Open chonghaozhan opened 1 year ago
Hi, thanks for your interest.
When inference with TLC, you need to modify your Model
.
In details, you need to create a new LocalModel
which inherit Local_Base
and YourModel
. For example, for MPRNet
w/ TLC (named MPRNetLocal
), we need to:
1、import MPRNet
2、create MPRNetLocal
which inherit Local_Base
and MPRNet
3、set training size
4、inference with MPRNetLocal
with proper base_size
For MPRNet, the training size is (1, 3, 256, 256) and the base size is 1.5 times lager than training size (i.e., 384). You can tu ne this hyper-parameter in your dataset.
Note: The current Local_Base only supports global avgpool and Instance Normalization.
Thanks for your reply.
Thank you for your sharing. If I have a trained dehazing network ready, How can I incorporate TLC into my test code