481-Rotation of output images in prostate_mri_anatomy
The transformations applied to the test data during preprocessing are not reversed during postprocessing. Thus, the resulting predictions have a different rotation compared to the original images and labels. I added an 'invertd()' function that reverses all the traceable transformations in the preprocessing.
Running the inference on a PC without a GPU leads to an error because the handlers function 'CheckpointLoader' attempts to load the model onto the GPU. I set the map_location argument to torch.device(device) so that it automatically loads the model onto the CPU when a GPU is not available.
Additionally, I changed some of the config directories as they were referring to the validation dataset rather than the test dataset.
Status
Ready
Please ensure all the checkboxes:
[x] Codeformat tests passed locally by running ./runtests.sh --codeformat.
[x] Update version and changelog in metadata.json if changing an existing bundle.
[x] Please ensure the naming rules in config files meet our requirements (please refer to: CONTRIBUTING.md).
[x] Ensure versions of packages such as monai, pytorch and numpy are correct in metadata.json.
[x] Descriptions should be consistent with the content, such as eval_metrics of the provided weights and TorchScript modules.
[x] Files larger than 25MB are excluded and replaced by providing download links in large_file.yml.
[x] Avoid using path that contains personal information within config files (such as use /home/your_name/ for "bundle_root").
wyli
commented
10 months ago
Description
481-Rotation of output images in prostate_mri_anatomy
Status
Ready
Please ensure all the checkboxes:
./runtests.sh --codeformat.versionandchangeloginmetadata.jsonif changing an existing bundle.CONTRIBUTING.md).monai,pytorchandnumpyare correct inmetadata.json.eval_metricsof the provided weights and TorchScript modules.large_file.yml./home/your_name/for"bundle_root").