seq-hybrid-detector 0.1.0

A sequential hybrid anomaly detection framework combining PyTorch GRUs and Isolation Forests.

seq_hybrid_detector

seq_hybrid_detector is a small Python package scaffold for sequence hybrid detection workflows.

What is included

The repository currently provides a clean src/ layout with starter modules for loading data, defining core models, and wiring a simple pipeline.

Project layout

seq_hybrid_detector/
├── LICENSE
├── README.md
├── pyproject.toml
└── src/
	└── seq_hybrid_detector/
		├── __init__.py
		├── data_engine.py
		├── models.py
		└── pipeline.py

Installation

From the project root, install the package in editable mode during development:

pip install -e .

Modules

• data_engine.py contains basic sequence loading and normalization helpers.
• models.py defines lightweight dataclasses for input samples and predictions.
• pipeline.py provides a starter pipeline class for fitting and predicting.

Status

This is a scaffold, not a finished detector. The modules are in place for you to extend with real preprocessing, model training, and scoring logic.