Advanced AI Explainability for computer vision. Support for CNNs, Vision Transformers, Classification, Object detection, Segmentation, Image similarity and more.
After fixing, the code will run successfully for CPU and GPU scenario, following is my test result:
$ python3 usage_examples/swinT_example.py --use-cuda
Using GPU for acceleration
100%|████████████████████████████████████████████████████████████████████████████████████████████████████████| 32/32 [00:07<00:00, 4.43it/s]
$ python3 usage_examples/swinT_example.py
Using CPU for computation
100%|████████████████████████████████████████████████████████████████████████████████████████████████████████| 32/32 [02:01<00:00, 3.78s/it]
Based on https://github.com/jacobgil/pytorch-grad-cam/issues/504,
use_cuda
already removed in https://github.com/jacobgil/pytorch-grad-cam/commit/00711a20b7a8dfb6d6d1d892216997daeeee6288, so user will get error when executing swin transformer example code currently.After fixing, the code will run successfully for CPU and GPU scenario, following is my test result: