Binary Classification/Transformer Comparison/Is This Tweet Referring to a Disaster or Not? - DeBERTa.ipynb

{
 "cells": [
  {
   "attachments": {},
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Is This Tweet Referring to a Disaster or Not?\n",
    "\n",
    "Dataset Source: https://www.kaggle.com/datasets/vstepanenko/disaster-tweets"
   ]
  },
  {
   "attachments": {},
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Import Necessary Libraries"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "NOTE: Redirects are currently not supported in Windows or MacOs.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Git LFS initialized.\n"
     ]
    }
   ],
   "source": [
    "import os, sys\n",
    "os.environ['TOKENIZERS_PARALLELISM']='false'\n",
    "\n",
    "import numpy as np\n",
    "import pandas as pd\n",
    "import matplotlib.pyplot as plt\n",
    "\n",
    "import datasets\n",
    "from datasets import Dataset, DatasetDict\n",
    "\n",
    "import torch\n",
    "\n",
    "import transformers\n",
    "from transformers import AutoModelForSequenceClassification, AutoTokenizer\n",
    "from transformers import TrainingArguments, Trainer, set_seed, DataCollatorWithPadding\n",
    "\n",
    "import evaluate\n",
    "\n",
    "!git lfs install"
   ]
  },
  {
   "attachments": {},
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Display Library Versions"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "        Python: 3.9.12\n",
      "         NumPy: 1.24.2\n",
      "        Pandas: 1.5.0\n",
      "      Datasets: 2.9.0\n",
      "  Transformers: 4.26.1\n",
      "         Torch: 1.13.1\n",
      "      Evaluate: 0.4.0\n"
     ]
    }
   ],
   "source": [
    "print(\"Python:\".rjust(15), sys.version[0:6])\n",
    "print(\"NumPy:\".rjust(15), np.__version__)\n",
    "print(\"Pandas:\".rjust(15), pd.__version__)\n",
    "print(\"Datasets:\".rjust(15), datasets.__version__)\n",
    "print(\"Transformers:\".rjust(15), transformers.__version__)\n",
    "print(\"Torch:\".rjust(15), torch.__version__)\n",
    "print(\"Evaluate:\".rjust(15), evaluate.__version__)"
   ]
  },
  {
   "attachments": {},
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Load & Start Preprocessing Dataset"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "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>text</th>\n",
       "      <th>labels</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>Communal violence in Bhainsa, Telangana. \"Ston...</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>Telangana: Section 144 has been imposed in Bha...</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>Arsonist sets cars ablaze at dealership https:...</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>Arsonist sets cars ablaze at dealership https:...</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>\"Lord Jesus, your love brings freedom and pard...</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>11365</th>\n",
       "      <td>Media should have warned us well in advance. T...</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>11366</th>\n",
       "      <td>i feel directly attacked 💀 i consider moonbin ...</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>11367</th>\n",
       "      <td>i feel directly attacked 💀 i consider moonbin ...</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>11368</th>\n",
       "      <td>ok who remember \"outcast\" nd the \"dora\" au?? T...</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>11369</th>\n",
       "      <td>Jake Corway wrecked while running 14th at IRP.</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>11370 rows × 2 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "                                                    text  labels\n",
       "0      Communal violence in Bhainsa, Telangana. \"Ston...       1\n",
       "1      Telangana: Section 144 has been imposed in Bha...       1\n",
       "2      Arsonist sets cars ablaze at dealership https:...       1\n",
       "3      Arsonist sets cars ablaze at dealership https:...       1\n",
       "4      \"Lord Jesus, your love brings freedom and pard...       0\n",
       "...                                                  ...     ...\n",
       "11365  Media should have warned us well in advance. T...       0\n",
       "11366  i feel directly attacked 💀 i consider moonbin ...       0\n",
       "11367  i feel directly attacked 💀 i consider moonbin ...       0\n",
       "11368  ok who remember \"outcast\" nd the \"dora\" au?? T...       0\n",
       "11369     Jake Corway wrecked while running 14th at IRP.       1\n",
       "\n",
       "[11370 rows x 2 columns]"
      ]
     },
     "execution_count": 3,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "parent_dir = \"/Users/briandunn/Desktop/nlp_ds/Disaster Tweets\"\n",
    "\n",
    "data_file = os.path.join(parent_dir, \"tweets.csv\")\n",
    "\n",
    "data = pd.read_csv(data_file)\n",
    "\n",
    "data.drop(columns=['id', 'keyword', 'location'], inplace=True)\n",
    "\n",
    "data.rename(columns={\"target\": \"labels\"}, inplace=True)\n",
    "\n",
    "data"
   ]
  },
  {
   "attachments": {},
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Define Dictionaries to Convert Between String & Integer Class Values"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [],
   "source": [
    "unique_labels = [\"non-disaster\", \"disaster\"]\n",
    "id2label = {idx:label for idx, label in enumerate(unique_labels)}\n",
    "label2id = {label:idx for idx, label in enumerate(unique_labels)}\n",
    "NUM_OF_LABELS = len(unique_labels)"
   ]
  },
  {
   "attachments": {},
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Visual: Histogram of Text Lengths"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 1800x800 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "The maximum length of Article Abstracts (in words) is 32 words.\n"
     ]
    }
   ],
   "source": [
    "data['text_length'] = data['text'].str.split().str.len()\n",
    "\n",
    "n_bin=35\n",
    "data.hist('text_length', grid=False, figsize=(18,8), range=(0,n_bin), bins=n_bin)\n",
    "plt.title('Histogram: Length of Article Abstracts')\n",
    "plt.xlabel('# of Words in Article Abstracts')\n",
    "plt.ylabel('# of Samples')\n",
    "plt.show()\n",
    "\n",
    "print(f\"The maximum length of Article Abstracts (in words) is {max(data['text_length'])} words.\")"
   ]
  },
  {
   "attachments": {},
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Visual: Tweet Length by Class"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 1000x400 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "data.boxplot(\"text_length\", by=\"labels\", grid=False, showfliers=False, color=\"Blue\", figsize=(10,4))\n",
    "\n",
    "plt.suptitle('')\n",
    "plt.title('Tweet Length by Type')\n",
    "plt.xlabel('')\n",
    "plt.ylabel('Words per Tweet')\n",
    "plt.show()"
   ]
  },
  {
   "attachments": {},
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Find Largest Tokenized Input Value"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Special tokens have been added in the vocabulary, make sure the associated word embeddings are fine-tuned or trained.\n",
      "/Users/briandunn/Documents/nlpnn/nlp_ml/lib/python3.9/site-packages/transformers/convert_slow_tokenizer.py:446: UserWarning: The sentencepiece tokenizer that you are converting to a fast tokenizer uses the byte fallback option which is not implemented in the fast tokenizers. In practice this means that the fast version of the tokenizer can produce unknown tokens whereas the sentencepiece version would have converted these unknown tokens into a sequence of byte tokens matching the original piece of text.\n",
      "  warnings.warn(\n",
      "Special tokens have been added in the vocabulary, make sure the associated word embeddings are fine-tuned or trained.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "The longest article length is 133 tokens.\n"
     ]
    }
   ],
   "source": [
    "MODEL_CKPT = \"microsoft/deberta-v3-small\"\n",
    "\n",
    "tokenizer = AutoTokenizer.from_pretrained(MODEL_CKPT, \n",
    "                                             truncation=True, \n",
