Metadata-Version: 2.2
Name: openretina
Version: 0.3.1
Summary: Open source retina model architectures and training setups
Author: Federico D'Agostino, Thomas Zenkel, Larissa Höfling
Project-URL: Homepage, https://github.com/open-retina/open-retina
Project-URL: Issues, https://github.com/open-retina/open-retina/issues
Requires-Python: >=3.10
Description-Content-Type: text/markdown
License-File: LICENSE
Requires-Dist: IPython
Requires-Dist: jaxtyping
Requires-Dist: matplotlib
Requires-Dist: numpy
Requires-Dist: pandas
Requires-Dist: seaborn
Requires-Dist: scipy
Requires-Dist: tenacity
Requires-Dist: tensorboard
Requires-Dist: torch
Requires-Dist: torchaudio
Requires-Dist: torchvision
Requires-Dist: einops
Requires-Dist: lightning
Requires-Dist: wandb
Requires-Dist: hydra-core
Requires-Dist: omegaconf
Requires-Dist: h5py
Requires-Dist: opencv-python-headless
Requires-Dist: imageio>=2.36
Requires-Dist: moviepy>=2.0.0
Requires-Dist: huggingface_hub>=0.25
Requires-Dist: jupyter
Requires-Dist: ipywidgets
Provides-Extra: dev
Requires-Dist: ruff>=0.9; extra == "dev"
Requires-Dist: mypy>=1.0; extra == "dev"
Requires-Dist: pytest; extra == "dev"
Requires-Dist: types-psutil; extra == "dev"
Requires-Dist: types-tqdm; extra == "dev"
Requires-Dist: types-PyYAML; extra == "dev"
Requires-Dist: types-requests; extra == "dev"
Requires-Dist: nbmake; extra == "dev"
Requires-Dist: jupyterlab; extra == "dev"
Provides-Extra: devmodels
Requires-Dist: neuralpredictors; extra == "devmodels"

# OpenRetina <img src="https://raw.githubusercontent.com/open-retina/open-retina/7aacfa64267930f787b16f24e4bc17047f285c25/assets/openretina_logo.png" align="right" width="120" />

[![Ruff](https://img.shields.io/endpoint?url=https://raw.githubusercontent.com/astral-sh/ruff/main/assets/badge/v2.json)](https://github.com/astral-sh/ruff)
[![mypy](https://img.shields.io/badge/type%20checked-mypy-039dfc)](https://github.com/python/mypy)
[![pytorch](https://img.shields.io/badge/PyTorch_2.0+-ee4c2c?logo=pytorch&logoColor=white)](https://pytorch.org/get-started/locally/)
[![lightning](https://img.shields.io/badge/-Lightning_2.0+-792ee5?logo=pytorchlightning&logoColor=white)](https://pytorchlightning.ai/)
[![hydra](https://img.shields.io/badge/Config-Hydra_1.3-89b8cd)](https://hydra.cc/)

[![huggingface](https://huggingface.co/datasets/huggingface/badges/resolve/main/dataset-on-hf-sm.svg)](https://huggingface.co/datasets/open-retina/open-retina)

Open-source repository containing neural network models of the retina.
The models in this repository are inspired by and partially contain adapted code of [sinzlab/neuralpredictors](https://github.com/sinzlab/neuralpredictors).

## Installation

For development and to have access to Jupyter notebooks:
```
git clone git@github.com:open-retina/open-retina.git
cd open-retina
pip install -e .
```

For normal usage:

```
pip install openretina
```

Test openretina by downloading a model and running a forward pass:
```python
import torch
from openretina.models import *

model = load_core_readout_from_remote("hoefling_2024_base_low_res", "cpu")
responses = model.forward(torch.rand(model.stimulus_shape(time_steps=50)))
```

## Contributing
Before raising a PR please run:
```
# Fix formatting of python files
make fix-formatting
# Run type checks and unit tests
make test-all
```

## Design decisions and structure
With this repository we provide already pre-trained retina models that can be used for inference and intepretability out of the box, and dataloaders together with model architectures to train new models.
For training new models, we rely on [pytorch lightning](https://lightning.ai/docs/pytorch/stable/) in combination with [hydra](https://hydra.cc/docs/intro/) to manage the configurations for training and dataloading.

The openretina package is structured as follows:
- modules: pytorch modules that define layers and losses
- models: pytorch lightning models that define models that can be trained and evaluated (i.e. models from specific papers)
- data_io: dataloaders to manage access of data to be used for training
- insilico: Methods perform _insilico_ experiments with above models
    - stimulus_optimization: optimize inputs for neurons of above models according to interpretable objectives (e.g. most exciting inputs)
    - future options: gradient analysis, data analysis
- utils: Utility functions that are used across above submodules


## Related papers

The Core + Readout model was developed in the paper [A chromatic feature detector in the retina signals visual context changes](https://elifesciences.org/articles/86860). All datasets used in openretina are shared under a CC-BY Share-Alike license, and we acknowledge and credit the original sources below:
- hoefling_2024: Originally published by Höfling et al. (2024), eLife: [A chromatic feature detector in the retina signals visual context changes](https://doi.org/10.7554/eLife.86860).
- karamanlis_2024: Originally published by Karamanlis et al. (2024), Nature: [Nonlinear receptive fields evoke redundant retinal coding of natural scenes](https://doi.org/10.1038/s41586-024-08212-3)
- maheswaranathan_2023: Originally published by Maheswaranathan et al. (2023), Neuron: [Interpreting the retinal neural code for natural scenes: From computations to neurons](https://doi.org/10.1016/j.neuron.2023.06.007)

The paper [Most discriminative stimuli for functional cell type clustering](https://openreview.net/forum?id=9W6KaAcYlr) explains the discriminatory stimulus objective we showcase in [notebooks/most_discriminative_stimulus](https://github.com/open-retina/open-retina/blob/main/notebooks/most_discriminative_stimulus.ipynb).
