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
Unable to reproduce seg_hrnet_w18_small_v1
#51
Comments
Hi, sorry for the late reply. The model 'hrnet_w18_for_mb' is the same as hrnet_w18_small_v1. |
No problem, thanks for replying! I noticed that the config in your training log includes |
For posterity, I was unable to reproduce Edit:
@sunke123 would you happen to have the mean/std of your runs? |
seg_hrnet_w18_small_v1
I only ran it 1 once. |
@sunke123
I realize I didn't initialize from the pretrained Imagenet weights, as stated in the README. Sorry for the bother -- I'll try again and update here. |
@sunke123 sorry for bothering, that fixed it! haha. I appreciate your help. These are probably the first reproducible results I've ever seen. The code is clean, the results are reproducible... couldn't ask for more. |
Congrats! hh~ |
Thanks for 27488d4, the configuration file is very helpful. With that said, training on 4 GPUs as prescribed, I'm unable to reproduce Cityscapes validation accuracy of 70.3% (attained 65.21%) https://github.com/HRNet/HRNet-Semantic-Segmentation#small-models.
Is
https://github.com/HRNet/HRNet-Semantic-Segmentation/blob/master/experiments/cityscapes/seg_hrnet_w18_small_v1_512x1024_sgd_lr1e-2_wd5e-4_bs_12_epoch484.yaml
verbatim the file used to produce 70.3% or does it need further hyperparameter tuning? (I'm on thepytorch-v1.1
branch.)In case it's helpful (although I'm sure this isn't informative), here are the cIoUs for the w18-v1 retrained model:
The text was updated successfully, but these errors were encountered: