course/videos/trainer_api.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "This notebook regroups the code sample of the video below, which is a part of the [Hugging Face course](https://huggingface.co/course)."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "cellView": "form"
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<iframe width=\"560\" height=\"315\" src=\"https://www.youtube.com/embed/nvBXf7s7vTI?rel=0&amp;controls=0&amp;showinfo=0\" frameborder=\"0\" allowfullscreen></iframe>"
      ],
      "text/plain": [
       "<IPython.core.display.HTML object>"
      ]
     },
     "execution_count": null,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "#@title\n",
    "from IPython.display import HTML\n",
    "\n",
    "HTML('<iframe width=\"560\" height=\"315\" src=\"https://www.youtube.com/embed/nvBXf7s7vTI?rel=0&amp;controls=0&amp;showinfo=0\" frameborder=\"0\" allowfullscreen></iframe>')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Install the Transformers and Datasets libraries to run this notebook."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "! pip install datasets transformers[sentencepiece]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Reusing dataset glue (/home/sgugger/.cache/huggingface/datasets/glue/mrpc/1.0.0/dacbe3125aa31d7f70367a07a8a9e72a5a0bfeb5fc42e75c9db75b96da6053ad)\n",
      "Loading cached processed dataset at /home/sgugger/.cache/huggingface/datasets/glue/mrpc/1.0.0/dacbe3125aa31d7f70367a07a8a9e72a5a0bfeb5fc42e75c9db75b96da6053ad/cache-8174fd92eed0af98.arrow\n",
      "Loading cached processed dataset at /home/sgugger/.cache/huggingface/datasets/glue/mrpc/1.0.0/dacbe3125aa31d7f70367a07a8a9e72a5a0bfeb5fc42e75c9db75b96da6053ad/cache-8c99fb059544bc96.arrow\n",
      "Loading cached processed dataset at /home/sgugger/.cache/huggingface/datasets/glue/mrpc/1.0.0/dacbe3125aa31d7f70367a07a8a9e72a5a0bfeb5fc42e75c9db75b96da6053ad/cache-e625eb72bcf1ae1f.arrow\n"
     ]
    }
   ],
   "source": [
    "from datasets import load_dataset\n",
    "from transformers import AutoTokenizer, DataCollatorWithPadding\n",
    "\n",
    "raw_datasets = load_dataset(\"glue\", \"mrpc\")\n",
    "checkpoint = \"bert-base-cased\"\n",
    "tokenizer = AutoTokenizer.from_pretrained(checkpoint)\n",
    "\n",
    "def tokenize_function(examples):\n",
    "    return tokenizer(examples[\"sentence1\"], examples[\"sentence2\"], truncation=True)\n",
    "\n",
    "tokenized_datasets = raw_datasets.map(tokenize_function, batched=True)\n",
    "data_collator = DataCollatorWithPadding(tokenizer)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Some weights of the model checkpoint at bert-base-cased were not used when initializing BertForSequenceClassification: ['cls.predictions.decoder.weight', 'cls.predictions.bias', 'cls.predictions.transform.LayerNorm.bias', 'cls.predictions.transform.dense.weight', 'cls.predictions.transform.dense.bias', 'cls.seq_relationship.bias', 'cls.seq_relationship.weight', 'cls.predictions.transform.LayerNorm.weight']\n",
      "- This IS expected if you are initializing BertForSequenceClassification 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 BertForSequenceClassification 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 BertForSequenceClassification were not initialized from the model checkpoint at bert-base-cased and are newly initialized: ['classifier.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": [
    "from transformers import AutoModelForSequenceClassification\n",
    "\n",
    "model = AutoModelForSequenceClassification.from_pretrained(checkpoint, num_labels=2)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "from transformers import TrainingArguments\n",
    "\n",
    "training_args = TrainingArguments(\"test-trainer\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "from transformers import TrainingArguments\n",
    "\n",
    "training_args = TrainingArguments(\n",
    "    \"test-trainer\",\n",
    "    per_device_train_batch_size=16,\n",
    "    per_device_eval_batch_size=16,\n",
    "    num_train_epochs=5,\n",
    "    learning_rate=2e-5,\n",
    "    weight_decay=0.01,\n",
    ")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "\n",
       "    <div>\n",
       "      \n",
       "      <progress value='1150' max='1150' style='width:300px; height:20px; vertical-align: middle;'></progress>\n",
       "      [1150/1150 01:57, Epoch 5/5]\n",
       "    </div>\n",
       "    <table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: left;\">\n",
       "      <th>Step</th>\n",
       "      <th>Training Loss</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <td>500</td>\n",
       "      <td>0.443800</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>1000</td>\n",
       "      <td>0.127500</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table><p>"
      ],
      "text/plain": [
       "<IPython.core.display.HTML object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/plain": [
       "TrainOutput(global_step=1150, training_loss=0.25629232904185417, metrics={'train_runtime': 118.2961, 'train_samples_per_second': 155.035, 'train_steps_per_second': 9.721, 'total_flos': 939591087838320.0, 'train_loss': 0.25629232904185417, 'epoch': 5.0})"
