Metadata-Version: 2.1
Name: pclib
Version: 2.0.0b2
Summary: A torch-like package for building Predictive Coding Neural Networks.
Author: Joe Griffith
Author-email: joeagriffith@gmail.com
Keywords: python,neural networks,deep learning,predictive coding
Classifier: Development Status :: 4 - Beta
Classifier: Intended Audience :: Science/Research
Classifier: Programming Language :: Python :: 3
Classifier: Operating System :: Unix
Classifier: Operating System :: MacOS :: MacOS X
Classifier: Operating System :: Microsoft :: Windows
Description-Content-Type: text/markdown
License-File: LICENSE


# PCLib





PCLib is a python package with a torch-like API for building and training Predictive Coding Networks.<br> 

Documentation can be found [here](https://joeagriffith.github.io/pclib/).



The package includes a fully-connected layer implementation, as well as a convolutional one. Both are customisable and can be used together or separately for building neural networks.



The package also includes a helper class for constructing fully-connected PCNs. This class has been designed to be extremely customiseable such that the network it builds can be used in a wide range of tasks: supervised/unsupervised, classic/inverted, etc. There is also a CNN class, however it is not customisable in shape. For more detailed explanations, please see the documentation.



## Installation

```

pip install pclib

```



## Example usage



In the examples folder you will find two different classification tasks which demonstrate the usage of this package.
