Colab-VsCode-Bridge

A Python package to complete functionalities missing from the VS Code Colab extension.

Overview

This package fills the gaps in VS Code's Colab extension by providing:

• Upload files from localhost to Colab kernel
• Download smaller files from Colab kernel to local PC
• Authenticate to Kaggle
• Upload from kernel to Kaggle datasets

Features

• Upload local files to Colab kernel while connected via VS Code
• Download files from Colab kernel to local machine
• Kaggle authentication integration
• Upload datasets from Colab to Kaggle
• Control dataset visibility (public/private)

Installation

!pip install colab-vscode-bridge

Quick Start

from colab_vscode_bridge import upload_to_colab, ColabFileDownloader, authenticate_kaggle, upload_to_kaggle

# Upload local file to Colab
upload_to_colab(destination_folder='/content/data.csv')

# Download from Colab to local
ColabFileDownloader("data/my_plot.png")

# Authenticate Kaggle
authenticate_kaggle()

# Upload to Kaggle
upload_to_kaggle(
    file_path=file_path,
    dataset_slug="kaggle-dataset-slug",
    dataset_title="kaggle-dataset-title",
    public=False,
)

Requirements

• Python >= 3.8
• VS Code with Colab extension
• Active connection to Colab kernel in VS Code
• Kaggle API credentials (for Kaggle features)

Note

Use the legacy Kaggle API token, as the Colab runtime currently uses Kaggle CLI v1.7.

Usage

Read https://github.com/ranidz/Colab-VsCode-Bridge/blob/main/docs/useDetails.md

Roadmap

• Upload files from localhost to Colab kernel
• Download files from Colab kernel to local PC
• Kaggle authentication
• Upload from kernel to Kaggle
• Batch file operations
• HuggingFace integration
• Google Drive support
• Compression for large files

Contributing

This is an early-stage project. Bug reports and feature requests are welcome via GitHub issues.

License

MIT License

---

Note : This is v0.1.1 - early release bridging VS Code Colab extension gaps.