- },
740
- "text/plain": [
741
- "model.safetensors: 0%| | 0.00/90.9M [00:00<?, ?B/s]"
742
- ]
743
- },
744
- "metadata": {},
745
- "output_type": "display_data"
746
- },
747
- {
748
- "data": {
749
- "application/vnd.jupyter.widget-view+json": {
750
- "model_id": "b9632a0063044733bd70e443fce6caed",
751
- "version_major": 2,
752
- "version_minor": 0
753
- },
754
- "text/plain": [
755
- "tokenizer_config.json: 0%| | 0.00/350 [00:00<?, ?B/s]"
756
- ]
757
- },
758
- "metadata": {},
759
- "output_type": "display_data"
760
- },
761
- {
762
- "data": {
763
- "application/vnd.jupyter.widget-view+json": {
764
- "model_id": "5d345282cdd6400d8bc229280df8766e",
765
- "version_major": 2,
766
- "version_minor": 0
767
- },
768
- "text/plain": [
769
- "vocab.txt: 0%| | 0.00/232k [00:00<?, ?B/s]"
770
- ]
771
- },
772
- "metadata": {},
773
- "output_type": "display_data"
774
- },
775
- {
776
- "data": {
777
- "application/vnd.jupyter.widget-view+json": {
778
- "model_id": "0473d385d77c406a94075d701d4565e1",
779
- "version_major": 2,
780
- "version_minor": 0
781
- },
782
- "text/plain": [
783
- "tokenizer.json: 0%| | 0.00/466k [00:00<?, ?B/s]"
784
- ]
785
- },
786
- "metadata": {},
787
- "output_type": "display_data"
788
- },
789
- {
790
- "data": {
791
- "application/vnd.jupyter.widget-view+json": {
792
- "model_id": "643a5bf87437468e8a168e55687814d5",
793
- "version_major": 2,
794
- "version_minor": 0
795
- },
796
- "text/plain": [
797
- "special_tokens_map.json: 0%| | 0.00/112 [00:00<?, ?B/s]"
798
- ]
799
- },
800
- "metadata": {},
801
- "output_type": "display_data"
802
- },
803
- {
804
- "data": {
805
- "application/vnd.jupyter.widget-view+json": {
806
- "model_id": "4560c43f64884ba9b236962626a1f784",
807
- "version_major": 2,
808
- "version_minor": 0
809
- },
810
- "text/plain": [
811
- "1_Pooling%2Fconfig.json: 0%| | 0.00/190 [00:00<?, ?B/s]"
812
- ]
813
- },
814
- "metadata": {},
815
- "output_type": "display_data"
816
- }
817
- ],
818
- "source": [
819
- "# Load model\n",
820
- "from sentence_transformers import SentenceTransformer # import the SentenceTransformer class\n",
821
- "model = SentenceTransformer(\"all-MiniLM-L6-v2\")\n",
822
- "\n",
823
- "# Converting text to embeddings\n",
824
- "X = model.encode(df_balanced[\"user_prompt\"].tolist(), convert_to_numpy=True)"
825
- ]
826
- },
827
- {
828
- "cell_type": "code",
829
- "execution_count": 18,
830
- "metadata": {
831
- "colab": {
832
- "base_uri": "https://localhost:8080/"
833
- },
834
- "id": "mH4_pI6YZa3E",
835
- "outputId": "667d2b0f-a60a-4afd-e2aa-270ef9d5b8de"
836
- },
837
- "outputs": [
838
- {
839
- "name": "stdout",
840
- "output_type": "stream",
841
- "text": [
842
- "Label Mapping: {'negative': 0, 'neutral': 1, 'positive': 2}\n"
843
- ]
844
- }
845
- ],
846
- "source": [
847
- "from sklearn.preprocessing import LabelEncoder\n",
848
- "\n",
849
- "# Encode labels\n",
850
- "label_encoder = LabelEncoder()\n",
851
- "y = label_encoder.fit_transform(df_balanced[\"answer\"])\n",
852
- "\n",
853
- "# Saving the mapping\n",
854
- "label_mapping = dict(zip(label_encoder.classes_, label_encoder.transform(label_encoder.classes_)))\n",
855
- "print(\"Label Mapping:\", label_mapping)\n"
856
- ]
857
- },
858
- {
859
- "cell_type": "code",
860
- "execution_count": 19,
861
- "metadata": {
862
- "id": "I5gpynBmZe1h"
863
- },
864
- "outputs": [],
865
- "source": [
866
- "X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=0.2, random_state=42)\n"
867
- ]
868
- },
869
- {
870
- "cell_type": "code",
871
- "execution_count": 20,
872
- "metadata": {
873
- "colab": {
874
- "base_uri": "https://localhost:8080/",
875
- "height": 80
876
- },
877
- "id": "jCUXUqgRgdeJ",
878
- "outputId": "8e27311c-e799-4a89-f319-f7b6f4e07e75"
879
- },
880
- "outputs": [
881
- {
882
- "data": {
883
- "text/html": [
884
- "<style>#sk-container-id-1 {\n",
885
- " /* Definition of color scheme common for light and dark mode */\n",
886
- " --sklearn-color-text: black;\n",
887
- " --sklearn-color-line: gray;\n",
888
- " /* Definition of color scheme for unfitted estimators */\n",
889
- " --sklearn-color-unfitted-level-0: #fff5e6;\n",
890
- " --sklearn-color-unfitted-level-1: #f6e4d2;\n",
891
- " --sklearn-color-unfitted-level-2: #ffe0b3;\n",
892
- " --sklearn-color-unfitted-level-3: chocolate;\n",
893
- " /* Definition of color scheme for fitted estimators */\n",
894
- " --sklearn-color-fitted-level-0: #f0f8ff;\n",
895
- " --sklearn-color-fitted-level-1: #d4ebff;\n",
896
- " --sklearn-color-fitted-level-2: #b3dbfd;\n",
897
- " --sklearn-color-fitted-level-3: cornflowerblue;\n",
898
- "\n",
899
