Skip to main content

Command-line interface for Refrakt ML/DL framework

Project description

refrakt_cli

PyPI version

refrakt_cli is a command-line interface for the refrakt_core deep learning and machine learning research framework. It enables rapid, reproducible, and flexible pipeline training by converting CLI arguments into YAML-based experiment configurations, seamlessly integrating with refrakt_core's modular system.


📦 Installation

Install the latest release from PyPI:

pip install refrakt_cli

🚀 What is refrakt_cli?

  • A CLI tool to launch, manage, and override ML/DL training pipelines using YAML configs.
  • Bridges user-friendly command-line workflows with the powerful abstractions of refrakt_core.
  • Supports dynamic hyperparameter overrides, modular experiment configs, and robust logging.

⚙️ Setup

git clone https://github.com/refrakt-hub/refrakt_cli.git
cd refrakt_cli

# (Recommended) Create and activate a virtual environment
python3 -m venv .venv
source .venv/bin/activate

# Install dependencies
pip install .

🧪 Usage: Running Experiments

Run with a config file

refrakt --config path/to/your_config.yaml

Override hyperparameters on-the-fly

refrakt --config path/to/your_config.yaml model.name=ResNet optimizer.lr=0.0005 trainer.epochs=20

Supported CLI Flags

Flag Description
--config Path to YAML config file
--log_type Logging backend: tensorboard, wandb, or both
--debug Enable debug mode with extra verbosity

🧩 Project Structure

src/refrakt_cli/
├── cli.py                # Main CLI entry point
├── helpers/              # Argument parsing, config overrides, pipeline orchestration
│   ├── argument_parser.py
│   ├── config_overrides.py
│   └── pipeline_orchestrator.py
├── hooks/                # Custom hooks (e.g., hyperparameter overrides)
│   └── hyperparameter_override.py
├── utils/                # Utility functions for pipeline management
│   └── pipeline_utils.py
└── __init__.py
  • cli.py: Main CLI logic and entry point.
  • helpers/: Argument parsing, config override logic, and pipeline orchestration.
  • hooks/: Custom hooks for advanced config/hyperparameter handling.
  • utils/: Utility functions for pipeline and config management.

Test files are located in the tests/ directory, mirroring the main module structure.


🤝 Contributing

Contributions, issues, and feature requests are welcome!

Please read our CONTRIBUTING.md for guidelines on how to get started and best practices for contributing to refrakt_cli.


📄 License

This repository is licensed under the MIT License. See LICENSE for full details.


👤 Credits

Akshath Mangudi

If you find issues, raise them. If you learn from this, share it. Built with love and curiosity :)

Project details


Download files

Download the file for your platform. If you're not sure which to choose, learn more about installing packages.

Source Distribution

refrakt_cli-0.1.2.tar.gz (13.8 kB view details)

Uploaded Source

Built Distribution

If you're not sure about the file name format, learn more about wheel file names.

refrakt_cli-0.1.2-py3-none-any.whl (14.4 kB view details)

Uploaded Python 3

File details

Details for the file refrakt_cli-0.1.2.tar.gz.

File metadata

  • Download URL: refrakt_cli-0.1.2.tar.gz
  • Upload date:
  • Size: 13.8 kB
  • Tags: Source
  • Uploaded using Trusted Publishing? No
  • Uploaded via: twine/6.1.0 CPython/3.11.13

File hashes

Hashes for refrakt_cli-0.1.2.tar.gz
Algorithm Hash digest
SHA256 18b63988eb4f225d28d206e52e4f660695ff59ba9a58179cc59d7cf13ee4fd3f
MD5 fbeb6ccbe2f4f5a849496acd6e5d6c21
BLAKE2b-256 620ef6cb9ad948edf19cc95912ffc60b9d4413e54359336fb05ed4d4ffaad9c6

See more details on using hashes here.

File details

Details for the file refrakt_cli-0.1.2-py3-none-any.whl.

File metadata

  • Download URL: refrakt_cli-0.1.2-py3-none-any.whl
  • Upload date:
  • Size: 14.4 kB
  • Tags: Python 3
  • Uploaded using Trusted Publishing? No
  • Uploaded via: twine/6.1.0 CPython/3.11.13

File hashes

Hashes for refrakt_cli-0.1.2-py3-none-any.whl
Algorithm Hash digest
SHA256 8fc95059d11e4fdbdda5ed0e8d4899212e0e452d49e2ddda75f0667be7d6d876
MD5 69ba9e7dbbe5172c5e1f1bd4a4e3254e
BLAKE2b-256 abe571e90561686e7222520d25d5e7fff1b55fbd7ab5cda400b9385beb918ac8

See more details on using hashes here.

Supported by

AWS Cloud computing and Security Sponsor Datadog Monitoring Depot Continuous Integration Fastly CDN Google Download Analytics Pingdom Monitoring Sentry Error logging StatusPage Status page