Igel is a Python-based machine learning tool designed to democratize ML by enabling users to train, evaluate, and deploy models without writing code. It caters to both technical and non-technical users seeking to rapidly prototype ML solutions or leverage AutoML capabilities.

How It Works

Igel abstracts complex ML workflows into a configuration-driven approach. Users define their tasks, data sources, preprocessing steps, model choices, and hyperparameters in a YAML or JSON file. The tool then executes these instructions via a command-line interface (CLI), supporting a wide range of scikit-learn models, AutoML for image and text tasks, and various data formats (CSV, Excel, JSON, raw data folders).

Quick Start & Requirements

• Install via pip: pip install -U igel
• Requires Python.
• Documentation: https://igel.readthedocs.io/

Highlighted Details

• Supports regression, classification, and clustering tasks.
• Offers AutoML for image and text classification/regression.
• Handles diverse data inputs including CSV, Excel, JSON, and raw data folders.
• Includes a CLI for training, evaluation, prediction, and model serving (REST API via FastAPI).
• Provides an optional GUI for terminal-averse users.

Maintenance & Community

• Project actively maintained by Nidhal Baccouri.
• Contributions are welcomed via pull requests and issues.
• Links to GitHub and Twitter are provided for community engagement.

Licensing & Compatibility

• MIT License.
• Permissive license suitable for commercial use and integration with closed-source projects.

Limitations & Caveats

• AutoML features are computationally expensive.
• When serving models, users must manually parse data into JSON for API requests, though a Python client example is provided.