…s()`

…tor & non-OpenCV normalizer

- Save a slide alignment with `WSI.alignment.save(path)`
- Re-apply a saved alignment with
`WSI.load_alignment(path)` or `WSI.apply_alignment(Alignment.load(path)`

- If a PyTorch, GPU-enabled normalizer has a device set, use this device when calculating DatasetFeatures
- Move normalizers to a device when setting the `.device` attribute
- Change the preferred device for Reinhard PyTorch normalizer from 'gpu' to 'cuda'

- Not dropping this batch can lead to shape errors, issues with batch normalization, and other problems.

…e size w/ CycleGAN

- Disable "use slide bounding boxes" option in Studio with cucim backend

- Alignment is now performed with reference to base slide dimensions (untransformed, without bounds or flipping/rotating). This is groundwork to allow alignments to be portable across tile sizes and slides (but potentially consistent between slide scanners)

- Reduce memory usage when generating images with CuCIM by using float32 instead of float64 during image conversion

- This initial support needs minor refactoring for efficiency and to ensure broader compatibility

…ch feature extractors.

…sorflow()

- Fix error where 'n' was not recognized as a valid augmentation in PyTorch.
- Improves error handling for stain augmentation, raising an error if 'n' is provided but not a stain normalizer.

…ction

- Use by setting `num_gpus` in `Project.generate_feature_bags()`
- Works in both pytorch and tensorflow

…atures

…site validation

… exist

Add TIFF suppport for:
- sf.io.write_tfrecords_single
- sf.io.write_tfrecords_multi

…preloading

… preloading

- Save MIL parameters (`mil_params.json`) when evaluating an MIL model
- New `get_mil_tile_predictions()` returns a pandas dataframe of predictions

- `drop_last`: drop the last non-full training batch (defaults to True)
- `save_monitor`: Set the monitored parameter during training that triggers a model to be saved. Defaults to "valid_loss"
- Forward compatibility added for upcoming FastAI configurations by ignoring (and warning about) unrecognized configuration parameters, rather than raising an error.

- This adds forward compatibility for new MIL hyperparameters. If an MIL model is loaded with unrecognized hyperparameters, the default behavior will be to log a warning message and ignore the hyperparameter. When loading a model with `sf.mil.util.load_model_weights()`, the `strict` parameter controls whether an unrecognized hyperparameter raises an error or if a warning message is instead logged.

…per patient

- Macenko (OpenCV and PyTorch)
- Reinhard (PyTorch)
- Define new `v2` augmentation space

…tion)

- Add multiclass support to segmentation models
- Various bug fixes with segmentation training

- Multilabel segmentation support for qc.Segment

- Error is raised when a dataset has old/outdated index files that need regenerated. This fix circumvents the problem by regenerating index files before calculating/exporting features.

…bnails/masks

- Add segmentation documentation
- Switch from `loss_mode` to `mode`; add `lr` parameter
- Auto-detect `out_classes` from segmentation labels

- Fix bug with GPU stain augmentation in PyTorch (ValueError: Stain augmentation (n) requires a stain normalizer, which was not provided)
- Fix "AssertionError: Input tensor must be float" for some PyTorch models deployed in Studio
- Fix edge case where there is 1 tile in a slide
- Fix bug in Studio in instances where there are no tiles in a slide (e.g. a JPEG image smaller than the tile size)