Skip to content

Long To Wide Format Via Longitudinal Dataset - #88

Merged
simonprovost merged 4 commits into
mainfrom
feat/long-to-wide-and-vice-versa
Apr 20, 2026
Merged

Long To Wide Format Via Longitudinal Dataset#88
simonprovost merged 4 commits into
mainfrom
feat/long-to-wide-and-vice-versa

Conversation

@simonprovost

Copy link
Copy Markdown
Owner

A long wanted, and needed feature! Thanks to @MathiasValla to push it a little bit through #80.

Solving #64.

Cheers

Expose `to_wide` / `to_long` so cohorts stored in tidy long format can be
pivoted into the wide layout that Sklong estimators expect (and back),
keeping `features_group` and `non_longitudinal_features` in sync without
silently dropping or duplicating rows.

Signed-off-by: Provost Simon <simon1.provost@epitech.eu>
Validate round-trips, padded groups, duplicate (id, time) rows, and
varying static columns so future changes to the reshape helpers cannot
regress data integrity.

Signed-off-by: Provost Simon <simon1.provost@epitech.eu>
Add a tutorial explaining how `LongitudinalDataset.to_wide` /
`to_long` move cohorts between layouts, with Apple-styled 4K-rendered
animations driving the explanation. Wire the page into the tutorials
overview and Zensical nav.

Signed-off-by: Provost Simon <simon1.provost@epitech.eu>
Remove the deprecated long-format ingestion shim from the TpT decision
tree estimators (and its private `_preprocessing` helper); reshaping is
now the caller's responsibility via `LongitudinalDataset.to_wide`,
keeping a single, validated reshape implementation in the codebase.

Signed-off-by: Provost Simon <simon1.provost@epitech.eu>
@simonprovost simonprovost self-assigned this Apr 20, 2026
@simonprovost simonprovost added documentation Improvements or additions to documentation enhancement New feature or request Preparation Related content to data preparation labels Apr 20, 2026
@simonprovost
simonprovost merged commit c8a0cbb into main Apr 20, 2026
6 of 7 checks passed
@simonprovost
simonprovost deleted the feat/long-to-wide-and-vice-versa branch April 22, 2026 09:12
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

documentation Improvements or additions to documentation enhancement New feature or request Preparation Related content to data preparation

Projects

None yet

Development

Successfully merging this pull request may close these issues.

1 participant