Gautham Venkataraman

@experiencor does adding that as an input increase the performance ? I don't get why that is an input to the model.

> I've never seen this pre-commit project before I'm surprised, maybe it is huge in the Python community cause we use it fairly extensively at work and there are tons...

I made a small POC which you can check out in my fork of cue - https://github.com/gauthsvenkat/cue/blob/master/.pre-commit-hooks.yaml. It turns out all that was needed was just a simple pre-commit-hooks.yaml.

Hey @myitcv ! The way pre-commit works is that the project offering pre-commit hook(s) needs to have a `.pre-commit-hooks.yaml` file at the root of the repository, that describes the hooks...

I agree with @mvdan's point and the following response by @riclima pretty much hits the nail. I checked it out and it works well. Based on this discussion, I'm closing...

@yilonghe Did you manage to find out the reason? My team and I are also running the same experiment and our results are also much lower than the results from...

> @yilonghe @gauthsvenkat Have you found the reason for the abnormal results? Not really. We had to do this as part of a project for a course but haven't really...

Okay, after reading up on "https://arxiv.org/abs/1506.02106" and watershed transforms I kinda understand the losses. How exactly are the blobs computed ? (How are they stored and made use of for...

What does it mean when blobs[ None ] is returned ? I've never seen this anywhere before. (Also counts[None])

I went through all the code in model.py and I don't understand much. Particularly in class FCN8 I understand the computation till fc7 but I get lost in the semantic...