Documenting this weird bug I ran into while working on microscopy files.
During the training/validation/testing steps, I get good results and predictions are visually OK. However, when trying to segment these images using the --segment command, the results are consistently bad. To make sure the --segment command was faulty, I tried segmenting an image using the --test command and a dummy gt. The results were fine so the problem is definitely coming from --segment.
To illustrate this, here is an output myelin segmentation using --test:
And here is the output of the same sample using --segment:
This is such a weird behavior, because as we can see, the --segment command does not produce complete rubbish. It seems the model identifies the right elements, produces a segmentation and then outputs the edges of the segmentation.
Moreover, I am not able to pinpoint how to reproduce this. @mariehbourget trained models on other microscopy datasets and she doesn't encounter anything like this. When I use her recently trained models, everything works as expected. This would suggest that something went wrong when I trained my models but my config files look almost identical to the ones she used.
Yesterday, I trained 2 new models on an up-to-date master branch (some PRs were merged in the meantime) and nothing changed.
Documenting this weird bug I ran into while working on microscopy files.
During the training/validation/testing steps, I get good results and predictions are visually OK. However, when trying to segment these images using the
--segment
command, the results are consistently bad. To make sure the--segment
command was faulty, I tried segmenting an image using the--test
command and a dummy gt. The results were fine so the problem is definitely coming from--segment
.To illustrate this, here is an output myelin segmentation using
--test
:And here is the output of the same sample using
--segment
:This is such a weird behavior, because as we can see, the
--segment
command does not produce complete rubbish. It seems the model identifies the right elements, produces a segmentation and then outputs the edges of the segmentation.Moreover, I am not able to pinpoint how to reproduce this. @mariehbourget trained models on other microscopy datasets and she doesn't encounter anything like this. When I use her recently trained models, everything works as expected. This would suggest that something went wrong when I trained my models but my config files look almost identical to the ones she used.
Yesterday, I trained 2 new models on an up-to-date master branch (some PRs were merged in the meantime) and nothing changed.