hest.bench.benchmark

hest.bench.benchmark(encoder: Module, enc_transf: Callable, precision: dtype, cli_args: dict = None, **kwargs) → Tuple[list, dict]

Benchmark a patch encoder on HEST-bench

Parameters:

• encoder (torch.nn.Module) – patch encoder to benchmark

• enc_transf (Callable) – transformation applied to encoder during inference

• precision (torch.dtype) – precision used by torch.amp.autocast(‘cuda’) during inference for encoder

• cli_args (dict) – cli_arguments. Defaults to None.

• **kwargs – lookup BenchmarkConfig for additional parameters