You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
The output images generated by the prediction module are not as expected. Here is the snapshot of the prediction from the 2D infection classifier trained for BJ5 cells generated using the test module:
This is the ground truth image of the infection score:
Here is the output from the prediction module: @ziw-liu, where do you think this can originate, and can you please help me debug this? Thanks!
The text was updated successfully, but these errors were encountered:
Sure @mattersoflight . The problem seems to come from the prediction module in my code here. The same line in the test module here gives the right result (first image above).
I will inspect the source fed to the model in each of these modules and update the issue.
I am closing the issue here. The issue has been resolved.
There were two issues here. The issue with the prediction was already resolved when I recreated a fresh conda environment using a fresh installation of VisCy.
Secondly, the issue seemed to persist as napari did not show me the new results. The wrong result kept showing even when I reopened a fresh napari window and created new zarr stores. On loading the same zarr today with a new OnDemand session it shows the expected result.
The output images generated by the prediction module are not as expected. Here is the snapshot of the prediction from the 2D infection classifier trained for BJ5 cells generated using the test module:
This is the ground truth image of the infection score:
Here is the output from the prediction module:
@ziw-liu, where do you think this can originate, and can you please help me debug this? Thanks!
The text was updated successfully, but these errors were encountered: