This is the readme file for the new PeekingDuck Training feature.
Run on default cifar10 dataset:
python peekingduck/training/main.pyTo use rsna dataset:
python ./peekingduck/training/main.py data_module=rsna project_name=rsnaTo use vegfru dataset:
python ./peekingduck/training/main.py data_module=vegfru project_name=vegfruTo use vegfru5 dataset:
python ./peekingduck/training/main.py data_module=vegfru5 project_name=vegfru5 debug=TrueTo log in to wandb
wandb loginCopy your key and paste it into your command line when asked to authorize your account. At the top of your training script, start a new run etc:
wandb.init(project="training-pipeline", entity="peekingduck")