brainglobe / brainglobe-workflows

Workflows that utilise BrainGlobe tools to perform data analysis and visualisation.
BSD 3-Clause "New" or "Revised" License
9 stars 2 forks source link

[Feature] A benchmark for classification on a large dataset #99

Open alessandrofelder opened 2 months ago

alessandrofelder commented 2 months ago

Is your feature request related to a problem? Please describe. We currently don't have a good handle on how classification performs on large dataset.

Describe the solution you'd like Published results of how the classification performs with associated specs.

Describe alternatives you've considered None so far.

Additional context A user reports that classifying 20 slices (with LOTS of cell candidates, whole stack of ~4500 slices has ~3.5 Million cell candidates) takes 2-7 hours, and more on Windows 🤔

alessandrofelder commented 1 month ago

How classification scales with number of cell candidates would be particulary interesting to explore in the context of people using cell finder on datasets with millions of cells