torchcell.loader

Batch loaders for experiment datasets. CpuExperimentLoaderMultiprocessing prefetches batches in worker processes on the CPU side; the dense_padding_data_loader module pads batches to a fixed node count for models that need dense inputs.

`cpu_experiment_loader`

CPU-side prefetching batch loaders backed by worker threads/processes.

Classes 

cpu_experiment_loader.CpuExperimentLoaderMultiprocessing

Process-based loader that prefetches dataset batches into a queue.

`dense_padding_data_loader`

Data loader and collater that pad batches to a dense, fixed node count.

Classes 

`dense_padding_data_loader.DensePaddingCollater`	Collate dataset objects into a mini-batch using dense padding.
`dense_padding_data_loader.DensePaddingDataLoader`	Data loader that merges dataset objects into mini-batches via dense padding.

Functions 

`dense_padding_data_loader.dense_padded_collate`	Collate a list of data objects into a single dense object of type cls.
`dense_padding_data_loader.dense_padded_from_data_list`	Construct a dense Batch object from a Python list of data objects.