Changelog
Early versions of this toolbox used a manually created changelog. As of March 2022, we have switched to using Github’s auto-generated changelog. If you would like to view the changelog for a particular release, you can do so on the Releases page: Each release contains a link for “Full Changelog”
Changelog for Versions before March 2022
0.1.14
Added
(#227) Add TransformerPooling.
(#179) Add GaussianBlur and RotationByMultiplesOf90 augmentations. Added torchvision and opencv to the environment file since it is necessary for the augmentations.
(#193) Add transformation adaptor to hi-ml-histopathology.
(#178) Add runner script for running ML experiments.
(#181) Add computational pathology tools in hi-ml-histopathology folder.
(#187) Add mean pooling layer for MIL.
(#186) Add inference to hi-ml runner.
(#198) Add cross-validation to hi-ml runner.
(#198) Improved editor setup for VSCode.
Changed
(#227) Pooling constructor is outside of DeepMIL and inside of BaseMIL now.
(#198) Model config loader is now more flexible, can accept fully qualified class name or just top-level module name and class (like histopathology.DeepSMILECrck)
(#198) Runner raises an error when Conda environment file contains a pip include (-r) statement
(#196) Show current workspace name in error message.
Fixed
([#267]https://github.com/microsoft/hi-ml/pull/267)) Correct PYTHONPATH for Windows in VS Code settings
([#266]https://github.com/microsoft/hi-ml/pull/266)) Pin jinja2 package to avoid ‘No attribute Markup’ bug in version 3.1.0
(#246) Added tolerance to
test_attentionlayers.py
.(#198) Dependencies for histopathology folder are no longer specified in
test_requirements.txt
, but correctly in the histopathology Conda environment.(#188) Updated DeepSMILES models. Now they are uptodate with innereye-dl.
(#179) HEDJitter was jittering the D channel as well. StainNormalization was relying on skimage.
(#195) Fix DeepMIL metrics bug whereby hard labels were used instead of probabilities.
Removed
Deprecated
0.1.13
Added
(#170) Add utils including bag sampling, bounding boxes, HEDJitter, StainNormalisation and add attention layers
Changed
(#173) Improve report tool: allow lists of tables, option for zipping report folder, option for base64 encoding images
Removed
Deprecated
0.1.12
Added
Changed
Fixed
(#161) Empty string as target folder for a dataset creates an invalid mounting path for the dataset in AzureML (fixes #160)
(#167) Fix bugs in logging hyperparameters: logging as name/value table, rather than one column per hyperparameter. Use string logging for all hyperparameters
(#174) Fix bugs in returned local_checkpoint_path when downloading checkpoints from AML run
Removed
Deprecated
0.1.10
Added
Changed
([#129])https://github.com/microsoft/hi-ml/pull/129)) Refactor command line tools’ arguments. Refactor health_azure.utils’ various get_run functions. Replace argparsing with parametrized classes.
Fixed
Removed
Deprecated
0.1.9 (2021-10-20)
Added
Changed
(#133) Made large breaking changes to module names, from
health.azure
tohealth_azure
.([#141])(https://github.com/microsoft/hi-ml/pull/141)) Update changelog for release and increase scope of test_register_environment to ensure that by default environments are registered with a version number
0.1.8 (2021-10-06)
Added
0.1.7 (2021-10-04)
Changed
(#112) Update himl_tensorboard to work with files not in ‘logs’ directory
(#106) Split into two packages. Most of existing package renamed to hi-ml-azure, remained remains hi-ml.
(#113) Add helper function to download files from AML Run, tidied up some command line args, and moved some functions from himl.py to azure_util.py
(#122) Add helper functions to upload to and download from AML Datastores
0.1.4 (2021-09-15)
This is the baseline release.