Thanks to visit codestin.com
Credit goes to github.com

Skip to content

Possible bug in sklearn 1.6.1 PartialDependenceDisplay.from_estimator when target and feature are both binary #30675

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Jump to bottom
Open
vnijs opened this issue Jan 20, 2025 · 4 comments
Open

Possible bug in sklearn 1.6.1 PartialDependenceDisplay.from_estimator when target and feature are both binary #30675

vnijs opened this issue Jan 20, 2025 · 4 comments
Labels
Bug Regression

Comments

@vnijs
Copy link

vnijs commented Jan 20, 2025 •
edited by jeremiedbb
Loading

Describe the bug

PartialDependenceDisplay.from_estimator does not seem able to handle dummy variables when the response variable is binary. See example below. The example works fine in 1.5.2 but returns ValueError: cannot reshape array of size 1 into shape (2) in 1.6.1

Steps/Code to Reproduce

import numpy as np
import pandas as pd
from sklearn.ensemble import RandomForestClassifier
from sklearn.inspection import PartialDependenceDisplay

np.random.seed(42)
n_samples = 1000
age = np.random.normal(35, 10, n_samples)
smoker = np.random.choice([0, 1], n_samples, p=[0.7, 0.3])
prob_disease = 1 / (1 + np.exp(-(age - 35) / 10 - 2 * smoker))
heart_disease = (np.random.random(n_samples) < prob_disease).astype(int)
df = pd.DataFrame({"age": age, "smoker": smoker, "heart_disease": heart_disease})
X = df[["age", "smoker"]]
y = df["heart_disease"]

rf_model = RandomForestClassifier(n_estimators=100, random_state=42)
rf_model.fit(X, y)

pdp_age = PartialDependenceDisplay.from_estimator(rf_model, X, features=[0, 1])

Expected Results

PDP plots for age and smoker.

Actual Results

---------------------------------------------------------------------------
ValueError                                Traceback (most recent call last)
Cell In[1], [line 19](vscode-notebook-cell:?execution_count=1&line=19)
     [16](vscode-notebook-cell:?execution_count=1&line=16) rf_model = RandomForestClassifier(n_estimators=100, random_state=42)
     [17](vscode-notebook-cell:?execution_count=1&line=17) rf_model.fit(X, y)
---> [19](vscode-notebook-cell:?execution_count=1&line=19) pdp_age = PartialDependenceDisplay.from_estimator(rf_model, X, features=[0, 1])

File ~/miniconda/envs/msba/lib/python3.12/site-packages/sklearn/inspection/_plot/partial_dependence.py:707, in PartialDependenceDisplay.from_estimator(cls, estimator, X, features, sample_weight, categorical_features, feature_names, target, response_method, n_cols, grid_resolution, percentiles, method, n_jobs, verbose, line_kw, ice_lines_kw, pd_line_kw, contour_kw, ax, kind, centered, subsample, random_state)
    [701](https://file+.vscode-resource.vscode-cdn.net/Users/vnijs/gh/pyrsm/examples/model/~/miniconda/envs/msba/lib/python3.12/site-packages/sklearn/inspection/_plot/partial_dependence.py:701)         raise ValueError(
    [702](https://file+.vscode-resource.vscode-cdn.net/Users/vnijs/gh/pyrsm/examples/model/~/miniconda/envs/msba/lib/python3.12/site-packages/sklearn/inspection/_plot/partial_dependence.py:702)             f"When a floating-point, subsample={subsample} should be in "
    [703](https://file+.vscode-resource.vscode-cdn.net/Users/vnijs/gh/pyrsm/examples/model/~/miniconda/envs/msba/lib/python3.12/site-packages/sklearn/inspection/_plot/partial_dependence.py:703)             "the (0, 1) range."
    [704](https://file+.vscode-resource.vscode-cdn.net/Users/vnijs/gh/pyrsm/examples/model/~/miniconda/envs/msba/lib/python3.12/site-packages/sklearn/inspection/_plot/partial_dependence.py:704)         )
    [706](https://file+.vscode-resource.vscode-cdn.net/Users/vnijs/gh/pyrsm/examples/model/~/miniconda/envs/msba/lib/python3.12/site-packages/sklearn/inspection/_plot/partial_dependence.py:706) # compute predictions and/or averaged predictions
--> [707](https://file+.vscode-resource.vscode-cdn.net/Users/vnijs/gh/pyrsm/examples/model/~/miniconda/envs/msba/lib/python3.12/site-packages/sklearn/inspection/_plot/partial_dependence.py:707) pd_results = Parallel(n_jobs=n_jobs, verbose=verbose)(
    [708](https://file+.vscode-resource.vscode-cdn.net/Users/vnijs/gh/pyrsm/examples/model/~/miniconda/envs/msba/lib/python3.12/site-packages/sklearn/inspection/_plot/partial_dependence.py:708)     delayed(partial_dependence)(
    [709](https://file+.vscode-resource.vscode-cdn.net/Users/vnijs/gh/pyrsm/examples/model/~/miniconda/envs/msba/lib/python3.12/site-packages/sklearn/inspection/_plot/partial_dependence.py:709)         estimator,
    [710](https://file+.vscode-resource.vscode-cdn.net/Users/vnijs/gh/pyrsm/examples/model/~/miniconda/envs/msba/lib/python3.12/site-packages/sklearn/inspection/_plot/partial_dependence.py:710)         X,
    [711](https://file+.vscode-resource.vscode-cdn.net/Users/vnijs/gh/pyrsm/examples/model/~/miniconda/envs/msba/lib/python3.12/site-packages/sklearn/inspection/_plot/partial_dependence.py:711)         fxs,
    [712](https://file+.vscode-resource.vscode-cdn.net/Users/vnijs/gh/pyrsm/examples/model/~/miniconda/envs/msba/lib/python3.12/site-packages/sklearn/inspection/_plot/partial_dependence.py:712)         sample_weight=sample_weight,
    [713](https://file+.vscode-resource.vscode-cdn.net/Users/vnijs/gh/pyrsm/examples/model/~/miniconda/envs/msba/lib/python3.12/site-packages/sklearn/inspection/_plot/partial_dependence.py:713)         feature_names=feature_names,
    [714](https://file+.vscode-resource.vscode-cdn.net/Users/vnijs/gh/pyrsm/examples/model/~/miniconda/envs/msba/lib/python3.12/site-packages/sklearn/inspection/_plot/partial_dependence.py:714)         categorical_features=categorical_features,
    [715](https://file+.vscode-resource.vscode-cdn.net/Users/vnijs/gh/pyrsm/examples/model/~/miniconda/envs/msba/lib/python3.12/site-packages/sklearn/inspection/_plot/partial_dependence.py:715)         response_method=response_method,
    [716](https://file+.vscode-resource.vscode-cdn.net/Users/vnijs/gh/pyrsm/examples/model/~/miniconda/envs/msba/lib/python3.12/site-packages/sklearn/inspection/_plot/partial_dependence.py:716)         method=method,
    [717](https://file+.vscode-resource.vscode-cdn.net/Users/vnijs/gh/pyrsm/examples/model/~/miniconda/envs/msba/lib/python3.12/site-packages/sklearn/inspection/_plot/partial_dependence.py:717)         grid_resolution=grid_resolution,
    [718](https://file+.vscode-resource.vscode-cdn.net/Users/vnijs/gh/pyrsm/examples/model/~/miniconda/envs/msba/lib/python3.12/site-packages/sklearn/inspection/_plot/partial_dependence.py:718)         percentiles=percentiles,
    [719](https://file+.vscode-resource.vscode-cdn.net/Users/vnijs/gh/pyrsm/examples/model/~/miniconda/envs/msba/lib/python3.12/site-packages/sklearn/inspection/_plot/partial_dependence.py:719)         kind=kind_plot,
    [720](https://file+.vscode-resource.vscode-cdn.net/Users/vnijs/gh/pyrsm/examples/model/~/miniconda/envs/msba/lib/python3.12/site-packages/sklearn/inspection/_plot/partial_dependence.py:720)     )
    [721](https://file+.vscode-resource.vscode-cdn.net/Users/vnijs/gh/pyrsm/examples/model/~/miniconda/envs/msba/lib/python3.12/site-packages/sklearn/inspection/_plot/partial_dependence.py:721)     for kind_plot, fxs in zip(kind_, features)
    [722](https://file+.vscode-resource.vscode-cdn.net/Users/vnijs/gh/pyrsm/examples/model/~/miniconda/envs/msba/lib/python3.12/site-packages/sklearn/inspection/_plot/partial_dependence.py:722) )
    [724](https://file+.vscode-resource.vscode-cdn.net/Users/vnijs/gh/pyrsm/examples/model/~/miniconda/envs/msba/lib/python3.12/site-packages/sklearn/inspection/_plot/partial_dependence.py:724) # For multioutput regression, we can only check the validity of target
    [725](https://file+.vscode-resource.vscode-cdn.net/Users/vnijs/gh/pyrsm/examples/model/~/miniconda/envs/msba/lib/python3.12/site-packages/sklearn/inspection/_plot/partial_dependence.py:725) # now that we have the predictions.
    [726](https://file+.vscode-resource.vscode-cdn.net/Users/vnijs/gh/pyrsm/examples/model/~/miniconda/envs/msba/lib/python3.12/site-packages/sklearn/inspection/_plot/partial_dependence.py:726) # Also note: as multiclass-multioutput classifiers are not supported,
    [727](https://file+.vscode-resource.vscode-cdn.net/Users/vnijs/gh/pyrsm/examples/model/~/miniconda/envs/msba/lib/python3.12/site-packages/sklearn/inspection/_plot/partial_dependence.py:727) # multiclass and multioutput scenario are mutually exclusive. So there is
    [728](https://file+.vscode-resource.vscode-cdn.net/Users/vnijs/gh/pyrsm/examples/model/~/miniconda/envs/msba/lib/python3.12/site-packages/sklearn/inspection/_plot/partial_dependence.py:728) # no risk of overwriting target_idx here.
    [729](https://file+.vscode-resource.vscode-cdn.net/Users/vnijs/gh/pyrsm/examples/model/~/miniconda/envs/msba/lib/python3.12/site-packages/sklearn/inspection/_plot/partial_dependence.py:729) pd_result = pd_results[0]  # checking the first result is enough

