edit baseline notebook

baselines.ipynb

@@ -206,7 +206,7 @@
    "metadata": {},
    "outputs": [],
    "source": [
-    "class ToxicDataset(Dataset):\n",
+    "class PersuasionDataset(Dataset):\n",
     "    \n",
     "    def __init__(self, tokenizer: AutoTokenizer, dataframe: pd.DataFrame, lazy: bool = False):\n",
     "        self.tokenizer = tokenizer\n",
@@ -252,9 +252,9 @@
     "    y = torch.stack(y)\n",
     "    return x.cuda(), y.cuda(), index\n",
     "\n",
-    "train_dataset = ToxicDataset(tokenizer, train_df, lazy=True)\n",
-    "dev_dataset = ToxicDataset(tokenizer, val_df, lazy=True)\n",
-    "test_dataset = ToxicDataset(tokenizer, test_df, lazy=True)\n",
+    "train_dataset = PersuasionDataset(tokenizer, train_df, lazy=True)\n",
+    "dev_dataset = PersuasionDataset(tokenizer, val_df, lazy=True)\n",
+    "test_dataset = PersuasionDataset(tokenizer, test_df, lazy=True)\n",
     "collate_fn = partial(collate_fn)\n",
     "BATCH_SIZE = 8\n",
     "train_sampler = RandomSampler(train_dataset)\n",
@@ -489,35 +489,14 @@
     "    evaluate(model, test_iterator, test_df)"
    ]
   },
-  {
-   "cell_type": "code",
-   "execution_count": 31,
-   "metadata": {},
-   "outputs": [],
-   "source": [
-    "# from transformers import BertTokenizer, DistillBertModel, AdamW, get_linear_schedule_with_warmup, DistillBertPreTrainedModel\n",
-    "from sklearn.metrics import roc_auc_score, f1_score, precision_score, recall_score, accuracy_score\n",
-    "from tqdm import tqdm\n",
-    "from pathlib import Path\n",
-    "import logging\n",
-    "import pickle\n",
-    "import random\n",
-    "from nltk.tokenize import sent_tokenize, word_tokenize\n",
-    "import itertools\n",
-    "import random\n",
-    "random.seed(0)\n",
-    "np.random.seed(0)\n",
-    "import json"
-   ]
-  },
   {
    "cell_type": "code",
    "execution_count": 29,
    "metadata": {},
    "outputs": [],
    "source": [
     "from sklearn.dummy import DummyClassifier\n",
-    "from sklearn.feature_extraction.text import TfidfVectorizer\n"
+    "from sklearn.feature_extraction.text import TfidfVectorizer"
    ]
   },
   {