Introducing topography in convolutional neural networks

Companion repository to "Introducing topography in convolutional neural networks".

• Provides the topography Python package.
• Examples and Python scripts to reproduce the results from the paper in notebooks and scripts.
• Fully documented and tested.
• The TopographicModel and TopographicLoss can be used with any PyTorch model with at least one nn.Conv2d layer.

Installation

Clone this repository:

git clone https://github.com/bootphon/topography.git
cd topography

Create the conda environment and install the package:

conda create -n topography pip python=3.10
conda activate topography
pip install .

You can also install in editable mode (pip must be at least at version 21.3 in this case):

conda create -n topography-dev pip=22 python=3.10
conda activate topography-dev
pip install -e .[dev,testing]