Cortical Tools

Cortical Tools offers dataset-specific Python clients for interacting with several CAVE datasets, in particular the Microns65 dataset and V1dd dataset. It wraps CAVEclient and related tooling with a more task-focused design than the infrastructure-focused CAVEclient. The public entry point is a preconfigured client for datasets.

• Get started with the Examples

Who can use it?

Cortical Tools can be used if you are working with either the publicly accessible datastacks minnie65_public or v1dd_public. In addition, if you are using the production datastacks for these same datastacks, you can use those as well.

Installation

pip install cortical-tools

(Really, you should probably be using uv and doing uv add cortical-tools to your project.)

Cortical Tools is opinionated, and installs not only CAVEclient, but a number of related tools like nglui, ossify, cloudvolume, and standard_transform that are commonly used in conjunction with CAVE datasets.

The mesh-to-skeleton vertex lookup tooling (VertexAssigner) depends on gpytoolbox, which can be slow to build. It is an optional extra, installed with:

pip install "cortical-tools[vertex_lookup]"

Quick start

from cortical_tools.datasets.minnie_public import client
client

This will return a MinniePublicClient object that is ready to use. For example, client.neuroglancer_url() will return a URL that opens the Neuroglancer viewer at the specified location. Your standard CAVEclient functionality is available under client.cave.

Who made this?

This is a totally unofficial tool that Casey Schneider-Mizell made to use for himself, and maybe you'll find some useful things here too.