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

关于optimizer和frozen_stage的设置 #91

Open
pingyang97 opened this issue Dec 15, 2023 · 2 comments
Open

关于optimizer和frozen_stage的设置 #91

pingyang97 opened this issue Dec 15, 2023 · 2 comments

Comments

@pingyang97
Copy link

您好,关于optimizer的设置,您这边设置了 'img_backbone': dict(lr_mult=0.1),请问这个有什么特殊的说法吗?

optimizer = dict(
type='AdamW',
lr=2e-4,
paramwise_cfg=dict(
custom_keys={
'img_backbone': dict(lr_mult=0.1),
}),
weight_decay=0.01)

在img_backbone这边,为什么选择frozen_stages=1呢?
img_backbone=dict(
type='ResNet',
depth=101,
num_stages=4,
out_indices=(1,2,3),
frozen_stages=1,
norm_cfg=dict(type='BN2d', requires_grad=False),
norm_eval=True,
style='caffe',
#with_cp=True, # using checkpoint to save GPU memory
dcn=dict(type='DCNv2', deform_groups=1, fallback_on_stride=False), # original DCNv2 will print log when perform load_state_dict
stage_with_dcn=(False, False, True, True)),

非常感谢您的阅读,期待能够尽快得到您会回复~

@weiyithu
Copy link
Owner

同学你好,这块是直接参考bevformer里的超参的,没有改过这个参数

@pingyang97
Copy link
Author

同学你好,这块是直接参考bevformer里的超参的,没有改过这个参数

收到,感谢回复~

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

2 participants