    "                                             padding=True,\n",
    "                                             batched=True, \n",
    "                                             batch_size=None)\n",
    "\n",
    "text_len = [len(tokenizer.encode(a)) for a in data[\"text\"]]\n",
    "\n",
    "print(f\"The longest article length is {max(text_len)} tokens.\")"
   ]
  },
  {
   "attachments": {},
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Final Data Preprocessing"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "<class 'pandas.core.frame.DataFrame'>\n",
      "RangeIndex: 11370 entries, 0 to 11369\n",
      "Data columns (total 2 columns):\n",
      " #   Column  Non-Null Count  Dtype \n",
      "---  ------  --------------  ----- \n",
      " 0   text    11370 non-null  object\n",
      " 1   labels  11370 non-null  int64 \n",
      "dtypes: int64(1), object(1)\n",
      "memory usage: 177.8+ KB\n"
     ]
    }
   ],
   "source": [
    "data.drop(columns=[\"text_length\"], inplace=True)\n",
    "data.info()"
   ]
  },
  {
   "attachments": {},
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Convert DataFrame to Dataset & Split Dataset into Train/Test/Evaluation Datasets"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Training Dataset Shape: (9096, 2)\n",
      "Testing Dataset Shape: (1137, 2)\n",
      "Evaluation Dataset Shape: (1137, 2)\n"
     ]
    }
   ],
   "source": [
    "dataset = Dataset.from_pandas(data)\n",
    "\n",
    "train_testEval = dataset.train_test_split(train_size=0.80)\n",
    "test_eval = train_testEval['test'].train_test_split(train_size=0.50)\n",
    "\n",
    "ds = DatasetDict({\n",
    "    'train' : train_testEval['train'],\n",
    "    'test' : test_eval['train'],\n",
    "    'eval' : test_eval['test'],\n",
    "})\n",
    "\n",
    "print(\"Training Dataset Shape:\", ds['train'].shape)\n",
    "print(\"Testing Dataset Shape:\", ds['test'].shape)\n",
    "print(\"Evaluation Dataset Shape:\", ds['eval'].shape)"
   ]
  },
  {
   "attachments": {},
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Basic Values/Constants"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "This project will utilize the cpu to train and evaluate.\n",
      "This project will have a maximum length for the input text of 136 tokens.\n",
      "The model checkpoint used for this project is: microsoft/deberta-v3-small\n"
     ]
    }
   ],
   "source": [
    "set_seed = 42\n",
    "NUM_OF_EPOCHS = 5\n",
    "\n",
    "BATCH_SIZE = 64\n",
    "LEARNING_RATE = 2e-5\n",
    "\n",
    "MAX_LENGTH = int(round(max(text_len) * 1.025, 0))\n",
    "MODEL_CKPT = \"microsoft/deberta-v3-small\"\n",
    "MODEL_NAME = MODEL_CKPT.split(\"/\")[-1] + \"-Tweet_About_Disaster_Or_Not\"\n",
    "\n",
    "METRIC_NAME = \"f1\"\n",
    "DEVICE = torch.device(\"cpu\")\n",
    "\n",
    "print(f\"This project will utilize the {DEVICE} to train and evaluate.\")\n",
    "print(f\"This project will have a maximum length for the input text of {MAX_LENGTH} tokens.\")\n",
    "print(f\"The model checkpoint used for this project is: {MODEL_CKPT}\")"
   ]
  },
  {
   "attachments": {},
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Tokenize Entire Dataset"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Special tokens have been added in the vocabulary, make sure the associated word embeddings are fine-tuned or trained.\n",
      "/Users/briandunn/Documents/nlpnn/nlp_ml/lib/python3.9/site-packages/transformers/convert_slow_tokenizer.py:446: UserWarning: The sentencepiece tokenizer that you are converting to a fast tokenizer uses the byte fallback option which is not implemented in the fast tokenizers. In practice this means that the fast version of the tokenizer can produce unknown tokens whereas the sentencepiece version would have converted these unknown tokens into a sequence of byte tokens matching the original piece of text.\n",
      "  warnings.warn(\n",
