Skip to content

Add Tutorial on Text Features #446

@popescu-v

Description

@popescu-v

Description

In v11, new variable (feature) types Text and TextList have been introduced. As of now, only basic illustration of their usage is included in the Core API samples (i.e. for basic parameterization of the maximum number of text features - max_text_features and of their type - text_features).

The goal of this issue is to add new tutorial (use-case-driven) which illustrates the usage of these features, especially for text_features="tokens" parameterization, whereby the user provides pre-tokenized text data to Khiops.

Note: max_text_features and text_features should also be made available to the Sklearn API (as n_text_features and text_features respectively), by adding them as parameters to the estimator classes (akin to n_trees which uses Core API max_trees, etc). ; cf. issue #39

Metadata

Metadata

Assignees

No one assigned

    Labels

    Status/DraftThe issue is still not well defined

    Type

    No type

    Projects

    No projects

    Milestone

    No milestone

    Relationships

    None yet

    Development

    No branches or pull requests

    Issue actions