- " /* Specific color for light theme */\n",
900
- " --sklearn-color-text-on-default-background: var(--sg-text-color, var(--theme-code-foreground, var(--jp-content-font-color1, black)));\n",
901
- " --sklearn-color-background: var(--sg-background-color, var(--theme-background, var(--jp-layout-color0, white)));\n",
902
- " --sklearn-color-border-box: var(--sg-text-color, var(--theme-code-foreground, var(--jp-content-font-color1, black)));\n",
903
- " --sklearn-color-icon: #696969;\n",
904
- "\n",
905
- " @media (prefers-color-scheme: dark) {\n",
906
- " /* Redefinition of color scheme for dark theme */\n",
907
- " --sklearn-color-text-on-default-background: var(--sg-text-color, var(--theme-code-foreground, var(--jp-content-font-color1, white)));\n",
908
- " --sklearn-color-background: var(--sg-background-color, var(--theme-background, var(--jp-layout-color0, #111)));\n",
909
- " --sklearn-color-border-box: var(--sg-text-color, var(--theme-code-foreground, var(--jp-content-font-color1, white)));\n",
910
- " --sklearn-color-icon: #878787;\n",
911
- " }\n",
912
- "}\n",
913
- "\n",
914
- "#sk-container-id-1 {\n",
915
- " color: var(--sklearn-color-text);\n",
916
- "}\n",
917
- "\n",
918
- "#sk-container-id-1 pre {\n",
919
- " padding: 0;\n",
920
- "}\n",
921
- "\n",
922
- "#sk-container-id-1 input.sk-hidden--visually {\n",
923
- " border: 0;\n",
924
- " clip: rect(1px 1px 1px 1px);\n",
925
- " clip: rect(1px, 1px, 1px, 1px);\n",
926
- " height: 1px;\n",
927
- " margin: -1px;\n",
928
- " overflow: hidden;\n",
929
- " padding: 0;\n",
930
- " position: absolute;\n",
931
- " width: 1px;\n",
932
- "}\n",
933
- "\n",
934
- "#sk-container-id-1 div.sk-dashed-wrapped {\n",
935
- " border: 1px dashed var(--sklearn-color-line);\n",
936
- " margin: 0 0.4em 0.5em 0.4em;\n",
937
- " box-sizing: border-box;\n",
938
- " padding-bottom: 0.4em;\n",
939
- " background-color: var(--sklearn-color-background);\n",
940
- "}\n",
941
- "\n",
942
- "#sk-container-id-1 div.sk-container {\n",
943
- " /* jupyter's `normalize.less` sets `[hidden] { display: none; }`\n",
944
- " but bootstrap.min.css set `[hidden] { display: none !important; }`\n",
945
- " so we also need the `!important` here to be able to override the\n",
946
- " default hidden behavior on the sphinx rendered scikit-learn.org.\n",
947
- " See: https://github.com/scikit-learn/scikit-learn/issues/21755 */\n",
948
- " display: inline-block !important;\n",
949
- " position: relative;\n",
950
- "}\n",
951
- "\n",
952
- "#sk-container-id-1 div.sk-text-repr-fallback {\n",
953
- " display: none;\n",
954
- "}\n",
955
- "\n",
956
- "div.sk-parallel-item,\n",
957
- "div.sk-serial,\n",
958
- "div.sk-item {\n",
959
- " /* draw centered vertical line to link estimators */\n",
960
- " background-image: linear-gradient(var(--sklearn-color-text-on-default-background), var(--sklearn-color-text-on-default-background));\n",
961
- " background-size: 2px 100%;\n",
962
- " background-repeat: no-repeat;\n",
963
- " background-position: center center;\n",
964
- "}\n",
965
- "\n",
966
- "/* Parallel-specific style estimator block */\n",
967
- "\n",
968
- "#sk-container-id-1 div.sk-parallel-item::after {\n",
969
- " content: \"\";\n",
970
- " width: 100%;\n",
971
- " border-bottom: 2px solid var(--sklearn-color-text-on-default-background);\n",
972
- " flex-grow: 1;\n",
973
- "}\n",
974
- "\n",
975
- "#sk-container-id-1 div.sk-parallel {\n",
976
- " display: flex;\n",
977
- " align-items: stretch;\n",
978
- " justify-content: center;\n",
979
- " background-color: var(--sklearn-color-background);\n",
980
- " position: relative;\n",
981
- "}\n",
982
- "\n",
983
- "#sk-container-id-1 div.sk-parallel-item {\n",
984
- " display: flex;\n",
985
- " flex-direction: column;\n",
986
- "}\n",
987
- "\n",
988
- "#sk-container-id-1 div.sk-parallel-item:first-child::after {\n",
989
- " align-self: flex-end;\n",
990
- " width: 50%;\n",
991
- "}\n",
992
- "\n",
993
- "#sk-container-id-1 div.sk-parallel-item:last-child::after {\n",
994
- " align-self: flex-start;\n",
995
- " width: 50%;\n",
996
- "}\n",
997
- "\n",
998
- "#sk-container-id-1 div.sk-parallel-item:only-child::after {\n",
999
- " width: 0;\n",
1000
- "}\n",
1001
- "\n",
1002
- "/* Serial-specific style estimator block */\n",
1003
- "\n",
1004
- "#sk-container-id-1 div.sk-serial {\n",
1005
- " display: flex;\n",
1006
- " flex-direction: column;\n",
1007
- " align-items: center;\n",
1008
- " background-color: var(--sklearn-color-background);\n",
1009
- " padding-right: 1em;\n",
1010
- " padding-left: 1em;\n",
1011
- "}\n",
1012
- "\n",
1013
- "\n",
1014
