Including distinction between macaque and human regions

brain-score / model-tools

Helper functions to extract model activations and translate from Machine Learning to Neuroscience

MIT License

8 stars 27 forks source link

Open ernestoBocini opened 11 months ago

ernestoBocini commented 11 months ago

This commit introduces new standardized region benchmarks for model mapping. It includes:

The addition of human's and macaque's specific STANDARD_REGION_BENCHMARKS dictionary to map regions to benchmark models.
Introduction of the 'whole_brain' entry to encompass all available layers for a given model. Mainly useful for reproducing Algonauts.