search

zhengchen1999 / CAT

PyTorch code for our NeurIPS 2022 paper "Cross Aggregation Transformer for Image Restoration"
Apache License 2.0
121 stars 8 forks source link

Geometric self-ensemble #2

Closed joshyZhou closed 1 year ago

joshyZhou commented 1 year ago

Hi, Thanks for your great work! Would you like to share another version code with self-ensemble strategy, since it is mentioned in your paper.

zhengchen1999 commented 1 year ago

Hi. Thanks for your interest in our work. We have added the code for self-ensemble (cat_se_model.py), and provided the test YML of CAT-R (SR, x2, x3, x4: test_CAT_R_sr_se_x2.yml, test_CAT_R_sr_se_x3.yml, test_CAT_R_sr_se_x4.yml). You can run the following scripts:

python basicsr/test.py -opt options/test/test_CAT_R_sr_se_x2.yml
python basicsr/test.py -opt options/test/test_CAT_R_sr_se_x3.yml
python basicsr/test.py -opt options/test/test_CAT_R_sr_se_x4.yml

If you want to test other models with self-ensemble strategy, you can modify the corresponding testing YML: model_type: CATModle, selfensemble_testing: True, similar to test_CAT_R_sr_se_x2.yml. If you have any other problem, please let us know. Thanks.

joshyZhou commented 1 year ago

Thanks for your quickly reply! My problem has been solved.