Skip to content

Dtm and topic models in a supervised setting #1

@nicolasdugue

Description

@nicolasdugue

Hi,

The tmtoolkit package is great and I really enjoyed working on my corpus with it. However, I am now working in a supervised setting, I have a train corpus and a test corpus. Thus, I want to build my dtm model on my train corpus and then use the same vocabulary to get the dtm of my test corpus. Similarly, I would like to train my topic model on the train corpus, and then, use the topics uncovered to get the document-topic matrix of the test corpus.
Is there a way to do so on tmtoolkit ? Or is it planned in the future ? I did not find any way and thus used ad-hoc code.

Regards

Metadata

Metadata

Assignees

No one assigned

    Labels

    No labels
    No labels

    Projects

    No projects

    Milestone

    No milestone

    Relationships

    None yet

    Development

    No branches or pull requests

    Issue actions