Getting started with the MacaqueRetina simulator

The simulator should be imported as package. When imported as a package, you have the option of changing parameters at runtime.

First, set the folder for your output data

Navigate to your local MacaqueRetina git repository root.

Open macaqueretina/parameters/core_parameters.yaml and change the model_root_path to point to an existing directory in your system. This is where all output data will be written.

Import as a package and run

Add the 'shell' plugin to your Poetry environment:

poetry self add poetry-plugin-shell

Activate your Poetry-managed environment:

poetry shell

Import into your Python environment:

import macaqueretina as mr
mr.load_parameters()

The mr.load_parameters() is necessary after every import.

Run the following for a quick example:

import macaqueretina as mr
import matplotlib.pyplot as plt
mr.load_parameters()

filename="my_response.gz"

mr.retina_constructor.construct()
mr.stimulus_factory.generate()
mr.retina_simulator.simulate(filename=filename)
mr.viz.show_all_gc_responses_after_simulate()
plt.show()

Variables saved to disk can be accessed with:

 import macaqueretina as mr 

 mr.load_parameters()

 my_data = mr.data_io.load_data(filename)

Troubleshooting

• unclear exception: First, empty your output and stimulus directories and retry. Old constructs, stimuli and cone noise will be reused if present and hash matches. Second, check your environment against pyproject.toml file dependencies. Third, raise an issue in github.