      "Special tokens have been added in the vocabulary, make sure the associated word embeddings are fine-tuned or trained.\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "c1ae9053f2d144978687cf6ae6007c9f",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "A Jupyter Widget"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "6c4b4f388a344d2bac4123905a0fcfa8",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "A Jupyter Widget"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "a67cbb639a2e40b3b780c0622fff3101",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "A Jupyter Widget"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "training data shape: (9096, 5)\n",
      "testing data shape: (1137, 5)\n",
      "validation data shape: (1137, 5)\n",
      "{'text': Value(dtype='string', id=None), 'labels': Value(dtype='int64', id=None), 'input_ids': Sequence(feature=Value(dtype='int32', id=None), length=-1, id=None), 'token_type_ids': Sequence(feature=Value(dtype='int8', id=None), length=-1, id=None), 'attention_mask': Sequence(feature=Value(dtype='int8', id=None), length=-1, id=None)}\n",
      "Dataset({\n",
      "    features: ['text', 'labels', 'input_ids', 'token_type_ids', 'attention_mask'],\n",
      "    num_rows: 9096\n",
      "})\n"
     ]
    }
   ],
   "source": [
    "# Define the Tokenizer\n",
    "tokenizer = AutoTokenizer.from_pretrained(MODEL_CKPT)\n",
    "\n",
    "# Method to Tokenize each Sample in the Dataset\n",
    "def tokenize(batch):\n",
    "    return tokenizer(batch['text'], \n",
    "                     padding=True, \n",
    "                     truncation=True, \n",
    "                     max_length=MAX_LENGTH, \n",
    "                     return_tensors=\"pt\")\n",
    "\n",
    "# Map the 'Tokenize' Method to All Samples in the Entire Dataset\n",
    "encoded_data = ds.map(tokenize, batched=True, batch_size=None)\n",
    "\n",
    "print('training data shape:', encoded_data['train'].shape)\n",
    "print('testing data shape:', encoded_data['test'].shape)\n",
    "print('validation data shape:', encoded_data['eval'].shape)\n",
    "\n",
    "print(encoded_data['train'].features)\n",
    "print(encoded_data['train'])"
   ]
  },
  {
   "attachments": {},
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Instantiate Model"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Some weights of the model checkpoint at microsoft/deberta-v3-small were not used when initializing DebertaV2ForSequenceClassification: ['lm_predictions.lm_head.LayerNorm.bias', 'mask_predictions.dense.weight', 'lm_predictions.lm_head.dense.bias', 'lm_predictions.lm_head.bias', 'mask_predictions.LayerNorm.bias', 'mask_predictions.classifier.weight', 'lm_predictions.lm_head.dense.weight', 'mask_predictions.LayerNorm.weight', 'lm_predictions.lm_head.LayerNorm.weight', 'mask_predictions.classifier.bias', 'mask_predictions.dense.bias']\n",
      "- This IS expected if you are initializing DebertaV2ForSequenceClassification from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model).\n",
      "- This IS NOT expected if you are initializing DebertaV2ForSequenceClassification from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model).\n",
      "Some weights of DebertaV2ForSequenceClassification were not initialized from the model checkpoint at microsoft/deberta-v3-small and are newly initialized: ['pooler.dense.weight', 'classifier.bias', 'pooler.dense.bias', 'classifier.weight']\n",
      "You should probably TRAIN this model on a down-stream task to be able to use it for predictions and inference.\n"
     ]
    }
   ],
   "source": [
    "model = (\n",
    "    AutoModelForSequenceClassification.from_pretrained(\n",
    "        MODEL_CKPT,\n",
    "        num_labels=NUM_OF_LABELS\n",
    "    ).to(DEVICE))"
   ]
  },
  {
   "attachments": {},
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Define Function to Compute Metrics"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {},
   "outputs": [],
   "source": [
    "def compute_metrics(p):\n",
    "    accuracy_metric = evaluate.load(\"accuracy\")\n",
    "    accuracy = accuracy_metric.compute(predictions=np.argmax(p.predictions, axis=1), references=p.label_ids)['accuracy']\n",
    "    \n",
    "    ### ------------------- F1 scores -------------------\n",
