v0.2.1

What changed

1. Second dummy data with noisy labels (see misc\synthetic_data.ipynb).
2. Removed the support for the Anaconda environment, all the packages are specified in requirements.txt (see this tutorial).
3. Upgraded pytorch-lightning and dask packages to the newer versions.
4. Fixed the issue with the data augmentation described in here.
5. Added new experiment (using two data sources) in workflow_two_datasets.ipynb.
6. Cleaned typos

v0.2.0

What changed

1. Second dummy data with noisy labels (see misc\synthetic_data.ipynb).
2. Removed the support for the Anaconda environment, all the packages are specified in requirements.txt (see this tutorial).
3. Upgraded pytorch-lightning and dask packages to the newer versions.
4. Fixed the issue with the data augmentation described in here.
5. Added new experiment (using two data sources) in workflow_two_datasets.ipynb.

v0.1.4

1. Cleaned the simple_workflow.ipynb
2. I fixed the issue with the dataloader that terminated training on the 3rd epoch.

v0.1.3

1. Added more annotations
2. Fixed TypeError: on_train_epoch_end() missing 1 required positional argument: 'output'

v0.1.2

Minor fixes

v0.1.1

Minor changes

0.1.0

Simple example in simple_workflow.ipynb (specification of the Anaconda environment are in environment.yml