ENH: Add TCN models + DAS datasets #630
Description
- Add DAS network architecture, converted to pytorch. From https://github.com/janclemenslab/das/blob/master/src/das/models.py
- Will require implementing the DAS-specific vendored version of TCN:
- As well as the vendored version of the Kapre layer:
- Ideally should have
torchvision-stylebuilderfunctions for each model that can specifies the correct network parameters and loads corresponding weights, i.e.das_bengalese_finches
- For training, add DAS to the FrameClassificationModel family
- Add datasets, from https://data.goettingen-research-online.de/dataverse/das?q=&types=dataverses%3Adatasets&sort=dateSort&order=desc&page=2
- Because these datasets are audio + annotations this will be one place to refactor
VocalDatasetas described in Refactor frame classification models to use singleWindowedFramesDatapipe#574 ENH: add BaseVocalDataset that usesvocles#539 and use it as a base for these specific datasets- Will need to make sure
itemfrom these datasets matches what's returned byAudioSequencein DAS: https://github.com/janclemenslab/das/blob/9ea349f13bdde7f44ba0506c5601078b2617cb45/src/das/data.py#L81
- Will need to make sure
- Because these datasets are audio + annotations this will be one place to refactor