File ~/miniconda/envs/msba/lib/python3.12/site-packages/sklearn/utils/parallel.py:77, in Parallel.__call__(self, iterable)
     [72](https://file+.vscode-resource.vscode-cdn.net/Users/vnijs/gh/pyrsm/examples/model/~/miniconda/envs/msba/lib/python3.12/site-packages/sklearn/utils/parallel.py:72) config = get_config()
     [73](https://file+.vscode-resource.vscode-cdn.net/Users/vnijs/gh/pyrsm/examples/model/~/miniconda/envs/msba/lib/python3.12/site-packages/sklearn/utils/parallel.py:73) iterable_with_config = (
     [74](https://file+.vscode-resource.vscode-cdn.net/Users/vnijs/gh/pyrsm/examples/model/~/miniconda/envs/msba/lib/python3.12/site-packages/sklearn/utils/parallel.py:74)     (_with_config(delayed_func, config), args, kwargs)
     [75](https://file+.vscode-resource.vscode-cdn.net/Users/vnijs/gh/pyrsm/examples/model/~/miniconda/envs/msba/lib/python3.12/site-packages/sklearn/utils/parallel.py:75)     for delayed_func, args, kwargs in iterable
     [76](https://file+.vscode-resource.vscode-cdn.net/Users/vnijs/gh/pyrsm/examples/model/~/miniconda/envs/msba/lib/python3.12/site-packages/sklearn/utils/parallel.py:76) )
---> [77](https://file+.vscode-resource.vscode-cdn.net/Users/vnijs/gh/pyrsm/examples/model/~/miniconda/envs/msba/lib/python3.12/site-packages/sklearn/utils/parallel.py:77) return super().__call__(iterable_with_config)

File ~/miniconda/envs/msba/lib/python3.12/site-packages/joblib/parallel.py:1918, in Parallel.__call__(self, iterable)
   [1916](https://file+.vscode-resource.vscode-cdn.net/Users/vnijs/gh/pyrsm/examples/model/~/miniconda/envs/msba/lib/python3.12/site-packages/joblib/parallel.py:1916)     output = self._get_sequential_output(iterable)
   [1917](https://file+.vscode-resource.vscode-cdn.net/Users/vnijs/gh/pyrsm/examples/model/~/miniconda/envs/msba/lib/python3.12/site-packages/joblib/parallel.py:1917)     next(output)
-> [1918](https://file+.vscode-resource.vscode-cdn.net/Users/vnijs/gh/pyrsm/examples/model/~/miniconda/envs/msba/lib/python3.12/site-packages/joblib/parallel.py:1918)     return output if self.return_generator else list(output)
   [1920](https://file+.vscode-resource.vscode-cdn.net/Users/vnijs/gh/pyrsm/examples/model/~/miniconda/envs/msba/lib/python3.12/site-packages/joblib/parallel.py:1920) # Let's create an ID that uniquely identifies the current call. If the
   [1921](https://file+.vscode-resource.vscode-cdn.net/Users/vnijs/gh/pyrsm/examples/model/~/miniconda/envs/msba/lib/python3.12/site-packages/joblib/parallel.py:1921) # call is interrupted early and that the same instance is immediately
   [1922](https://file+.vscode-resource.vscode-cdn.net/Users/vnijs/gh/pyrsm/examples/model/~/miniconda/envs/msba/lib/python3.12/site-packages/joblib/parallel.py:1922) # re-used, this id will be used to prevent workers that were
   [1923](https://file+.vscode-resource.vscode-cdn.net/Users/vnijs/gh/pyrsm/examples/model/~/miniconda/envs/msba/lib/python3.12/site-packages/joblib/parallel.py:1923) # concurrently finalizing a task from the previous call to run the
   [1924](https://file+.vscode-resource.vscode-cdn.net/Users/vnijs/gh/pyrsm/examples/model/~/miniconda/envs/msba/lib/python3.12/site-packages/joblib/parallel.py:1924) # callback.
   [1925](https://file+.vscode-resource.vscode-cdn.net/Users/vnijs/gh/pyrsm/examples/model/~/miniconda/envs/msba/lib/python3.12/site-packages/joblib/parallel.py:1925) with self._lock:

