You can deploy this model and start earning money today!

Rishabh323/
text-classification-finetuning-using-BERT

Name: text-classification-finetuning-using-BERT
Availability: InStock

Views

No views yet

BERT Text Classification Fine-tuning

This project demonstrates how to fine-tune a BERT model for text classification tasks using the Hugging Face Transformers library.

Setup

Install the required dependencies:

pip install -r requirements.txt

Run the training script:

python train.py

Project Structure

train.py: Main training script

Model

The fine-tuned model will be uploaded to Hugging Face Hub and can be accessed using the model ID provided after training.

#code import torch from transformers import ( AutoModelForSequenceClassification, AutoTokenizer, Trainer, TrainingArguments, DataCollatorWithPadding ) from datasets import load_dataset from sklearn.metrics import accuracy_score, precision_recall_fscore_support import numpy as np from huggingface_hub import login, HfFolder import os import sys

def compute_metrics(pred): labels = pred.label_ids preds = pred.predictions.argmax(-1) precision, recall, f1, _ = precision_recall_fscore_support(labels, preds, average='weighted') acc = accuracy_score(labels, preds) return { 'accuracy': acc, 'f1': f1, 'precision': precision, 'recall': recall }

def main(): try: # Set Hugging Face token token = "your - HF - token- key" os.environ["HF_TOKEN"] = token

    # Login to Hugging Face Hub
    print("Logging in to Hugging Face Hub...")
    login(token=token)
    print("Successfully logged in to Hugging Face Hub")

    # Load dataset (using IMDB as an example)
    print("Loading IMDB dataset...")
    dataset = load_dataset("imdb")
    
    # Load tokenizer and model
    print("Loading BERT model and tokenizer...")
    model_name = "bert-base-uncased"
    tokenizer = AutoTokenizer.from_pretrained(model_name)
    model = AutoModelForSequenceClassification.from_pretrained(model_name, num_labels=2)

    # Tokenize the dataset
    print("Tokenizing dataset...")
    def tokenize_function(examples):
        return tokenizer(examples["text"], padding="max_length", truncation=True)

    tokenized_datasets = dataset.map(tokenize_function, batched=True)

    # Get username from token
    username = HfFolder.get_username()
    if not username:
        raise ValueError("Could not get username from token")

    # Prepare training arguments
    training_args = TrainingArguments(
        output_dir="./results",
        learning_rate=2e-5,
        per_device_train_batch_size=8,  # Reduced batch size to prevent memory issues
        per_device_eval_batch_size=8,
        num_train_epochs=3,
        weight_decay=0.01,
        evaluation_strategy="epoch",
        save_strategy="epoch",
        load_best_model_at_end=True,
        push_to_hub=True,
        hub_model_id=f"{username}/bert-imdb-classifier",
        logging_steps=100,
        save_total_limit=2,
        fp16=True,  # Enable mixed precision training
    )

    # Initialize trainer
    print("Initializing trainer...")
    trainer = Trainer(
        model=model,
        args=training_args,
        train_dataset=tokenized_datasets["train"],
        eval_dataset=tokenized_datasets["test"],
        tokenizer=tokenizer,
        data_collator=DataCollatorWithPadding(tokenizer),
        compute_metrics=compute_metrics,
    )

    # Train the model
    print("Starting training...")
    trainer.train()

    # Push to Hub
    print("Pushing model to Hugging Face Hub...")
    trainer.push_to_hub()
    print(f"Model successfully uploaded to: https://huggingface.co/{username}/bert-imdb-classifier")

except Exception as e:
    print(f"An error occurred: {str(e)}")
    sys.exit(1)

if name == "main": main()

Rishabh323/text-classification-finetuning-using-BERT

BERT Text Classification Fine-tuning

Setup

Project Structure

Model

Rishabh323/
text-classification-finetuning-using-BERT