Skip to content

Conversation

@shaneahmed
Copy link
Member

  • Replace deprecated input pretrained with weights according to the new API.

- Replace `pretrained` with `weights` as this is replaced in torch by the new API.

Signed-off-by: Shan E Ahmed Raza <[email protected]>
- `weights` variable was not used

Signed-off-by: Shan E Ahmed Raza <[email protected]>
@shaneahmed shaneahmed added the refactoring Code Refactoring label Jun 15, 2023
@shaneahmed shaneahmed added this to the Release v1.4.1 milestone Jun 15, 2023
@shaneahmed shaneahmed self-assigned this Jun 15, 2023
@shaneahmed shaneahmed requested a review from measty June 15, 2023 12:41
- `DEFAULT` weights might be case-sensitive

Signed-off-by: Shan E Ahmed Raza <[email protected]>
@codecov
Copy link

codecov bot commented Jun 15, 2023

Codecov Report

Merging #621 (013ce29) into develop (971ef74) will increase coverage by 0.00%.
The diff coverage is 100.00%.

@@           Coverage Diff            @@
##           develop     #621   +/-   ##
========================================
  Coverage    99.78%   99.78%           
========================================
  Files           64       64           
  Lines         6819     6820    +1     
  Branches      1362     1362           
========================================
+ Hits          6804     6805    +1     
  Misses           7        7           
  Partials         8        8           
Impacted Files Coverage Δ
tiatoolbox/models/architecture/vanilla.py 100.00% <100.00%> (ø)
tiatoolbox/tools/registration/wsi_registration.py 100.00% <100.00%> (ø)

📣 We’re building smart automated test selection to slash your CI/CD build times. Learn more

Copy link
Collaborator

@measty measty left a comment

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Looks good

@shaneahmed shaneahmed merged commit 25f5f3d into develop Jun 16, 2023
@shaneahmed shaneahmed deleted the dev-update-pretrained-to-weights branch June 16, 2023 05:54
@shaneahmed shaneahmed mentioned this pull request Dec 15, 2023
shaneahmed added a commit that referenced this pull request Dec 15, 2023
## 1.5.0 (2023-12-15)

### Major Updates and Feature Improvements

- Adds the bokeh visualization tool. #684
  - The tool allows a user to launch a server on their machine to visualise whole slide images, overlay the results of deep learning algorithms or to select a patch from whole slide image and run TIAToolbox deep learning engines.
  - This tool powers the TIA demos server. For details please see https://tiademos.dcs.warwick.ac.uk/.
- Extends Annotation to Support Init from WKB #639
- Adds `IOConfig` for NuClick in `pretrained_model.yaml` #709
- Adds functions to save the TIAToolbox Engine outputs to Zarr and AnnotationStore files. #724
- Adds Support for QuPath Annotation Imports #721

### Changes to API

- Adds `model.to(device)` and `model.load_model_from_file()` functionality to make it compatible with PyTorch API. #733
- Replaces `pretrained` with `weights` to make the engines compatible with the new PyTorch API. #621
- Adds support for high-level imports for various utility functions and classes such as `WSIReader`, `PatchPredictor` and `imread` #606, #607,
- Adds `tiatoolbox.typing` for type hints. #619
- Fixes incorrect file size saved by `save_tiles`, issue with certain WSIs raised by @TomastpPereira
- TissueMasker transform now returns mask instead of a list. #748
  - Fixes #732

### Bug Fixes and Other Changes

- Fixes `pixman` incompability error on Colab #601
- Removes `shapely.speedups`. The module no longer has any affect in Shapely >=2.0. #622
- Fixes errors in the slidegraph example notebook #608
- Fixes bugs in WSI Registration #645, #670, #693
- Fixes the situation where PatchExtractor.get_coords() can return patch coords which lie fully outside the bounds of a slide. #712
  - Fixes #710
- Fixes #738 raised by @xiachenrui

### Development related changes

- Replaces `flake8` and `isort` with `ruff` #625, #666
- Adds `mypy` checks to `root` and `utils` package. This will be rolled out in phases to other modules. #723
- Adds a module to detect file types using magic number/signatures #616
- Uses `poetry` for version updates instead of `bump2version`. #638
- Removes `setup.cfg` and uses `pyproject.toml` for project configurations.
- Reduces runtime for some unit tests e.g., #627, #630, #631, #629
- Reuses models and datasets in tests on GitHub actions by utilising cache #641, #644
- Set up parallel tests locally #671

**Full Changelog:** v1.4.0...v1.5.0

---------

Co-authored-by: pre-commit-ci[bot] <66853113+pre-commit-ci[bot]@users.noreply.github.com>
Co-authored-by: mostafajahanifar <[email protected]>
Co-authored-by: John Pocock <[email protected]>
Co-authored-by: DavidBAEpstein <[email protected]>
Co-authored-by: David Epstein <[email protected]>
Co-authored-by: Ruqayya Awan <[email protected]>
Co-authored-by: Mark Eastwood <[email protected]>
Co-authored-by: adamshephard <[email protected]>
Co-authored-by: adamshephard <[email protected]>
Co-authored-by: Abdol <[email protected]>
Co-authored-by: Jiaqi-Lv <[email protected]>
Co-authored-by: Abishek <[email protected]>
Co-authored-by: Dmitrii Blaginin <[email protected]>
@shaneahmed shaneahmed mentioned this pull request Dec 15, 2023
shaneahmed added a commit that referenced this pull request Dec 15, 2023
## 1.5.0 (2023-12-15)

### Major Updates and Feature Improvements

- Adds the bokeh visualization tool. #684
  - The tool allows a user to launch a server on their machine to visualise whole slide images, overlay the results of deep learning algorithms or to select a patch from whole slide image and run TIAToolbox deep learning engines.
  - This tool powers the TIA demos server. For details please see https://tiademos.dcs.warwick.ac.uk/.
- Extends Annotation to Support Init from WKB #639
- Adds `IOConfig` for NuClick in `pretrained_model.yaml` #709
- Adds functions to save the TIAToolbox Engine outputs to Zarr and AnnotationStore files. #724
- Adds Support for QuPath Annotation Imports #721

### Changes to API

- Adds `model.to(device)` and `model.load_model_from_file()` functionality to make it compatible with PyTorch API. #733
- Replaces `pretrained` with `weights` to make the engines compatible with the new PyTorch API. #621
- Adds support for high-level imports for various utility functions and classes such as `WSIReader`, `PatchPredictor` and `imread` #606, #607,
- Adds `tiatoolbox.typing` for type hints. #619
- Fixes incorrect file size saved by `save_tiles`, issue with certain WSIs raised by @TomastpPereira
- TissueMasker transform now returns mask instead of a list. #748
  - Fixes #732

### Bug Fixes and Other Changes

- Fixes `pixman` incompability error on Colab #601
- Removes `shapely.speedups`. The module no longer has any affect in Shapely >=2.0. #622
- Fixes errors in the slidegraph example notebook #608
- Fixes bugs in WSI Registration #645, #670, #693
- Fixes the situation where PatchExtractor.get_coords() can return patch coords which lie fully outside the bounds of a slide. #712
  - Fixes #710
- Fixes #738 raised by @xiachenrui

### Development related changes

- Replaces `flake8` and `isort` with `ruff` #625, #666
- Adds `mypy` checks to `root` and `utils` package. This will be rolled out in phases to other modules. #723
- Adds a module to detect file types using magic number/signatures #616
- Uses `poetry` for version updates instead of `bump2version`. #638
- Removes `setup.cfg` and uses `pyproject.toml` for project configurations.
- Reduces runtime for some unit tests e.g., #627, #630, #631, #629
- Reuses models and datasets in tests on GitHub actions by utilising cache #641, #644
- Set up parallel tests locally #671

**Full Changelog:** v1.4.0...v1.5.0
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

refactoring Code Refactoring

Projects

None yet

Development

Successfully merging this pull request may close these issues.

3 participants