- "/* Toggleable style: style used for estimator/Pipeline/ColumnTransformer box that is\n",
1015
- "clickable and can be expanded/collapsed.\n",
1016
- "- Pipeline and ColumnTransformer use this feature and define the default style\n",
1017
- "- Estimators will overwrite some part of the style using the `sk-estimator` class\n",
1018
- "*/\n",
1019
- "\n",
1020
- "/* Pipeline and ColumnTransformer style (default) */\n",
1021
- "\n",
1022
- "#sk-container-id-1 div.sk-toggleable {\n",
1023
- " /* Default theme specific background. It is overwritten whether we have a\n",
1024
- " specific estimator or a Pipeline/ColumnTransformer */\n",
1025
- " background-color: var(--sklearn-color-background);\n",
1026
- "}\n",
1027
- "\n",
1028
- "/* Toggleable label */\n",
1029
- "#sk-container-id-1 label.sk-toggleable__label {\n",
1030
- " cursor: pointer;\n",
1031
- " display: block;\n",
1032
- " width: 100%;\n",
1033
- " margin-bottom: 0;\n",
1034
- " padding: 0.5em;\n",
1035
- " box-sizing: border-box;\n",
1036
- " text-align: center;\n",
1037
- "}\n",
1038
- "\n",
1039
- "#sk-container-id-1 label.sk-toggleable__label-arrow:before {\n",
1040
- " /* Arrow on the left of the label */\n",
1041
- " content: \"▸\";\n",
1042
- " float: left;\n",
1043
- " margin-right: 0.25em;\n",
1044
- " color: var(--sklearn-color-icon);\n",
1045
- "}\n",
1046
- "\n",
1047
- "#sk-container-id-1 label.sk-toggleable__label-arrow:hover:before {\n",
1048
- " color: var(--sklearn-color-text);\n",
1049
- "}\n",
1050
- "\n",
1051
- "/* Toggleable content - dropdown */\n",
1052
- "\n",
1053
- "#sk-container-id-1 div.sk-toggleable__content {\n",
1054
- " max-height: 0;\n",
1055
- " max-width: 0;\n",
1056
- " overflow: hidden;\n",
1057
- " text-align: left;\n",
1058
- " /* unfitted */\n",
1059
- " background-color: var(--sklearn-color-unfitted-level-0);\n",
1060
- "}\n",
1061
- "\n",
1062
- "#sk-container-id-1 div.sk-toggleable__content.fitted {\n",
1063
- " /* fitted */\n",
1064
- " background-color: var(--sklearn-color-fitted-level-0);\n",
1065
- "}\n",
1066
- "\n",
1067
- "#sk-container-id-1 div.sk-toggleable__content pre {\n",
1068
- " margin: 0.2em;\n",
1069
- " border-radius: 0.25em;\n",
1070
- " color: var(--sklearn-color-text);\n",
1071
- " /* unfitted */\n",
1072
- " background-color: var(--sklearn-color-unfitted-level-0);\n",
1073
- "}\n",
1074
- "\n",
1075
- "#sk-container-id-1 div.sk-toggleable__content.fitted pre {\n",
1076
- " /* unfitted */\n",
1077
- " background-color: var(--sklearn-color-fitted-level-0);\n",
1078
- "}\n",
1079
- "\n",
1080
- "#sk-container-id-1 input.sk-toggleable__control:checked~div.sk-toggleable__content {\n",
1081
- " /* Expand drop-down */\n",
1082
- " max-height: 200px;\n",
1083
- " max-width: 100%;\n",
1084
- " overflow: auto;\n",
1085
- "}\n",
1086
- "\n",
1087
- "#sk-container-id-1 input.sk-toggleable__control:checked~label.sk-toggleable__label-arrow:before {\n",
1088
- " content: \"▾\";\n",
1089
- "}\n",
1090
- "\n",
1091
- "/* Pipeline/ColumnTransformer-specific style */\n",
1092
- "\n",
1093
- "#sk-container-id-1 div.sk-label input.sk-toggleable__control:checked~label.sk-toggleable__label {\n",
1094
- " color: var(--sklearn-color-text);\n",
1095
- " background-color: var(--sklearn-color-unfitted-level-2);\n",
1096
- "}\n",
1097
- "\n",
1098
- "#sk-container-id-1 div.sk-label.fitted input.sk-toggleable__control:checked~label.sk-toggleable__label {\n",
1099
- " background-color: var(--sklearn-color-fitted-level-2);\n",
1100
- "}\n",
1101
- "\n",
1102
- "/* Estimator-specific style */\n",
1103
- "\n",
1104
- "/* Colorize estimator box */\n",
1105
- "#sk-container-id-1 div.sk-estimator input.sk-toggleable__control:checked~label.sk-toggleable__label {\n",
1106
- " /* unfitted */\n",
1107
- " background-color: var(--sklearn-color-unfitted-level-2);\n",
1108
- "}\n",
1109
- "\n",
1110
- "#sk-container-id-1 div.sk-estimator.fitted input.sk-toggleable__control:checked~label.sk-toggleable__label {\n",
1111
- " /* fitted */\n",
1112
- " background-color: var(--sklearn-color-fitted-level-2);\n",
1113
- "}\n",
1114
- "\n",
1115
- "#sk-container-id-1 div.sk-label label.sk-toggleable__label,\n",
1116
- "#sk-container-id-1 div.sk-label label {\n",
1117
- " /* The background is the default theme color */\n",
1118
- " color: var(--sklearn-color-text-on-default-background);\n",
1119
- "}\n",
1120
- "\n",
1121
- "/* On hover, darken the color of the background */\n",
1122
- "#sk-container-id-1 div.sk-label:hover label.sk-toggleable__label {\n",
1123
- " color: var(--sklearn-color-text);\n",
1124
- " background-color: var(--sklearn-color-unfitted-level-2);\n",
1125
- "}\n",
1126
- "\n",
1127
