🧑💻 Define PatchPredictor
#783
Conversation
shaneahmed
commented
Feb 9, 2024
•
shaneahmed
commented
- Redesigns PatchPredictor engine using the new EngineABC base class.
- The WSIs are now processed using the same code as for the processing the patches using WSI based dataloader.
- The intermediate output is saved as zarr for the WSIs to resolve memory issues.
- The output of model architectures should now be a dictionary.
- The output can be specified as AnnotationStore for visualisation using TIAViz.
for more information, see https://pre-commit.ci
…geAnalytics/tiatoolbox into dev-update-engine-abc-wsi
…ImageAnalytics/tiatoolbox into dev-update-engine-abc-wsi
for more information, see https://pre-commit.ci
for more information, see https://pre-commit.ci
…ImageAnalytics/tiatoolbox into dev-update-engine-abc-wsi
for more information, see https://pre-commit.ci
…ImageAnalytics/tiatoolbox into dev-update-engine-abc-wsi
…ImageAnalytics/tiatoolbox into dev-update-engine-abc-wsi
for more information, see https://pre-commit.ci
…ImageAnalytics/tiatoolbox into dev-update-engine-abc-wsi
for more information, see https://pre-commit.ci
…ls.misc.dict_to_store() method
for more information, see https://pre-commit.ci
Signed-off-by: Shan E Ahmed Raza <13048456+shaneahmed@users.noreply.github.com>
Signed-off-by: Shan E Ahmed Raza <13048456+shaneahmed@users.noreply.github.com>
Signed-off-by: Shan E Ahmed Raza <13048456+shaneahmed@users.noreply.github.com>
Signed-off-by: Shan E Ahmed Raza <13048456+shaneahmed@users.noreply.github.com>
|
@mostafajahanifar and @adamshephard I have pushed the updates as discussed during the meeting. |
Signed-off-by: Shan E Ahmed Raza <13048456+shaneahmed@users.noreply.github.com>
Signed-off-by: Shan E Ahmed Raza <13048456+shaneahmed@users.noreply.github.com>
There was a problem hiding this comment.
Thank you for your excellent work on this PR, Shan. The extent of your contributions is truly impressive.
While the current design appears functional, I believe we can enhance the flexibility of the toolbox by addressing the use of ModelABC for models. To achieve this, I suggest relocating methods that specifically pertain to ModelABC to the engines. This will allow for greater customization and the ability to seamlessly swap engines and models based on user preferences. Predefined (default) functions for common model types can also be implemented to streamline the process.
Anyway, I understand that you may decide to include these changes in a separate PR.