    "    \n",
    "    f1_score_metric = evaluate.load(\"f1\")\n",
    "    f1_score = f1_score_metric.compute(predictions=np.argmax(p.predictions, axis=1), references=p.label_ids)[\"f1\"]\n",
    "    \n",
    "    ### ------------------- recall -------------------\n",
    "    \n",
    "    recall_metric = evaluate.load(\"recall\")\n",
    "    recall = recall_metric.compute(predictions=np.argmax(p.predictions, axis=1), references=p.label_ids)[\"recall\"]\n",
    "    \n",
    "    ### ------------------- precision -------------------\n",
    "    \n",
    "    precision_metric = evaluate.load(\"precision\")\n",
    "    precision = precision_metric.compute(predictions=np.argmax(p.predictions, axis=1), references=p.label_ids)[\"precision\"]\n",
    "    \n",
    "    return {\"accuracy\" : accuracy, \n",
    "            \"F1\" : f1_score,\n",
    "            \"Recall\" : recall,\n",
    "            \"Precision\" : precision,\n",
    "            }"
   ]
  },
  {
   "attachments": {},
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Define Training Arguments"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {},
   "outputs": [],
   "source": [
    "training_args = TrainingArguments(\n",
    "    output_dir=MODEL_NAME,\n",
    "    num_train_epochs=NUM_OF_EPOCHS,\n",
    "    learning_rate=LEARNING_RATE,\n",
    "    per_device_train_batch_size=BATCH_SIZE, \n",
    "    per_device_eval_batch_size=BATCH_SIZE, \n",
    "    weight_decay=0.01,\n",
    "    log_level=\"error\",\n",
    "    logging_strategy='epoch', \n",
    "    evaluation_strategy=\"epoch\",\n",
    "    save_strategy=\"epoch\",\n",
    "    logging_first_step=True,\n",
    "    load_best_model_at_end=True,\n",
    "    disable_tqdm=False,\n",
    "    hub_private_repo=True,\n",
    "    push_to_hub=True\n",
    ")"
   ]
  },
  {
   "attachments": {},
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Define Data Collator"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {},
   "outputs": [],
   "source": [
    "data_collator = DataCollatorWithPadding(tokenizer, padding=True)"
   ]
  },
  {
   "attachments": {},
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Subclass Trainer to Handle Class Imbalance"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {},
   "outputs": [],
   "source": [
    "class CustomTrainer(Trainer):\n",
    "    def compute_loss(self, model, inputs, return_outputs=False):\n",
    "        labels = inputs.get(\"labels\")\n",
    "        outputs = model(**inputs)\n",
    "        logits = outputs.get(\"logits\")\n",
    "        loss_fct = torch.nn.CrossEntropyLoss(weight=torch.tensor([1.0, 2.0]))\n",
    "        loss = loss_fct(logits.view(-1, self.model.config.num_labels), labels.view(-1))\n",
    "        return (loss, outputs) if return_outputs else loss"
   ]
  },
  {
   "attachments": {},
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Instantiate Trainer"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Cloning https://huggingface.co/DunnBC22/deberta-v3-small-Tweet_About_Disaster_Or_Not into local empty directory.\n"
     ]
    }
   ],
   "source": [
    "trainer = CustomTrainer(model=model, \n",
    "                  args=training_args, \n",
    "                  compute_metrics=compute_metrics,\n",
    "                  train_dataset=encoded_data['train'], \n",
    "                  eval_dataset=encoded_data['eval'], \n",
    "                  data_collator=data_collator,\n",
    "                  tokenizer=tokenizer)"
   ]
  },
  {
   "attachments": {},
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Train Model"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/Users/briandunn/Documents/nlpnn/nlp_ml/lib/python3.9/site-packages/transformers/optimization.py:306: FutureWarning: This implementation of AdamW is deprecated and will be removed in a future version. Use the PyTorch implementation torch.optim.AdamW instead, or set `no_deprecation_warning=True` to disable this warning\n",
