I was wondering if someone could explain how the data augmentation works within DLC.
I just passed
deeplabcut.create_training_dataset(config_path, augmenter_type='imgaug') and then started training. Was that the right thing to do? I see “UnaugmentedDataSet” in my training-datasets, but nothing else.
What kind of augmentation is done with imgaug? In the functionDetails, I see that there are a lot of possibilities but I didn’t see any options in the example pose_cfg.yaml other than scale jitter and batch size. Though the example on github also looks pretty different from the one generated when the training set was created.