- "/* Label box, darken color on hover, fitted */\n",
1128
- "#sk-container-id-1 div.sk-label.fitted:hover label.sk-toggleable__label.fitted {\n",
1129
- " color: var(--sklearn-color-text);\n",
1130
- " background-color: var(--sklearn-color-fitted-level-2);\n",
1131
- "}\n",
1132
- "\n",
1133
- "/* Estimator label */\n",
1134
- "\n",
1135
- "#sk-container-id-1 div.sk-label label {\n",
1136
- " font-family: monospace;\n",
1137
- " font-weight: bold;\n",
1138
- " display: inline-block;\n",
1139
- " line-height: 1.2em;\n",
1140
- "}\n",
1141
- "\n",
1142
- "#sk-container-id-1 div.sk-label-container {\n",
1143
- " text-align: center;\n",
1144
- "}\n",
1145
- "\n",
1146
- "/* Estimator-specific */\n",
1147
- "#sk-container-id-1 div.sk-estimator {\n",
1148
- " font-family: monospace;\n",
1149
- " border: 1px dotted var(--sklearn-color-border-box);\n",
1150
- " border-radius: 0.25em;\n",
1151
- " box-sizing: border-box;\n",
1152
- " margin-bottom: 0.5em;\n",
1153
- " /* unfitted */\n",
1154
- " background-color: var(--sklearn-color-unfitted-level-0);\n",
1155
- "}\n",
1156
- "\n",
1157
- "#sk-container-id-1 div.sk-estimator.fitted {\n",
1158
- " /* fitted */\n",
1159
- " background-color: var(--sklearn-color-fitted-level-0);\n",
1160
- "}\n",
1161
- "\n",
1162
- "/* on hover */\n",
1163
- "#sk-container-id-1 div.sk-estimator:hover {\n",
1164
- " /* unfitted */\n",
1165
- " background-color: var(--sklearn-color-unfitted-level-2);\n",
1166
- "}\n",
1167
- "\n",
1168
- "#sk-container-id-1 div.sk-estimator.fitted:hover {\n",
1169
- " /* fitted */\n",
1170
- " background-color: var(--sklearn-color-fitted-level-2);\n",
1171
- "}\n",
1172
- "\n",
1173
- "/* Specification for estimator info (e.g. \"i\" and \"?\") */\n",
1174
- "\n",
1175
- "/* Common style for \"i\" and \"?\" */\n",
1176
- "\n",
1177
- ".sk-estimator-doc-link,\n",
1178
- "a:link.sk-estimator-doc-link,\n",
1179
- "a:visited.sk-estimator-doc-link {\n",
1180
- " float: right;\n",
1181
- " font-size: smaller;\n",
1182
- " line-height: 1em;\n",
1183
- " font-family: monospace;\n",
1184
- " background-color: var(--sklearn-color-background);\n",
1185
- " border-radius: 1em;\n",
1186
- " height: 1em;\n",
1187
- " width: 1em;\n",
1188
- " text-decoration: none !important;\n",
1189
- " margin-left: 1ex;\n",
1190
- " /* unfitted */\n",
1191
- " border: var(--sklearn-color-unfitted-level-1) 1pt solid;\n",
1192
- " color: var(--sklearn-color-unfitted-level-1);\n",
1193
- "}\n",
1194
- "\n",
1195
- ".sk-estimator-doc-link.fitted,\n",
1196
- "a:link.sk-estimator-doc-link.fitted,\n",
1197
- "a:visited.sk-estimator-doc-link.fitted {\n",
1198
- " /* fitted */\n",
1199
- " border: var(--sklearn-color-fitted-level-1) 1pt solid;\n",
1200
- " color: var(--sklearn-color-fitted-level-1);\n",
1201
- "}\n",
1202
- "\n",
1203
- "/* On hover */\n",
1204
- "div.sk-estimator:hover .sk-estimator-doc-link:hover,\n",
1205
- ".sk-estimator-doc-link:hover,\n",
1206
- "div.sk-label-container:hover .sk-estimator-doc-link:hover,\n",
1207
- ".sk-estimator-doc-link:hover {\n",
1208
- " /* unfitted */\n",
1209
- " background-color: var(--sklearn-color-unfitted-level-3);\n",
1210
- " color: var(--sklearn-color-background);\n",
1211
- " text-decoration: none;\n",
1212
- "}\n",
1213
- "\n",
1214
- "div.sk-estimator.fitted:hover .sk-estimator-doc-link.fitted:hover,\n",
1215
- ".sk-estimator-doc-link.fitted:hover,\n",
1216
- "div.sk-label-container:hover .sk-estimator-doc-link.fitted:hover,\n",
1217
- ".sk-estimator-doc-link.fitted:hover {\n",
1218
- " /* fitted */\n",
1219
- " background-color: var(--sklearn-color-fitted-level-3);\n",
1220
- " color: var(--sklearn-color-background);\n",
1221
- " text-decoration: none;\n",
1222
- "}\n",
1223
- "\n",
1224
- "/* Span, style for the box shown on hovering the info icon */\n",
1225
- ".sk-estimator-doc-link span {\n",
1226
- " display: none;\n",
1227
- " z-index: 9999;\n",
1228
- " position: relative;\n",
1229
- " font-weight: normal;\n",
1230
- " right: .2ex;\n",
1231
- " padding: .5ex;\n",
1232
- " margin: .5ex;\n",
1233
- " width: min-content;\n",
1234
- " min-width: 20ex;\n",
1235
- " max-width: 50ex;\n",
1236
- " color: var(--sklearn-color-text);\n",
1237
- " box-shadow: 2pt 2pt 4pt #999;\n",
1238
- " /* unfitted */\n",
1239
- " background: var(--sklearn-color-unfitted-level-0);\n",
1240
- " border: .5pt solid var(--sklearn-color-unfitted-level-3);\n",
1241
- "}\n",
1242
- "\n",
1243
- ".sk-estimator-doc-link.fitted span {\n",
1244
- " /* fitted */\n",
1245
- " background: var(--sklearn-color-fitted-level-0);\n",
1246
- " border: var(--sklearn-color-fitted-level-3);\n",
