Issues

chore: add `from future import annotations` to every module so signatures and help() print types as written

Data-Simply/openretailscience#635 · opened

on Nov 1, 2026

feat(api): keep each package's help() page a short map that names the next call

Data-Simply/openretailscience#633 · opened

on Oct 31, 2026

1

build: replace pyright with pyrefly and fix source and test type errors

Data-Simply/openretailscience#631 · opened

on Oct 31, 2026

refactor!: make public API arguments after the data input keyword-only

Data-Simply/openretailscience#630 · opened

on Oct 31, 2026

1

fix(validation): an unexpected keyword argument names the parameters a class takes, without dumping every option

Data-Simply/openretailscience#629 · opened

on Oct 31, 2026

fix(cohort)!: periods past the end of the data are 0, the same as a period with no buyers

Data-Simply/openretailscience#628 · opened

on Oct 30, 2026

fix: prefer the nearest .git when install_skills picks the project root

Data-Simply/openretailscience#623 · opened

on Oct 28, 2026

docs: serve a Markdown copy of each page for AI agents

Data-Simply/openretailscience#621 · opened

on Oct 27, 2026

docs: get the docs site indexed, unblock /examples for AI crawlers, and describe the package in its metadata

Data-Simply/openretailscience#617 · opened

on Oct 27, 2026

fix(period_on_period): a date-only period end drops later rows on the end day

Data-Simply/openretailscience#613 · opened

on Oct 26, 2026

perf: vectorise the GainLoss decomposition and the CustomerDecisionHierarchy Yule's Q matrix

Data-Simply/openretailscience#611 · opened

on Oct 26, 2026

perf(gain_loss): GainLoss with group_col builds every group x customer combination

Data-Simply/openretailscience#610 · opened

on Oct 26, 2026

feat(segmentation): option for tied values to share a Threshold/HML segment

Data-Simply/openretailscience#608 · opened

on Oct 26, 2026

fix(revenue_tree): a single group loses its group label

Data-Simply/openretailscience#602 · opened

on Oct 22, 2026

fix(rfm): recency is wrong for string transaction dates that aren't zero-padded

Data-Simply/openretailscience#600 · opened

on Oct 22, 2026

fix(date): filter_and_label_by_periods drops the rest of a date-only end day on timestamp columns

Data-Simply/openretailscience#599 · opened

on Oct 22, 2026

fix(bar): percentage_by_bar_group fails when an x value repeats

Data-Simply/openretailscience#598 · opened

on Oct 22, 2026

fix(time): plot adds a transaction_period column to the caller's DataFrame

Data-Simply/openretailscience#597 · opened

on Oct 22, 2026

fix(area): plot with group_col and x_col=None raises KeyError: None

Data-Simply/openretailscience#596 · opened

on Oct 22, 2026

fix(gain_loss): list masks crash, a non-unique index gives a false overlap error, and the length check is unreachable

Data-Simply/openretailscience#595 · opened

on Oct 22, 2026

fix(product_association): target_item rejects integer product ids

Data-Simply/openretailscience#594 · opened

on Oct 22, 2026

fix(gain_loss): 'Boolean Series key will be reindexed' warnings when df has rows outside p1/p2

Data-Simply/openretailscience#588 · opened

on Oct 14, 2026

fix(types): annotation mismatches in product_association, gain_loss and SubplotBase returns

Data-Simply/openretailscience#587 · opened

on Oct 14, 2026

Add group_col to RFMSegmentation

Data-Simply/openretailscience#580 · opened

on Sep 28, 2026

Migrate documentation from Material for MkDocs to Zensical

Data-Simply/openretailscience#560 · opened

on Aug 18, 2026

Navigation Menu

Repository navigation

Issues

chore: add `from future import annotations` to every module so signatures and help() print types as written

feat(api): keep each package's help() page a short map that names the next call

build: replace pyright with pyrefly and fix source and test type errors

refactor!: make public API arguments after the data input keyword-only

fix(validation): an unexpected keyword argument names the parameters a class takes, without dumping every option

fix(cohort)!: periods past the end of the data are 0, the same as a period with no buyers

fix: prefer the nearest .git when install_skills picks the project root

docs: serve a Markdown copy of each page for AI agents

docs: get the docs site indexed, unblock /examples for AI crawlers, and describe the package in its metadata

fix(period_on_period): a date-only period end drops later rows on the end day

perf: vectorise the GainLoss decomposition and the CustomerDecisionHierarchy Yule's Q matrix

perf(gain_loss): GainLoss with group_col builds every group x customer combination

feat(segmentation): option for tied values to share a Threshold/HML segment

fix(revenue_tree): a single group loses its group label

fix(rfm): recency is wrong for string transaction dates that aren't zero-padded

fix(date): filter_and_label_by_periods drops the rest of a date-only end day on timestamp columns

fix(bar): percentage_by_bar_group fails when an x value repeats

fix(time): plot adds a transaction_period column to the caller's DataFrame

fix(area): plot with group_col and x_col=None raises KeyError: None

fix(gain_loss): list masks crash, a non-unique index gives a false overlap error, and the length check is unreachable

fix(product_association): target_item rejects integer product ids

fix(gain_loss): 'Boolean Series key will be reindexed' warnings when df has rows outside p1/p2

fix(types): annotation mismatches in product_association, gain_loss and SubplotBase returns

Add group_col to RFMSegmentation

Migrate documentation from Material for MkDocs to Zensical