Metadata-Version: 2.1
Name: degann
Version: 0.2.4
Summary: Library for generating artificial neural networks for modeling the behavior of dynamic systems
Home-page: https://github.com/Krekep/degann
Author: Pavel Alimov
Author-email: <pashaalimov@gmail.com>
License: MIT
Keywords: python,ode,differential equation,neural network
Classifier: Development Status :: 3 - Alpha
Classifier: Intended Audience :: Education
Classifier: Intended Audience :: Science/Research
Classifier: License :: OSI Approved :: MIT License
Classifier: Operating System :: POSIX :: Linux
Classifier: Operating System :: Microsoft :: Windows
Classifier: Programming Language :: Python :: 3
Classifier: Topic :: Scientific/Engineering :: Physics
Requires-Python: >=3.10
Description-Content-Type: text/markdown
License-File: LICENSE
Requires-Dist: keras (~=2.12.0)
Requires-Dist: matplotlib (~=3.7.1)
Requires-Dist: numpy (~=1.23.5)
Requires-Dist: scipy (~=1.10.1)
Requires-Dist: tensorflow (~=2.12.0)

# DEGANN

[![Check tests](https://github.com/Krekep/degann/actions/workflows/tests.yml/badge.svg)](https://github.com/Krekep/degann/actions/workflows/tests.yml)
[![License](https://img.shields.io/badge/license-MIT-orange)](https://github.com/Krekep/degann/blob/main/LICENSE)

**DEGANN** is a library generating neural networks for approximating solutions to differential equations. As a backend for working with neural networks, tensorflow is used, but with the ability to expand with your own tools.

**Features**
- Generation of neural networks by parameters.
- Construction of tables with the numerical solution of ordinary differential equations of the first order
- Construction of tables with numerical solution of systems of ordinary differential equations of the first order
- Choosing the Best Neural Network from Several for Fixed Training Parameters
- Iterating over training parameters with choosing the best neural network for each set
- Export neural networks as a function in c++
- Export Neural Networks as a Parameter Set
- Import Neural Networks from a Parameter Set
- Building a dataset with complete training results for approximating the solution of a differential equation for each neural network that participated in training