      ]
     },
     "execution_count": null,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "from transformers import Trainer\n",
    "\n",
    "trainer = Trainer(\n",
    "    model,\n",
    "    training_args,\n",
    "    train_dataset=tokenized_datasets[\"train\"],\n",
    "    eval_dataset=tokenized_datasets[\"validation\"],\n",
    "    data_collator=data_collator,\n",
    "    tokenizer=tokenizer,\n",
    ")\n",
    "trainer.train()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "\n",
       "    <div>\n",
       "      \n",
       "      <progress value='26' max='26' style='width:300px; height:20px; vertical-align: middle;'></progress>\n",
       "      [26/26 00:00]\n",
       "    </div>\n",
       "    "
      ],
      "text/plain": [
       "<IPython.core.display.HTML object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "(408, 2) (408,)\n"
     ]
    }
   ],
   "source": [
    "predictions = trainer.predict(tokenized_datasets[\"validation\"])\n",
    "print(predictions.predictions.shape, predictions.label_ids.shape)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "{'accuracy': 0.8627450980392157, 'f1': 0.9050847457627118}"
      ]
     },
     "execution_count": null,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "import numpy as np\n",
    "from datasets import load_metric\n",
    "\n",
    "metric = load_metric(\"glue\", \"mrpc\")\n",
    "preds = np.argmax(predictions.predictions, axis=-1)\n",
    "metric.compute(predictions=preds, references=predictions.label_ids)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "metric = load_metric(\"glue\", \"mrpc\")\n",
    "\n",
    "def compute_metrics(eval_preds):\n",
    "    logits, labels = eval_preds\n",
    "    predictions = np.argmax(logits, axis=-1)\n",
    "    return metric.compute(predictions=predictions, references=labels)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Some weights of the model checkpoint at bert-base-cased were not used when initializing BertForSequenceClassification: ['cls.predictions.decoder.weight', 'cls.predictions.bias', 'cls.predictions.transform.LayerNorm.bias', 'cls.predictions.transform.dense.weight', 'cls.predictions.transform.dense.bias', 'cls.seq_relationship.bias', 'cls.seq_relationship.weight', 'cls.predictions.transform.LayerNorm.weight']\n",
      "- This IS expected if you are initializing BertForSequenceClassification 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 BertForSequenceClassification 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 BertForSequenceClassification were not initialized from the model checkpoint at bert-base-cased and are newly initialized: ['classifier.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"
     ]
    },
    {
     "data": {
      "text/html": [
       "\n",
       "    <div>\n",
       "      \n",
       "      <progress value='1377' max='1377' style='width:300px; height:20px; vertical-align: middle;'></progress>\n",
       "      [1377/1377 01:19, Epoch 3/3]\n",
       "    </div>\n",
       "    <table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: left;\">\n",
       "      <th>Epoch</th>\n",
       "      <th>Training Loss</th>\n",
       "      <th>Validation Loss</th>\n",
       "      <th>Accuracy</th>\n",
       "      <th>F1</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <td>1</td>\n",
       "      <td>No log</td>\n",
       "      <td>0.410291</td>\n",
       "      <td>0.823529</td>\n",
       "      <td>0.877966</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>2</td>\n",
       "      <td>0.510500</td>\n",
       "      <td>0.405150</td>\n",
       "      <td>0.845588</td>\n",
       "      <td>0.893401</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>3</td>\n",
       "      <td>0.258300</td>\n",
       "      <td>0.713577</td>\n",
       "      <td>0.862745</td>\n",
       "      <td>0.902778</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table><p>"
      ],
      "text/plain": [
       "<IPython.core.display.HTML object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/plain": [
       "TrainOutput(global_step=1377, training_loss=0.306962023633238, metrics={'train_runtime': 79.4823, 'train_samples_per_second': 138.446, 'train_steps_per_second': 17.325, 'total_flos': 530544172170240.0, 'train_loss': 0.306962023633238, 'epoch': 3.0})"
      ]
     },
     "execution_count": null,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "training_args = TrainingArguments(\"test-trainer\", evaluation_strategy=\"epoch\")\n",
    "model = AutoModelForSequenceClassification.from_pretrained(checkpoint, num_labels=2)\n",
    "\n",
    "trainer = Trainer(\n",
    "    model,\n",
    "    training_args,\n",
    "    train_dataset=tokenized_datasets[\"train\"],\n",
    "    eval_dataset=tokenized_datasets[\"validation\"],\n",
    "    data_collator=data_collator,\n",
    "    tokenizer=tokenizer,\n",
    "    compute_metrics=compute_metrics\n",
    ")\n",
    "trainer.train()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "colab": {
   "name": "The Trainer API",
   "provenance": []
  }
 },
 "nbformat": 4,
 "nbformat_minor": 4
}