search

AllenCell / benchmarking_representations

Benchmarking different methods for extracting unsupervised representations from images
Other
0 stars 0 forks source link

Fix the naming for the config files #30

Closed fatwir closed 1 week ago

fatwir commented 1 week ago

Issue

There was a slight mismatch in the naming of the datapaths in the experiment config files for the npm1_variance and npm1_perturb experiments.

Changes

Made the below changes for the npm1 experiments (for the data config overrides)

  1. image_sdf.yaml to classical_image_sdf.yaml
  2. image_seg.yaml to classical_image_seg.yaml

This change was made for the other_punctuate_pc_so3 experiment (for the data config override)

  1. pc_intensity.yaml to pc_intensity_structurenorm.yaml

Testing

You will need to follow steps to train the models from the main readme file (just that the experiments will now be those that the proposed changes aim to fix)

fatwir commented 1 week ago

This PR has been taken care of with the next updated PR (https://github.com/AllenCell/benchmarking_representations/pull/31)