Skip to content

distill #160

@Mahhos

Description

@Mahhos

Hi! Thanks for the great repo. I'm trying to distill a model2vec model from a relatively small sentence transformer, but it's taking several hours on GPU. Do you have any advice on optimizing this?

from model2vec.distill import distill
m2v_model = distill("MY MODEL", pca_dims=256)
m2v_model.save_pretrained("m2v_model")

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