File size: 11,443 Bytes
5584c5d |
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258 259 260 261 262 263 264 265 266 267 268 269 270 271 272 273 274 275 276 277 278 279 280 281 282 283 284 285 286 287 288 289 290 291 292 293 294 295 296 297 298 299 300 301 302 303 304 305 306 307 308 309 310 311 312 313 314 315 316 317 318 319 320 321 322 323 324 325 326 327 328 329 330 |
{
"cells": [
{
"cell_type": "code",
"execution_count": 1,
"metadata": {},
"outputs": [
{
"data": {
"text/plain": [
"LlamaForSequenceClassification(\n",
" (model): LlamaModel(\n",
" (embed_tokens): Embedding(49153, 576, padding_idx=49152)\n",
" (layers): ModuleList(\n",
" (0-29): 30 x LlamaDecoderLayer(\n",
" (self_attn): LlamaSdpaAttention(\n",
" (q_proj): Linear(in_features=576, out_features=576, bias=False)\n",
" (k_proj): Linear(in_features=576, out_features=192, bias=False)\n",
" (v_proj): Linear(in_features=576, out_features=192, bias=False)\n",
" (o_proj): Linear(in_features=576, out_features=576, bias=False)\n",
" (rotary_emb): LlamaRotaryEmbedding()\n",
" )\n",
" (mlp): LlamaMLP(\n",
" (gate_proj): Linear(in_features=576, out_features=1536, bias=False)\n",
" (up_proj): Linear(in_features=576, out_features=1536, bias=False)\n",
" (down_proj): Linear(in_features=1536, out_features=576, bias=False)\n",
" (act_fn): SiLU()\n",
" )\n",
" (input_layernorm): LlamaRMSNorm((576,), eps=1e-05)\n",
" (post_attention_layernorm): LlamaRMSNorm((576,), eps=1e-05)\n",
" )\n",
" )\n",
" (norm): LlamaRMSNorm((576,), eps=1e-05)\n",
" (rotary_emb): LlamaRotaryEmbedding()\n",
" )\n",
" (score): Linear(in_features=576, out_features=2, bias=False)\n",
")"
]
},
"execution_count": 1,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"from transformers import GPT2Tokenizer, LlamaForSequenceClassification\n",
"\n",
"# Load the GPT2 tokenizer and Llama model for sequence classification\n",
"model_path = r\"C:\\Users\\jatin\\OneDrive\\Desktop\\plagiarism-detection\\smolLM-fined-tuned-for-PLAGAIRISM-Detection\\model\"\n",
"tokenizer = GPT2Tokenizer.from_pretrained(model_path, local_files_only=True)\n",
"model = LlamaForSequenceClassification.from_pretrained(model_path, local_files_only=True)\n",
"\n",
"# Set model to evaluation mode\n",
"model.eval()"
]
},
{
"cell_type": "code",
"execution_count": 2,
"metadata": {},
"outputs": [
{
"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>sentence1</th>\n",
" <th>sentence2</th>\n",
" <th>label</th>\n",
" </tr>\n",
" </thead>\n",
" <tbody>\n",
" <tr>\n",
" <th>0</th>\n",
" <td>A person on a horse jumps over a broken down a...</td>\n",
" <td>A person is at a diner, ordering an omelette.</td>\n",
" <td>0</td>\n",
" </tr>\n",
" <tr>\n",
" <th>1</th>\n",
" <td>A person on a horse jumps over a broken down a...</td>\n",
" <td>A person is outdoors, on a horse.</td>\n",
" <td>1</td>\n",
" </tr>\n",
" <tr>\n",
" <th>2</th>\n",
" <td>Children smiling and waving at camera</td>\n",
" <td>There are children present</td>\n",
" <td>1</td>\n",
" </tr>\n",
" <tr>\n",
" <th>3</th>\n",
" <td>Children smiling and waving at camera</td>\n",
" <td>The kids are frowning</td>\n",
" <td>0</td>\n",
" </tr>\n",
" <tr>\n",
" <th>4</th>\n",
" <td>A boy is jumping on skateboard in the middle o...</td>\n",
" <td>The boy skates down the sidewalk.</td>\n",
" <td>0</td>\n",
" </tr>\n",
" </tbody>\n",
"</table>\n",
"</div>"
],
"text/plain": [
" sentence1 \\\n",
"0 A person on a horse jumps over a broken down a... \n",
"1 A person on a horse jumps over a broken down a... \n",
"2 Children smiling and waving at camera \n",
"3 Children smiling and waving at camera \n",
"4 A boy is jumping on skateboard in the middle o... \n",
"\n",
" sentence2 label \n",
"0 A person is at a diner, ordering an omelette. 0 \n",
"1 A person is outdoors, on a horse. 1 \n",
"2 There are children present 1 \n",
"3 The kids are frowning 0 \n",
"4 The boy skates down the sidewalk. 0 "
]
},
"execution_count": 2,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"import torch\n",
"import pandas as pd\n",
"\n",
"df = pd.read_csv(\"train_snli.txt\", delimiter='\\t', header=None, names=['sentence1', 'sentence2', 'label'])\n",
"\n",
"df.head()"
]
},
{
"cell_type": "code",
"execution_count": 3,
"metadata": {},
"outputs": [],
"source": [
"from torch.utils.data import Dataset, DataLoader\n",
"device = torch.device(\"cuda\" if torch.cuda.is_available() else \"cpu\")\n",
