NLP with Transformers: Fundamentals and Core ApplicationsChapter 97

7. Step 4: Fine-Tuning BERT

Section 7 of 10-~ 12 min read-Synced from Cuantum content

We’ll fine-tune a pre-trained BERT model for binary classification (positive or negative sentiment).

Code Example: Fine-Tuning

# Load pre-trained BERT modelmodel = BertForSequenceClassification.from_pretrained("bert-base-uncased", num_labels=2) # Define training argumentstraining_args = TrainingArguments(    output_dir="./results",    evaluation_strategy="epoch",    learning_rate=2e-5,    per_device_train_batch_size=8,    num_train_epochs=3,    weight_decay=0.01,) # Initialize Trainertrainer = Trainer(    model=model,    args=training_args,    train_dataset=tokenized_train,    eval_dataset=tokenized_test,) # Train the modeltrainer.train()