Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Loss Curve #6

Open
lzyhha opened this issue Apr 6, 2021 · 6 comments
Open

Loss Curve #6

lzyhha opened this issue Apr 6, 2021 · 6 comments

Comments

@lzyhha
Copy link

lzyhha commented Apr 6, 2021

Thanks for your great work.
Could you please show your loss curve in traing? Or could you let me know the loss value of PixPro at the end of training?

@YasserdahouML
Copy link

Hello, I have the same quetion, in fact, I trained the model on the COCO dataset, the loss seems to quickly goes to -2 and stagtnates there

@shiyongde
Copy link

same question

@shiyongde
Copy link

i got loss_pixpro: -2.7215 , it's confuse

@etasnadi
Copy link

Hello, I have the same quetion, in fact, I trained the model on the COCO dataset, the loss seems to quickly goes to -2 and stagtnates there

The same here (I trained with the default parameters on imagenette[a subset of imagenet], but observed the same behaviour with my own dataset).

@MaitaYuki
Copy link

@etasnadi , @shiyongde , @YasserdahouML ,

I trained on my own dataset and got the initial losses at the first epoch:
0.089 ([1/100][0/466]), -2.872([1/100][100/466]), -3.116([1/100][200/466]), -3.088([1/100][300/466]), -3.031([1/100][400/466])
The losses at the 100th epoch:
-2.1([1/100][0/466]), -2.2([1/100][100/466]), -2.195([1/100][200/466]), -2.185([1/100][300/466]), -2.161([1/100][400/466])

How about yours?

@tonitopark
Copy link

Any answer on this yet?

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

No branches or pull requests

6 participants