"\n",
"class PlagiarismDataset(Dataset):\n",
" def __init__(self, df, tokenizer, max_length=128):\n",
" self.df = df\n",
" self.tokenizer = tokenizer\n",
" self.max_length = max_length\n",
"\n",
" def __len__(self):\n",
" return len(self.df)\n",
"\n",
" def __getitem__(self, index):\n",
" row = self.df.iloc[index]\n",
"\n",
" # Ensure the sentences are strings; convert or skip if not\n",
" sentence1 = str(row['sentence1']) if not pd.isna(row['sentence1']) else \"\"\n",
" sentence2 = str(row['sentence2']) if not pd.isna(row['sentence2']) else \"\"\n",
"\n",
" inputs = self.tokenizer(\n",
" sentence1, sentence2,\n",
" add_special_tokens=True,\n",
" max_length=self.max_length,\n",
" padding='max_length',\n",
" truncation=True,\n",
" return_tensors=\"pt\"\n",
" )\n",
"\n",
" label = torch.tensor(row['label'], dtype=torch.long)\n",
"\n",
" return {\n",
" 'input_ids': inputs['input_ids'].squeeze(0),\n",
" 'attention_mask': inputs['attention_mask'].squeeze(0),\n",
" 'label': label\n",
" }\n",
"\n",
"def collate_fn(batch):\n",
" input_ids = torch.stack([item['input_ids'] for item in batch])\n",
" attention_masks = torch.stack([item['attention_mask'] for item in batch])\n",
" labels = torch.stack([item['label'] for item in batch])\n",
"\n",
" return {\n",
" 'input_ids': input_ids,\n",
" 'attention_mask': attention_masks,\n",
" 'label': labels\n",
" }"
]
},
{
"cell_type": "code",
"execution_count": 4,
"metadata": {},
"outputs": [
{
"data": {
"text/plain": [
"device(type='cuda')"
]
},
"execution_count": 4,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"device"
]
},
{
"cell_type": "code",
"execution_count": 5,
"metadata": {},
"outputs": [],
"source": [
"# Assuming you have a separate test set or validation set (e.g., df_test)\n",
"df_test = df[3_66_900:]\n",
"# Add padding token if not already\n",
"tokenizer.add_special_tokens({'pad_token': '[PAD]'})\n",
"\n",
"# Resize the model's token embeddings to fit the new tokenizer\n",
"model.resize_token_embeddings(len(tokenizer))\n",
"\n",
"# Create DataLoader for the test set\n",
"test_dataset = PlagiarismDataset(df_test, tokenizer)\n",
"test_data_loader = DataLoader(test_dataset, batch_size=16, shuffle=False, collate_fn=collate_fn)"
]
},
{
"cell_type": "code",
"execution_count": 6,
"metadata": {},
"outputs": [
{
"name": "stdout",
"output_type": "stream",
"text": [
"Classification Report:\n",
" precision recall f1-score support\n",
"\n",
" 0 1.00 1.00 1.00 236\n",
" 1 1.00 1.00 1.00 237\n",
"\n",
" accuracy 1.00 473\n",
" macro avg 1.00 1.00 1.00 473\n",
"weighted avg 1.00 1.00 1.00 473\n",
"\n"
]
}
],
"source": [
"from sklearn.metrics import classification_report\n",
"# Function to evaluate model on the test set\n",
"# Set up device\n",
"device = torch.device(\"cuda\" if torch.cuda.is_available() else \"cpu\")\n",
"\n",
"# Move model to the appropriate device\n",
"model = model.to(device)\n",
"\n",
"# Function to evaluate the model\n",
"def evaluate_model(model, data_loader):\n",
" model.eval() # Set model to evaluation mode\n",
" preds_list = []\n",
" labels_list = []\n",
"\n",
" with torch.no_grad(): # Disable gradient calculation for evaluation\n",
" for batch in data_loader:\n",
" # Move input tensors to the same device as the model\n",
" input_ids = batch['input_ids'].to(device)\n",
" attention_mask = batch['attention_mask'].to(device)\n",
" labels = batch['label'].to(device)\n",
" \n",
" # Get model outputs\n",
" outputs = model(input_ids=input_ids, attention_mask=attention_mask)\n",
" preds = torch.argmax(outputs.logits, dim=1)\n",
"\n",
" # Append predictions and true labels to respective lists\n",
" preds_list.extend(preds.cpu().numpy())\n",
" labels_list.extend(labels.cpu().numpy())\n",
" \n",
" # Compute evaluation metrics\n",
" from sklearn.metrics import classification_report\n",
" report = classification_report(labels_list, preds_list)\n",
" print(\"Classification Report:\\n\", report)\n",
"\n",
"# Evaluate the model\n",
"evaluate_model(model, test_data_loader)"
]
}
],
"metadata": {
"kernelspec": {
"display_name": "LLM",
"language": "python",
"name": "python3"
},
"language_info": {
"codemirror_mode": {
"name": "ipython",
"version": 3
},
"file_extension": ".py",
"mimetype": "text/x-python",
"name": "python",
"nbconvert_exporter": "python",
"pygments_lexer": "ipython3",
"version": "3.9.20"
}
},
"nbformat": 4,
"nbformat_minor": 2
}
|