Framework for training and tracking deep learning models built on pytorch-lightning and hydra. Soon to encompass both image classification, semantic segmentation and object detection.
Currently there are three use-cases with configs intermingled with the framework. Consider to separate them out from the framework. As more use-cases examples come along the framework code will get bloated.
The initial thought is to add an experiments folder at the root level of this repository and add experiments as sub-folders to that folder. This means moving out code and configs related to MNIST, CIFAR10 etc.
Currently there are three use-cases with configs intermingled with the framework. Consider to separate them out from the framework. As more use-cases examples come along the framework code will get bloated.
The initial thought is to add an
experimentsfolder at the root level of this repository and add experiments as sub-folders to that folder. This means moving out code and configs related to MNIST, CIFAR10 etc.