1247
- "}\n",
1248
- "\n",
1249
- ".sk-estimator-doc-link:hover span {\n",
1250
- " display: block;\n",
1251
- "}\n",
1252
- "\n",
1253
- "/* \"?\"-specific style due to the `<a>` HTML tag */\n",
1254
- "\n",
1255
- "#sk-container-id-1 a.estimator_doc_link {\n",
1256
- " float: right;\n",
1257
- " font-size: 1rem;\n",
1258
- " line-height: 1em;\n",
1259
- " font-family: monospace;\n",
1260
- " background-color: var(--sklearn-color-background);\n",
1261
- " border-radius: 1rem;\n",
1262
- " height: 1rem;\n",
1263
- " width: 1rem;\n",
1264
- " text-decoration: none;\n",
1265
- " /* unfitted */\n",
1266
- " color: var(--sklearn-color-unfitted-level-1);\n",
1267
- " border: var(--sklearn-color-unfitted-level-1) 1pt solid;\n",
1268
- "}\n",
1269
- "\n",
1270
- "#sk-container-id-1 a.estimator_doc_link.fitted {\n",
1271
- " /* fitted */\n",
1272
- " border: var(--sklearn-color-fitted-level-1) 1pt solid;\n",
1273
- " color: var(--sklearn-color-fitted-level-1);\n",
1274
- "}\n",
1275
- "\n",
1276
- "/* On hover */\n",
1277
- "#sk-container-id-1 a.estimator_doc_link:hover {\n",
1278
- " /* unfitted */\n",
1279
- " background-color: var(--sklearn-color-unfitted-level-3);\n",
1280
- " color: var(--sklearn-color-background);\n",
1281
- " text-decoration: none;\n",
1282
- "}\n",
1283
- "\n",
1284
- "#sk-container-id-1 a.estimator_doc_link.fitted:hover {\n",
1285
- " /* fitted */\n",
1286
- " background-color: var(--sklearn-color-fitted-level-3);\n",
1287
- "}\n",
1288
- "</style><div id=\"sk-container-id-1\" class=\"sk-top-container\"><div class=\"sk-text-repr-fallback\"><pre>RandomForestClassifier(random_state=42)</pre><b>In a Jupyter environment, please rerun this cell to show the HTML representation or trust the notebook. <br />On GitHub, the HTML representation is unable to render, please try loading this page with nbviewer.org.</b></div><div class=\"sk-container\" hidden><div class=\"sk-item\"><div class=\"sk-estimator fitted sk-toggleable\"><input class=\"sk-toggleable__control sk-hidden--visually\" id=\"sk-estimator-id-1\" type=\"checkbox\" checked><label for=\"sk-estimator-id-1\" class=\"sk-toggleable__label fitted sk-toggleable__label-arrow fitted\">&nbsp;&nbsp;RandomForestClassifier<a class=\"sk-estimator-doc-link fitted\" rel=\"noreferrer\" target=\"_blank\" href=\"https://scikit-learn.org/1.4/modules/generated/sklearn.ensemble.RandomForestClassifier.html\">?<span>Documentation for RandomForestClassifier</span></a><span class=\"sk-estimator-doc-link fitted\">i<span>Fitted</span></span></label><div class=\"sk-toggleable__content fitted\"><pre>RandomForestClassifier(random_state=42)</pre></div> </div></div></div></div>"
1289
- ],
1290
- "text/plain": [
1291
- "RandomForestClassifier(random_state=42)"
1292
- ]
1293
- },
1294
- "execution_count": 20,
1295
- "metadata": {},
1296
- "output_type": "execute_result"
1297
- }
1298
- ],
1299
- "source": [
1300
- "clf = RandomForestClassifier(n_estimators=100, random_state=42)\n",
1301
- "clf.fit(X_train, y_train)\n"
1302
- ]
1303
- },
1304
- {
1305
- "cell_type": "code",
1306
- "execution_count": 36,
1307
- "metadata": {},
1308
- "outputs": [
1309
- {
1310
- "name": "stdout",
1311
- "output_type": "stream",
1312
- "text": [
1313
- " precision recall f1-score support\n",
1314
- "\n",
1315
- " 0 0.66 0.52 0.58 277\n",
1316
- " 1 0.62 0.80 0.70 237\n",
1317
- " 2 0.55 0.52 0.54 228\n",
1318
- "\n",
1319
- " accuracy 0.61 742\n",
1320
- " macro avg 0.61 0.61 0.61 742\n",
1321
- "weighted avg 0.61 0.61 0.60 742\n",
1322
- "\n"
1323
- ]
1324
- }
1325
- ],
1326
- "source": [
1327
- "from sentence_transformers import SentenceTransformer\n",
1328
- "from sklearn.ensemble import RandomForestClassifier\n",
1329
- "from sklearn.model_selection import train_test_split\n",
1330
- "from sklearn.preprocessing import LabelEncoder\n",
1331
- "from sklearn.metrics import classification_report\n",
1332
- "\n",
1333
- "# Load model (already done)\n",
1334
- "model = SentenceTransformer(\"all-MiniLM-L6-v2\")\n",
1335
- "\n",
1336
- "# Converting text to embeddings\n",
1337
- "X = model.encode(df_balanced[\"user_prompt\"].tolist(), convert_to_numpy=True)\n",
1338
- "\n",
1339
- "# Encode labels (already done)\n",
1340
- "label_encoder = LabelEncoder()\n",
1341
- "y = label_encoder.fit_transform(df_balanced[\"answer\"])\n",
1342
- "\n",
1343
- "# Train-test split (already done)\n",
1344
- "X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=0.2, random_state=42)\n",
1345
- "\n",
1346
- "# Initialize and train RandomForestClassifier\n",
1347
- "clf = RandomForestClassifier(n_estimators=100, random_state=42)\n",
