Migrate BaseOutlier and WinsorizerBase to narwhals, add polars support - #1033
Open
solegalli wants to merge 1 commit into
Open
Migrate BaseOutlier and WinsorizerBase to narwhals, add polars support#1033solegalli wants to merge 1 commit into
solegalli wants to merge 1 commit into
Conversation
Shared base for all outlier transformers (ArbitraryOutlierCapper extends BaseOutlier directly; Winsoriser/OutlierTrimmer extend WinsorizerBase): column reorder + NA/Inf checks in _check_transform_input_and_state(), the fold-limit estimation in WinsorizerBase.fit() (gaussian/iqr/mad/ quantiles), and the capping step in BaseOutlier._transform() are now dataframe-agnostic. Capping (np.clip against per-column bounds) was benchmarked three ways at 10k/50k/100k rows x 1/2/10 columns: pandas-native .clip() loop vs. a single narwhals with_columns(nw.col(v).clip(lo, hi) for v in ...) vs. grouping columns by which bound(s) apply and running up to 3 vectorized numpy calls (np.clip/minimum/maximum) via to_numpy()/new_series(), mirroring ReciprocalTransformer's numpy-acceleration pattern. narwhals-generic alone was already close to parity (0.95-1.49x pandas-native - minimal loss, mergeable per the imputation-base precedent), but the numpy-grouped version was faster still: 0.16-0.82x of pandas-native on the homogeneous case (single tail, all columns share the same bound - the common Winsoriser/ OutlierTrimmer case) and 0.42-1.52x on mixed-coverage dicts (the ArbitraryOutlierCapper case, up to 3 groups). Adopted the numpy-grouped version as the single merged code path for both backends. A first numpy attempt used a blanket -inf/inf sentinel for the missing side per column (like RelativeFeatures-style bound arrays) - that's a correctness bug, not just a style choice: mixing an int64 numpy array with a float -inf/inf bound upcasts the whole column to float64 even when the real, present bound is an int (e.g. ArbitraryOutlierCapper's own docstring example, `max_capping_dict=dict(x1=8)`, expects int64 out). Grouping columns into "both bounds" / "right only" / "left only" buckets and calling np.clip/minimum/maximum with only the bounds that actually exist avoids ever introducing an inf, so dtype promotion matches pandas .clip() exactly - verified byte-for-byte against the old pandas-only implementation across all 4 capping methods x 3 tails, plus the int-dtype and mixed-dict-coverage cases. Also found and fixed a real bug introduced while migrating fit(): plain np.mean/np.std/np.quantile/np.median propagate NaN, unlike pandas' mean/std/quantile/median which skip NaN by default. With missing_values="ignore" and NaN present, this silently produced NaN caps instead of the caps computed from non-null data. Fixed by using the nan-aware numpy variants (np.nanmean/nanstd/nanquantile/nanmedian). Caught by tests/test_outliers/test_winsorizer.py::test_transformer_ignores_na_in_df, which predates this migration but exercises exactly this path. variables/feature names can be int or str; passing a plain list to narwhals' .select() only works for string columns, so every .select() call here uses nw.col(*variables) instead - .select(list_of_ints) raises InvalidIntoExprError. Verified: tests/test_outliers full suite - 83 passed, 3 pre-existing failures in test_check_estimator_outliers.py (sklearn's check_estimator feeds raw numpy arrays, which check_X() has always rejected per the narwhals migration's dataframe-only contract; identical failure set before and after this change). flake8 and mypy clean on the file. Module imports and runs fit/_transform end-to-end on polars with pandas import fully blocked. sphinx -W build clean (only the pre-existing unrelated linkcode_resolve warning). All 4 capping-method x tail combinations and the Winsoriser/OutlierTrimmer/ArbitraryOutlierCapper docstring examples produce byte-identical output to the pre-migration code (checked exact numeric values and dtypes). Not migrated here (belongs to the 3 follow-on transformer branches): ArbitraryOutlierCapper.fit()/transform(), Winsoriser's add_indicators branch (pd.concat), and OutlierTrimmer.transform() (its own .le/.ge/.loc row-filtering, which doesn't go through BaseOutlier._transform at all) all still import pandas directly. Existing tests in tests/test_outliers were left pandas-only rather than parametrized over polars, since they exercise those still-pandas-only subclasses, not BaseOutlier/ WinsorizerBase directly - parametrizing them now would fail on reasons unrelated to this file. Co-Authored-By: Claude Sonnet 5 <noreply@anthropic.com>
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
1 participant
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Shared base for all outlier transformers (
ArbitraryOutlierCapperextendsBaseOutlierdirectly;Winsoriser/OutlierTrimmerextendWinsorizerBase). Migrates the column reorder + NA/Inf checks in_check_transform_input_and_state(), the fold-limit estimation inWinsorizerBase.fit()(gaussian / iqr / mad / quantiles), and the capping step inBaseOutlier._transform()to be dataframe-agnostic.Capping (merge vs split): benchmarked pandas-native
.clip()loop vs a single narwhalswith_columns(...clip...)vs grouping columns by which bound(s) apply and running up to 3 vectorized numpy calls (np.clip/minimum/maximum) — 10k/50k/100k rows × 1/2/10 cols. narwhals-generic alone was already near parity (0.95–1.49x), but the numpy-grouped version was faster: 0.16–0.82x on the homogeneous case (single tail, all columns share a bound — the common Winsoriser/OutlierTrimmer case) and 0.42–1.52x on mixed-coverage dicts (ArbitraryOutlierCapper). Adopted the numpy-grouped version as the single merged path for both backends.-inf/infsentinel for the missing side per column — that's a correctness bug: mixing an int64 array with a floatinfbound upcasts the whole column to float64 even when the real present bound is int (e.g.max_capping_dict=dict(x1=8)expects int64 out). Grouping columns into "both bounds" / "right only" / "left only" buckets and passing only the bounds that exist avoids ever introducing an inf — dtype promotion now matches pandas.clip()exactly (verified byte-for-byte across all 4 methods × 3 tails, plus int-dtype and mixed-dict cases).fit()): plainnp.mean/std/quantile/medianpropagate NaN, unlike pandas' NaN-skipping defaults. Withmissing_values="ignore"and NaN present this silently produced NaN caps. Fixed with the nan-aware variants. Caught bytest_winsorizer.py::test_transformer_ignores_na_in_df.variables/ feature names can be int or str;.select(list_of_ints)raisesInvalidIntoExprError, so every.select()call usesnw.col(*variables).Verified:
tests/test_outliers— 83 passed, 3 pre-existingcheck_estimatorfailures (raw numpy-array input, rejected bycheck_X()under the narwhals dataframe-only contract; identical before and after). flake8 / mypy clean, sphinx -W clean. Runsfit/_transformend-to-end on polars with pandas import blocked.Not migrated here (belongs to the follow-on transformer PRs):
ArbitraryOutlierCapper.fit()/transform(), Winsoriser'sadd_indicatorsbranch,OutlierTrimmer.transform().