plismbench.models.hkust module

Models from Hong Kong University of Science and Technology.

class plismbench.models.hkust.GPFM(device: int | list[int] | None = -1, mixed_precision: bool = False)

Bases: Extractor

GPFM model developped by HKUST (1).

Note

1. Ma, J., Guo, Z., Zhou, F., Wang, Y., Xu, Y., et al. (2024).

Towards a generalizable pathology foundation model via unified knowledge distillation (arXiv No. 2407.18449). arXiv. https://arxiv.org/abs/2407.18449

Parameters:

• device (int | list[int] | None = DEFAULT_DEVICE,) – Compute resources to use. If None, will use all available GPUs. If -1, extraction will run on CPU.

• mixed_precision (bool = True) – Whether to use mixed_precision.

property transform: Compose

Transform method to apply element wise.