Metadata-Version: 2.4
Name: dantabnn
Version: 0.1.0
Summary: Dual-Attention Neural Networks for tabular data classification and regression
Author-email: Aleksandr Rybin <rybinalex2023@gmail.com>
License: MIT
Project-URL: Homepage, https://github.com/yourusername/dantabnn
Project-URL: Repository, https://github.com/yourusername/dantabnn.git
Project-URL: Issues, https://github.com/yourusername/dantabnn/issues
Project-URL: Changelog, https://github.com/yourusername/dantabnn/blob/main/CHANGELOG.md
Project-URL: Documentation, https://yourusername.github.io/dantabnn
Keywords: deep-learning,tabular-data,attention-mechanism,neural-network,pytorch,machine-learning,classification,regression,optuna,danet
Classifier: Development Status :: 3 - Alpha
Classifier: Intended Audience :: Developers
Classifier: Intended Audience :: Science/Research
Classifier: License :: OSI Approved :: MIT License
Classifier: Operating System :: OS Independent
Classifier: Programming Language :: Python :: 3
Classifier: Programming Language :: Python :: 3.9
Classifier: Programming Language :: Python :: 3.10
Classifier: Programming Language :: Python :: 3.11
Classifier: Programming Language :: Python :: 3.12
Classifier: Topic :: Scientific/Engineering :: Artificial Intelligence
Classifier: Topic :: Software Development :: Libraries :: Python Modules
Requires-Python: >=3.9
Description-Content-Type: text/markdown
License-File: LICENSE
Requires-Dist: torch>=2.0.0
Requires-Dist: pandas>=2.0.0
Requires-Dist: numpy>=1.24.0
Requires-Dist: scikit-learn>=1.3.0
Requires-Dist: optuna>=3.5.0
Provides-Extra: dev
Requires-Dist: pytest>=7.0; extra == "dev"
Requires-Dist: pytest-cov>=4.0; extra == "dev"
Requires-Dist: black>=23.0; extra == "dev"
Requires-Dist: ruff>=0.1.0; extra == "dev"
Requires-Dist: mypy>=1.0; extra == "dev"
Requires-Dist: pre-commit>=3.0; extra == "dev"
Provides-Extra: docs
Requires-Dist: mkdocs>=1.5.0; extra == "docs"
Requires-Dist: mkdocs-material>=9.0; extra == "docs"
Requires-Dist: mkdocstrings[python]>=0.23.0; extra == "docs"
Provides-Extra: tune
Requires-Dist: optuna-dashboard>=0.13.0; extra == "tune"
Dynamic: license-file

# DANet Pipeline

A PyTorch-based deep learning pipeline for tabular data classification and regression, featuring **Dual-Attention Networks (DANet)** with feature-wise self-attention and optional sample-wise attention mechanisms.

## Features

- **Dual-Attention Architecture**: Feature-level self-attention for learning complex feature interactions, plus optional sample-level attention
- **End-to-End Pipeline**: Handles preprocessing (scaling, encoding), training, evaluation, hyperparameter tuning, and model persistence
- **Hyperparameter Optimization**: Built-in Optuna integration with Bayesian optimization and early pruning
- **Production Ready**: Save/load full pipelines with preprocessing artifacts, reproducible training with seed control
- **Extensible Design**: Abstract base class makes it easy to add new task types (regression, binary/multiclass classification)

## Installation

```bash
pip install danet-pipeline
