Skip to content

Dataset preprocessing is slow #12

Description

@shervinea

Context. At training time, the enzymes' volumes are precomputed with the check_precomputed() method, e.g. for the training set:

training_generator.check_precomputed()

Since enzymes' volumes are independent from each other, this task is parallelizable by nature.

Current behavior. This process runs on a single core.

Desired behavior. This process is parallelized as much as possible.

Metadata

Metadata

Assignees

No one assigned

    Projects

    No projects

    Milestone

    No milestone

    Relationships

    None yet

    Development

    No branches or pull requests

    Issue actions