#!/usr/bin/env python3 """ Upload Bengali AI model to Hugging Face Hub Repository: megharudushi/Sheikh """ import os from huggingface_hub import HfApi, create_repo, upload_folder from huggingface_hub.utils import RepositoryNotFoundError def upload_model_to_hf(): """Upload the Bengali AI model to Hugging Face""" print("šŸš€ Uploading Bengali AI to Hugging Face Hub...") print("=" * 50) # Initialize Hugging Face API api = HfApi() # Repository details repo_id = "megharudushi/Sheikh" local_dir = "./ready_bengali_ai" # Check if local directory exists if not os.path.exists(local_dir): print(f"āŒ Error: Directory {local_dir} not found!") return False try: # Login prompt print("šŸ“ Please login to Hugging Face Hub") print("This will open your browser for authentication...") # Login (this will prompt for token) api.login() print(f"šŸ”— Creating repository: {repo_id}") # Create repository try: repo_url = create_repo( repo_id=repo_id, exist_ok=True, repo_type="model" ) print(f"āœ… Repository created/accessed: {repo_url}") except Exception as e: print(f"āš ļø Repository creation issue: {e}") print("Continuing with upload...") print(f"šŸ“¤ Uploading model files from {local_dir}...") # Upload all files upload_folder( folder_path=local_dir, repo_id=repo_id, commit_message="Upload complete Bengali AI model with tokenizer and config" ) print("šŸŽ‰ Model uploaded successfully!") print(f"🌐 View your model at: https://huggingface.co/{repo_id}") # Show repository info print("\nšŸ“Š Repository Information:") print(f" Repository ID: {repo_id}") print(f" Local files: {len(os.listdir(local_dir))} files") print(f" Model size: {os.path.getsize(f'{local_dir}/model.bin') / (1024*1024*1024):.2f} GB") return True except Exception as e: print(f"āŒ Upload failed: {e}") print("\nšŸ”§ Troubleshooting:") print("1. Make sure you're logged in to Hugging Face") print("2. Check your internet connection") print("3. Verify repository permissions") return False def create_model_card(): """Create a model card for the repository""" model_card = """# Bengali AI Model ## Model Description This is a Bengali (Bangla) language AI model trained on instruction-following datasets. The model can understand and respond to Bengali text queries. ## Model Details - **Base Model**: microsoft/DialoGPT-medium - **Language**: Bengali (Bangla) - **Parameters**: 355M - **Training Data**: Alpaca Bangla dataset - **Model Size**: 1.4GB ## Usage ```python from transformers import AutoTokenizer, AutoModelForCausalLM import torch # Load model tokenizer = AutoTokenizer.from_pretrained("megharudushi/Sheikh") model = AutoModelForCausalLM.from_pretrained("megharudushi/Sheikh") # Generate response input_text = "বাংলাদেশের ą¦°ą¦¾ą¦œą¦§ą¦¾ą¦Øą§€ কী?" inputs = tokenizer.encode(input_text, return_tensors="pt") outputs = model.generate(inputs, max_length=150, temperature=0.8) response = tokenizer.decode(outputs[0], skip_special_tokens=True) print(response) ``` ## Training Details - Trained on Bengali instruction-following data - Fine-tuned for educational and general knowledge queries - Supports Bengali language understanding and generation ## License Please check the repository for license information. ## Citation If you use this model in your research, please cite the original base model and dataset sources. """ with open("./ready_bengali_ai/README.md", "w", encoding="utf-8") as f: f.write(model_card) print("šŸ“„ Created model card (README.md)") if __name__ == "__main__": print("šŸ‡§šŸ‡© BANGLI AI HUGGING FACE UPLOAD") print("=" * 40) # Create model card create_model_card() # Upload to Hugging Face success = upload_model_to_hf() if success: print("\nšŸŽ‰ UPLOAD COMPLETE!") print("Your Bengali AI model is now live on Hugging Face Hub!") print("Repository: https://huggingface.co/megharudushi/Sheikh") else: print("\nāŒ Upload failed. Please check the error messages above.")