      "  warnings.warn(\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "7a413792260d474d99104feb8e2087de",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "A Jupyter Widget"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "{'loss': 0.6828, 'learning_rate': 1.9972027972027975e-05, 'epoch': 0.01}\n",
      "{'loss': 0.4167, 'learning_rate': 1.6000000000000003e-05, 'epoch': 1.0}\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "5fbae71d715b489a9228bd4a080df3d9",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "A Jupyter Widget"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "{'eval_loss': 0.31481608748435974, 'eval_accuracy': 0.8830255057167986, 'eval_F1': 0.7164179104477612, 'eval_Recall': 0.7924528301886793, 'eval_Precision': 0.6536964980544747, 'eval_runtime': 112.0574, 'eval_samples_per_second': 10.147, 'eval_steps_per_second': 0.161, 'epoch': 1.0}\n",
      "{'loss': 0.255, 'learning_rate': 1.2e-05, 'epoch': 2.0}\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "924f9d86c241411c8c74d167b86f4bda",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "A Jupyter Widget"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "{'eval_loss': 0.2942464351654053, 'eval_accuracy': 0.9050131926121372, 'eval_F1': 0.7452830188679244, 'eval_Recall': 0.7452830188679245, 'eval_Precision': 0.7452830188679245, 'eval_runtime': 107.0503, 'eval_samples_per_second': 10.621, 'eval_steps_per_second': 0.168, 'epoch': 2.0}\n",
      "{'loss': 0.1935, 'learning_rate': 8.000000000000001e-06, 'epoch': 3.0}\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "45cb9f26f8314495ac67567513372e76",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "A Jupyter Widget"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "{'eval_loss': 0.30222010612487793, 'eval_accuracy': 0.8874230430958663, 'eval_F1': 0.7288135593220341, 'eval_Recall': 0.8113207547169812, 'eval_Precision': 0.6615384615384615, 'eval_runtime': 97.6108, 'eval_samples_per_second': 11.648, 'eval_steps_per_second': 0.184, 'epoch': 3.0}\n",
      "{'loss': 0.1512, 'learning_rate': 4.000000000000001e-06, 'epoch': 4.0}\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "94d4c1b91c9248fabaf46e5780083905",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "A Jupyter Widget"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "{'eval_loss': 0.34052926301956177, 'eval_accuracy': 0.8786279683377308, 'eval_F1': 0.7172131147540985, 'eval_Recall': 0.8254716981132075, 'eval_Precision': 0.6340579710144928, 'eval_runtime': 109.9221, 'eval_samples_per_second': 10.344, 'eval_steps_per_second': 0.164, 'epoch': 4.0}\n",
      "{'loss': 0.1192, 'learning_rate': 0.0, 'epoch': 5.0}\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "b224daa4304446b8bca9290c8e2806fb",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "A Jupyter Widget"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "{'eval_loss': 0.36176347732543945, 'eval_accuracy': 0.8909410729991205, 'eval_F1': 0.7372881355932203, 'eval_Recall': 0.8207547169811321, 'eval_Precision': 0.6692307692307692, 'eval_runtime': 106.1779, 'eval_samples_per_second': 10.708, 'eval_steps_per_second': 0.17, 'epoch': 5.0}\n",
      "{'train_runtime': 26600.6068, 'train_samples_per_second': 1.71, 'train_steps_per_second': 0.027, 'train_loss': 0.2275060863761635, 'epoch': 5.0}\n"
     ]
    }
   ],
   "source": [
    "train_results = trainer.train()"
   ]
  },
  {
   "attachments": {},
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Save Model & Metrics on Training Dataset"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Several commits (2) will be pushed upstream.\n",
      "The progress bars may be unreliable.\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "0e0ba8207ae04a4c8ee65e329982462a",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "A Jupyter Widget"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "878920c76e484ab693bcfd1e963567ff",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "A Jupyter Widget"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "remote: Scanning LFS files for validity...        \n",
      "remote: LFS file scan complete.        \n",
      "To https://huggingface.co/DunnBC22/deberta-v3-small-Tweet_About_Disaster_Or_Not\n",
      "   faae496..6fd538f  main -> main\n",