1348
- "clf.fit(X_train, y_train)\n",
1349
- "\n",
1350
- "# Make predictions on the test set\n",
1351
- "y_pred = clf.predict(X_test)\n",
1352
- "\n",
1353
- "# Print classification report to evaluate performance\n",
1354
- "print(classification_report(y_test, y_pred))\n"
1355
- ]
1356
- },
1357
- {
1358
- "cell_type": "code",
1359
- "execution_count": 35,
1360
- "metadata": {},
1361
- "outputs": [
1362
- {
1363
- "name": "stderr",
1364
- "output_type": "stream",
1365
- "text": [
1366
- "huggingface/tokenizers: The current process just got forked, after parallelism has already been used. Disabling parallelism to avoid deadlocks...\n",
1367
- "To disable this warning, you can either:\n",
1368
- "\t- Avoid using `tokenizers` before the fork if possible\n",
1369
- "\t- Explicitly set the environment variable TOKENIZERS_PARALLELISM=(true | false)\n",
1370
- "huggingface/tokenizers: The current process just got forked, after parallelism has already been used. Disabling parallelism to avoid deadlocks...\n",
1371
- "To disable this warning, you can either:\n",
1372
- "\t- Avoid using `tokenizers` before the fork if possible\n",
1373
- "\t- Explicitly set the environment variable TOKENIZERS_PARALLELISM=(true | false)\n",
1374
- "huggingface/tokenizers: The current process just got forked, after parallelism has already been used. Disabling parallelism to avoid deadlocks...\n",
1375
- "To disable this warning, you can either:\n",
1376
- "\t- Avoid using `tokenizers` before the fork if possible\n",
1377
- "\t- Explicitly set the environment variable TOKENIZERS_PARALLELISM=(true | false)\n",
1378
- "huggingface/tokenizers: The current process just got forked, after parallelism has already been used. Disabling parallelism to avoid deadlocks...\n",
1379
- "To disable this warning, you can either:\n",
1380
- "\t- Avoid using `tokenizers` before the fork if possible\n",
1381
- "\t- Explicitly set the environment variable TOKENIZERS_PARALLELISM=(true | false)\n",
1382
- "huggingface/tokenizers: The current process just got forked, after parallelism has already been used. Disabling parallelism to avoid deadlocks...\n",
1383
- "To disable this warning, you can either:\n",
1384
- "\t- Avoid using `tokenizers` before the fork if possible\n",
1385
- "\t- Explicitly set the environment variable TOKENIZERS_PARALLELISM=(true | false)\n",
1386
- "huggingface/tokenizers: The current process just got forked, after parallelism has already been used. Disabling parallelism to avoid deadlocks...\n",
1387
- "To disable this warning, you can either:\n",
1388
- "\t- Avoid using `tokenizers` before the fork if possible\n",
1389
- "\t- Explicitly set the environment variable TOKENIZERS_PARALLELISM=(true | false)\n",
1390
- "huggingface/tokenizers: The current process just got forked, after parallelism has already been used. Disabling parallelism to avoid deadlocks...\n",
1391
- "To disable this warning, you can either:\n",
1392
- "\t- Avoid using `tokenizers` before the fork if possible\n",
1393
- "\t- Explicitly set the environment variable TOKENIZERS_PARALLELISM=(true | false)\n",
1394
- "huggingface/tokenizers: The current process just got forked, after parallelism has already been used. Disabling parallelism to avoid deadlocks...\n",
1395
- "To disable this warning, you can either:\n",
1396
- "\t- Avoid using `tokenizers` before the fork if possible\n",
1397
- "\t- Explicitly set the environment variable TOKENIZERS_PARALLELISM=(true | false)\n",
1398
- "huggingface/tokenizers: The current process just got forked, after parallelism has already been used. Disabling parallelism to avoid deadlocks...\n",
1399
- "To disable this warning, you can either:\n",
1400
- "\t- Avoid using `tokenizers` before the fork if possible\n",
1401
- "\t- Explicitly set the environment variable TOKENIZERS_PARALLELISM=(true | false)\n"
1402
- ]
1403
- },
1404
- {
1405
- "name": "stdout",
1406
- "output_type": "stream",
1407
- "text": [
1408
- "Best Parameters: {'max_depth': 20, 'min_samples_split': 5, 'n_estimators': 200}\n"
1409
- ]
1410
- }
1411
- ],
1412
- "source": [
1413
- "from sklearn.model_selection import GridSearchCV\n",
1414
- "\n",
1415
- "param_grid = {\n",
1416
- " 'n_estimators': [50, 100, 200],\n",
1417
- " 'max_depth': [10, 20, 30],\n",
1418
- " 'min_samples_split': [2, 5, 10]\n",
1419
- "}\n",
1420
- "\n",
1421
- "grid_search = GridSearchCV(estimator=clf, param_grid=param_grid, cv=3, n_jobs=-1)\n",
1422
- "grid_search.fit(X_train, y_train)\n",
1423
- "print(\"Best Parameters:\", grid_search.best_params_)\n"
1424
- ]
1425
- },
1426
- {
1427
- "cell_type": "code",
1428
