Skip to content

interesting, for movie recommend task, code maybe like this #17

@jt120

Description

@jt120

like we have 1000 users, 2000 movies and each movie have some tags, total tags is 200.

and we have user rating, from 1 to 5

the train_x maybe users+movies+tags, like nx3200 matrix, and train_y maybe nx1

so we got U, V, Z from cmf model

and we can prediction score from V@Z_T

am I right?

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