      "\n",
      "To https://huggingface.co/DunnBC22/deberta-v3-small-Tweet_About_Disaster_Or_Not\n",
      "   6fd538f..d278d3e  main -> main\n",
      "\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "***** train metrics *****\n",
      "  epoch                    =        5.0\n",
      "  train_loss               =     0.2275\n",
      "  train_runtime            = 7:23:20.60\n",
      "  train_samples_per_second =       1.71\n",
      "  train_steps_per_second   =      0.027\n"
     ]
    }
   ],
   "source": [
    "trainer.save_model()\n",
    "trainer.log_metrics(\"train\", train_results.metrics)\n",
    "trainer.save_metrics(\"train\", train_results.metrics)\n",
    "trainer.save_state()"
   ]
  },
  {
   "attachments": {},
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Evaluate Model"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "00b597064da84b269fa4d5312af27bab",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "A Jupyter Widget"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "***** eval metrics *****\n",
      "  epoch                   =        5.0\n",
      "  eval_F1                 =     0.7453\n",
      "  eval_Precision          =     0.7453\n",
      "  eval_Recall             =     0.7453\n",
      "  eval_accuracy           =      0.905\n",
      "  eval_loss               =     0.2942\n",
      "  eval_runtime            = 0:01:49.70\n",
      "  eval_samples_per_second =     10.364\n",
      "  eval_steps_per_second   =      0.164\n"
     ]
    }
   ],
   "source": [
    "metrics = trainer.evaluate(encoded_data['eval'])\n",
    "trainer.log_metrics(\"eval\", metrics)\n",
    "trainer.save_metrics(\"eval\", metrics)"
   ]
  },
  {
   "attachments": {},
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Push Project to HuggingFace Hub"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "37b54c8304a548048d1f03a294beb828",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "A Jupyter Widget"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "remote: Scanning LFS files for validity...        \n",
      "remote: LFS file scan complete.        \n",
      "To https://huggingface.co/DunnBC22/deberta-v3-small-Tweet_About_Disaster_Or_Not\n",
      "   d278d3e..5482d25  main -> main\n",
      "\n",
      "To https://huggingface.co/DunnBC22/deberta-v3-small-Tweet_About_Disaster_Or_Not\n",
      "   5482d25..d69b456  main -> main\n",
      "\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "'https://huggingface.co/DunnBC22/deberta-v3-small-Tweet_About_Disaster_Or_Not/commit/5482d256484e25f0fb77af0527829feccfe7636f'"
      ]
     },
     "execution_count": 22,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "trainer.push_to_hub(commit_message=\"All Dunn!!!\")"
   ]
  },
  {
   "attachments": {},
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Notes & Other Takeaways From This Project\n",
    "****\n",
    "- This project is part of a series of binary classification models that I am using the same dataset with virtually everything identical with the exception of the transformer used (& potentially the maximum length of tokens to accommodate the various transformer requirements). The idea is to see which one is most effective and efficient.\n",
    "****"
   ]
  },
  {
   "attachments": {},
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Citations\n",
    "\n",
    "- Model Checkpoint\n",
    "    - @inproceedings{\n",
    "he2021deberta,\n",
    "title={DEBERTA: DECODING-ENHANCED BERT WITH DISENTANGLED ATTENTION},\n",
    "author={Pengcheng He and Xiaodong Liu and Jianfeng Gao and Weizhu Chen},\n",
    "booktitle={International Conference on Learning Representations},\n",
    "year={2021},\n",
    "url={https://openreview.net/forum?id=XPZIaotutsD}\n",
    "}\n"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "nlp_ml",
   "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": "3.9.12"
  },
  "orig_nbformat": 4,
  "vscode": {
   "interpreter": {
    "hash": "0b7a754c5aaaa44a21e2f2148f50b931a5abdffe7ef71537f5954179554f39df"
   }
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}