upload.py

#!/usr/bin/env python3
# -*- coding: utf-8 -*-
"""
Script to upload robot trajectory dataset to Hugging Face Hub
"""

import os
import sys
from pathlib import Path
from huggingface_hub import HfApi, create_repo, upload_folder
from huggingface_hub.utils import HfHubHTTPError

def upload_dataset_to_hf(
    local_folder_path=".",
    repo_id=None,
    token=None,
    private=False,
    repo_type="dataset"
):
    """
    Upload a local folder to Hugging Face as a dataset.
    
    Args:
        local_folder_path: Path to the local folder to upload (default: current directory)
        repo_id: HuggingFace repo ID (format: 'username/dataset-name')
        token: HuggingFace API token
        private: Whether to make the repo private (default: False)
        repo_type: Type of repo (default: 'dataset')
    """
    
    # Initialize the API
    api = HfApi(token=token)
    
    # Get the absolute path
    local_folder_path = Path(local_folder_path).absolute()
    
    print(f"📁 Preparing to upload from: {local_folder_path}")
    
    # Count files to upload (optional, for progress indication)
    total_files = sum(1 for _ in local_folder_path.rglob("*") if _.is_file())
    print(f"📊 Found {total_files} files to upload")
    
    try:
        # Create the repository if it doesn't exist
        print(f"🔧 Creating repository: {repo_id}")
        create_repo(
            repo_id=repo_id,
            token=token,
            private=private,
            repo_type=repo_type,
            exist_ok=True  # Don't error if repo already exists
        )
        print(f"✅ Repository ready: {repo_id}")
        
        # Upload the entire folder
        print(f"📤 Starting upload...")
        print(f"   This may take a while depending on the size of your dataset...")
        
        # Upload folder with all its contents
        upload_folder(
            folder_path=str(local_folder_path),
            repo_id=repo_id,
            repo_type=repo_type,
            token=token,
            ignore_patterns=["*.pyc", "__pycache__", ".git", ".DS_Store"],  # Ignore common unwanted files
            commit_message="Upload robot trajectory dataset"
        )
        
        print(f"✅ Successfully uploaded dataset to: https://huggingface.co/datasets/{repo_id}")
        
    except HfHubHTTPError as e:
        print(f"❌ HTTP Error occurred: {e}")
        sys.exit(1)
    except Exception as e:
        print(f"❌ An error occurred: {e}")
        sys.exit(1)

def main():
    """Main function to run the upload script."""
    
    # Configuration - MODIFY THESE VALUES
    # ====================================
    
    # Your Hugging Face username and desired dataset name
    REPO_ID = "ASethi04/robot-trajectories-dataset"  # <- CHANGE THIS
    
    # Your Hugging Face token (get it from https://huggingface.co/settings/tokens)
    # You can also set this as an environment variable: HF_TOKEN
    HF_TOKEN = os.getenv("HF_TOKEN", None)  # <- SET THIS or use environment variable
    
    # Path to your local dataset folder (. for current directory)
    LOCAL_FOLDER = "."
    
    # Whether to make the dataset private
    PRIVATE = False  # Set to True if you want a private dataset
    
    # ====================================
    
    # Validate inputs
    if not HF_TOKEN:
        print("❌ Error: Hugging Face token not provided!")
        print("Please either:")
        print("  1. Set the HF_TOKEN environment variable: export HF_TOKEN='your-token-here'")
        print("  2. Modify the HF_TOKEN variable in this script")
        print("\nGet your token from: https://huggingface.co/settings/tokens")
        sys.exit(1)
    
    if REPO_ID == "your-username/robot-trajectories-dataset":
        print("❌ Error: Please update the REPO_ID with your Hugging Face username and desired dataset name!")
        print("Format: 'username/dataset-name'")
        sys.exit(1)
    
    # Confirm before uploading
    print("=" * 50)
    print("📋 Upload Configuration:")
    print(f"   Repository: {REPO_ID}")
    print(f"   Local Path: {Path(LOCAL_FOLDER).absolute()}")
    print(f"   Private: {PRIVATE}")
    print("=" * 50)
    
    response = input("\n⚠️  Do you want to proceed with the upload? (yes/no): ").lower().strip()
    
    if response not in ['yes', 'y']:
        print("❌ Upload cancelled.")
        sys.exit(0)
    
    # Run the upload
    upload_dataset_to_hf(
        local_folder_path=LOCAL_FOLDER,
        repo_id=REPO_ID,
        token=HF_TOKEN,
        private=PRIVATE
    )

if __name__ == "__main__":
    main()