Unsolved references in train.py and test_online.py

[dahyun-kang]

Dear authors.

Hello,
I found that train.py shows many unsolved references. For example,
optimizer = optim.SGD([{'params': get_10x_lr_params(model), 'lr': args.lr}], lr=args.lr, momentum=args.momentum, weight_decay=args.weight_decay)
and
from Network import network.
References for get_10x_lr_params and Network are missing.

test_online.py, which you refered to test.py at README, is not working either. It uses dataset_val, while you have applied multi-scale testing of dataset_test according to this issue.

I assume these are typos from refactoring.
Could you please double-check if your code runs?

Thank you.

[Wi-sc]

Sorry for that. The function get_10x_lr_params should be get_lr_params and Network should be Brinet. I have corrected them.

[lewnus]

Sorry for that. The function get_10x_lr_params should be get_lr_params and Network should be Brinet. I have corrected them.

The test_online.py has not yet been corrected.

I try to correct it by myself and the MIoU of the model you provided can only reach 64.6 in pascal voc fold1, which is 67.20 in the paper. Could you please provide a correct test script?