- "execution_count": null,
1429
- "metadata": {},
1430
- "outputs": [],
1431
- "source": []
1432
- },
1433
- {
1434
- "cell_type": "code",
1435
- "execution_count": null,
1436
- "metadata": {
1437
- "colab": {
1438
- "base_uri": "https://localhost:8080/",
1439
- "height": 339
1440
- },
1441
- "id": "QasSqfQhnsqs",
1442
- "outputId": "ca0b33bf-d2b2-46a5-9e4f-9a68ff77abeb"
1443
- },
1444
- "outputs": [
1445
- {
1446
- "ename": "ValueError",
1447
- "evalue": "No columns in the dataset match the model's forward method signature. The following columns have been ignored: [user_prompt, answer]. Please check the dataset and model. You may need to set `remove_unused_columns=False` in `TrainingArguments`.",
1448
- "output_type": "error",
1449
- "traceback": [
1450
- "\u001b[0;31m---------------------------------------------------------------------------\u001b[0m",
1451
- "\u001b[0;31mValueError\u001b[0m Traceback (most recent call last)",
1452
- "\u001b[0;32m<ipython-input-127-6d82a26ee1d5>\u001b[0m in \u001b[0;36m<cell line: 0>\u001b[0;34m()\u001b[0m\n\u001b[1;32m 32\u001b[0m )\n\u001b[1;32m 33\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m---> 34\u001b[0;31m \u001b[0mtrainer\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mtrain\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m",
1453
- "\u001b[0;32m/usr/local/lib/python3.11/dist-packages/transformers/trainer.py\u001b[0m in \u001b[0;36mtrain\u001b[0;34m(self, resume_from_checkpoint, trial, ignore_keys_for_eval, **kwargs)\u001b[0m\n\u001b[1;32m 2169\u001b[0m \u001b[0mhf_hub_utils\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0menable_progress_bars\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 2170\u001b[0m \u001b[0;32melse\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m-> 2171\u001b[0;31m return inner_training_loop(\n\u001b[0m\u001b[1;32m 2172\u001b[0m \u001b[0margs\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0margs\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 2173\u001b[0m \u001b[0mresume_from_checkpoint\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0mresume_from_checkpoint\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
1454
- "\u001b[0;32m/usr/local/lib/python3.11/dist-packages/transformers/trainer.py\u001b[0m in \u001b[0;36m_inner_training_loop\u001b[0;34m(self, batch_size, args, resume_from_checkpoint, trial, ignore_keys_for_eval)\u001b[0m\n\u001b[1;32m 2198\u001b[0m \u001b[0mlogger\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mdebug\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34mf\"Currently training with a batch size of: {self._train_batch_size}\"\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 2199\u001b[0m \u001b[0;31m# Data loader and number of training steps\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m-> 2200\u001b[0;31m \u001b[0mtrain_dataloader\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mget_train_dataloader\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m 2201\u001b[0m \u001b[0;32mif\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mis_fsdp_xla_v2_enabled\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 2202\u001b[0m \u001b[0mtrain_dataloader\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mtpu_spmd_dataloader\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mtrain_dataloader\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
1455
- "\u001b[0;32m/usr/local/lib/python3.11/dist-packages/transformers/trainer.py\u001b[0m in \u001b[0;36mget_train_dataloader\u001b[0;34m(self)\u001b[0m\n\u001b[1;32m 998\u001b[0m \u001b[0mdata_collator\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mdata_collator\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 999\u001b[0m \u001b[0;32mif\u001b[0m \u001b[0mis_datasets_available\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m)\u001b[0m \u001b[0;32mand\u001b[0m \u001b[0misinstance\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mtrain_dataset\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mdatasets\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mDataset\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m-> 1000\u001b[0;31m \u001b[0mtrain_dataset\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0m_remove_unused_columns\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mtrain_dataset\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mdescription\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0;34m\"training\"\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m 1001\u001b[0m \u001b[0;32melse\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 1002\u001b[0m \u001b[0mdata_collator\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0m_get_collator_with_removed_columns\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mdata_collator\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mdescription\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0;34m\"training\"\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
1456
- "\u001b[0;32m/usr/local/lib/python3.11/dist-packages/transformers/trainer.py\u001b[0m in \u001b[0;36m_remove_unused_columns\u001b[0;34m(self, dataset, description)\u001b[0m\n\u001b[1;32m 924\u001b[0m \u001b[0mcolumns\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0;34m[\u001b[0m\u001b[0mk\u001b[0m \u001b[0;32mfor\u001b[0m \u001b[0mk\u001b[0m \u001b[0;32min\u001b[0m \u001b[0msignature_columns\u001b[0m \u001b[0;32mif\u001b[0m \u001b[0mk\u001b[0m \u001b[0;32min\u001b[0m \u001b[0mdataset\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mcolumn_names\u001b[0m\u001b[0;34m]\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 925\u001b[0m \u001b[0;32mif\u001b[0m \u001b[0mlen\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mcolumns\u001b[0m\u001b[0;34m)\u001b[0m \u001b[0;34m==\u001b[0m \u001b[0;36m0\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m--> 926\u001b[0;31m raise ValueError(\n\u001b[0m\u001b[1;32m 927\u001b[0m \u001b[0;34m\"No columns in the dataset match the model's forward method signature. \"\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 928\u001b[0m \u001b[0;34mf\"The following columns have been ignored: [{', '.join(ignored_columns)}]. \"\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
1457
- "\u001b[0;31mValueError\u001b[0m: No columns in the dataset match the model's forward method signature. The following columns have been ignored: [user_prompt, answer]. Please check the dataset and model. You may need to set `remove_unused_columns=False` in `TrainingArguments`."
1458
- ]
1459
- }
1460
- ],
1461
- "source": [
1462
- "from transformers import BertForSequenceClassification, Trainer, TrainingArguments\n",
1463
- "from datasets import Dataset\n",
1464
- "\n",
1465
- "\n",
1466
- "dataset = Dataset.from_pandas(df_balanced)\n",
1467
- "\n",
1468
- "\n",
1469
- "#dataset = dataset.filter(lambda e: e['answer'] is not None and len(e['answer']) > 0)\n",
1470
- "\n",
1471
- "\n",
1472
- "#dataset = dataset.map(lambda e: {'labels': label_encoder.transform([e['answer']])[0]}, batched=False) # Transform expects a list\n",
1473
- "\n",
1474
- "\n",
1475
- "#model = BertForSequenceClassification.from_pretrained('bert-base-uncased', num_labels=len(label_encoder.classes_))\n",
1476
- "\n",
1477
- "\n",
1478
- "#training_args = TrainingArguments(\n",
1479
- " output_dir='./results',\n",
1480
- " num_train_epochs=3,\n",
1481
- " per_device_train_batch_size=8,\n",
1482
- " per_device_eval_batch_size=16,\n",
1483
- " warmup_steps=500,\n",
1484
- " weight_decay=0.01,\n",
1485
- " logging_dir='./logs',\n",
1486
- ")\n",
1487
- "\n",
1488
- "trainer = Trainer(\n",
1489
- " model=model,\n",
1490
- " args=training_args,\n",
1491
- " train_dataset=dataset,\n",
1492
- " eval_dataset=dataset,\n",
1493
- ")\n",
1494
- "\n",
1495
- "trainer.train()"
1496
- ]
1497
- },
1498
- {
1499
- "cell_type": "code",
1500
- "execution_count": 41,
1501
- "metadata": {
1502
- "colab": {
1503
- "base_uri": "https://localhost:8080/"
1504
- },
1505
- "id": "v8DE8aAzg4jQ",
1506
- "outputId": "5ce78149-c53b-45f3-994f-5f6c7d21b819"
1507
- },
1508
- "outputs": [
1509
- {
1510
- "name": "stdout",
1511
- "output_type": "stream",
1512
- "text": [
1513
- "Predicted Label: neutral\n"
1514
- ]
1515
- }
1516
- ],
1517
- "source": [
1518
- "new_texts = [\"The company is doing OK\"]\n",
1519
- "new_embeddings = model.encode(new_texts, convert_to_numpy=True)\n",
1520
- "predicted_label = clf.predict(new_embeddings)\n",
1521
- "\n",
1522
- "# Convert back to original label names\n",
1523
- "decoded_label = label_encoder.inverse_transform(predicted_label)\n",
1524
- "print(\"Predicted Label:\", decoded_label[0])\n"
1525
- ]
1526
- }
1527
- ],
1528
- "metadata": {
1529
- "colab": {
1530
- "provenance": []
1531
- },
1532
- "kernelspec": {
1533
- "display_name": "Python 3",
1534
- "name": "python3"
1535
- },
1536
- "language_info": {
1537
- "codemirror_mode": {
1538
- "name": "ipython",
1539
- "version": 3
1540
- },
1541
- "file_extension": ".py",
1542
- "mimetype": "text/x-python",
1543
- "name": "python",
1544
- "nbconvert_exporter": "python",
1545
- "pygments_lexer": "ipython3",
1546
- "version": "3.12.2"
1547
- }
1548
- },
1549
- "nbformat": 4,
1550
- "nbformat_minor": 0
1551
- }