mango-autoencoder

All notable changes to the mango_autoencoder project will be documented in this file.

The format is based on Keep a Changelog, and this project adheres to Semantic Versioning.

[Unreleased]

[1.0.2] - 2026-10-27

Changed

  • Upgraded tensorflow to 2.20.0

  • Added upper bounds to plotly, polars and scikit-learn

Security

  • Require mango>=1.0.3

  • Refreshed uv.lock to patched versions of vulnerable dependencies (Pillow, keras, protobuf, werkzeug, setuptools, wheel)

[1.0.1] - 2025-09-30

Added

  • Correct mango dependency version

  • Neural network autoencoder implementation

  • Encoder and decoder modules with customizable architectures

  • Anomaly detection using reconstruction error

  • Data processing and sequence handling utilities

  • Visualization tools for autoencoder results

  • Comprehensive logging and error handling

  • Support for various input data formats

Features

  • Flexible neural network architectures

  • Customizable encoder and decoder layers

  • Anomaly detection algorithms

  • Data preprocessing and normalization

  • Model training and evaluation

  • Result visualization and analysis

  • Integration with popular ML frameworks

  • Efficient data processing pipelines