Skip to content

Restarting model training from a saved checkpoint. #96

@ediriwas

Description

@ediriwas

I am training a scgen model using the following command:

model.train(max_epochs=100,
batch_size=32,
early_stopping=True,
early_stopping_patience=25,
enable_checkpointing=True
)

The integration is taking a long time and the job is ultimately cancelled, however, I have the saved checkpoint file. How can re start the model training using the information saved from the checkpoint?

Metadata

Metadata

Assignees

No one assigned

    Labels

    No labels
    No labels

    Type

    No type

    Projects

    No projects

    Milestone

    No milestone

    Relationships

    None yet

    Development

    No branches or pull requests

    Issue actions