File ~/miniconda/envs/msba/lib/python3.12/site-packages/joblib/parallel.py:1847, in Parallel._get_sequential_output(self, iterable)
   [1845](https://file+.vscode-resource.vscode-cdn.net/Users/vnijs/gh/pyrsm/examples/model/~/miniconda/envs/msba/lib/python3.12/site-packages/joblib/parallel.py:1845) self.n_dispatched_batches += 1
   [1846](https://file+.vscode-resource.vscode-cdn.net/Users/vnijs/gh/pyrsm/examples/model/~/miniconda/envs/msba/lib/python3.12/site-packages/joblib/parallel.py:1846) self.n_dispatched_tasks += 1
-> [1847](https://file+.vscode-resource.vscode-cdn.net/Users/vnijs/gh/pyrsm/examples/model/~/miniconda/envs/msba/lib/python3.12/site-packages/joblib/parallel.py:1847) res = func(*args, **kwargs)
   [1848](https://file+.vscode-resource.vscode-cdn.net/Users/vnijs/gh/pyrsm/examples/model/~/miniconda/envs/msba/lib/python3.12/site-packages/joblib/parallel.py:1848) self.n_completed_tasks += 1
   [1849](https://file+.vscode-resource.vscode-cdn.net/Users/vnijs/gh/pyrsm/examples/model/~/miniconda/envs/msba/lib/python3.12/site-packages/joblib/parallel.py:1849) self.print_progress()

File ~/miniconda/envs/msba/lib/python3.12/site-packages/sklearn/utils/parallel.py:139, in _FuncWrapper.__call__(self, *args, **kwargs)
    [137](https://file+.vscode-resource.vscode-cdn.net/Users/vnijs/gh/pyrsm/examples/model/~/miniconda/envs/msba/lib/python3.12/site-packages/sklearn/utils/parallel.py:137)     config = {}
    [138](https://file+.vscode-resource.vscode-cdn.net/Users/vnijs/gh/pyrsm/examples/model/~/miniconda/envs/msba/lib/python3.12/site-packages/sklearn/utils/parallel.py:138) with config_context(**config):
--> [139](https://file+.vscode-resource.vscode-cdn.net/Users/vnijs/gh/pyrsm/examples/model/~/miniconda/envs/msba/lib/python3.12/site-packages/sklearn/utils/parallel.py:139)     return self.function(*args, **kwargs)

File ~/miniconda/envs/msba/lib/python3.12/site-packages/sklearn/utils/_param_validation.py:216, in validate_params.<locals>.decorator.<locals>.wrapper(*args, **kwargs)
    [210](https://file+.vscode-resource.vscode-cdn.net/Users/vnijs/gh/pyrsm/examples/model/~/miniconda/envs/msba/lib/python3.12/site-packages/sklearn/utils/_param_validation.py:210) try:
    [211](https://file+.vscode-resource.vscode-cdn.net/Users/vnijs/gh/pyrsm/examples/model/~/miniconda/envs/msba/lib/python3.12/site-packages/sklearn/utils/_param_validation.py:211)     with config_context(
    [212](https://file+.vscode-resource.vscode-cdn.net/Users/vnijs/gh/pyrsm/examples/model/~/miniconda/envs/msba/lib/python3.12/site-packages/sklearn/utils/_param_validation.py:212)         skip_parameter_validation=(
    [213](https://file+.vscode-resource.vscode-cdn.net/Users/vnijs/gh/pyrsm/examples/model/~/miniconda/envs/msba/lib/python3.12/site-packages/sklearn/utils/_param_validation.py:213)             prefer_skip_nested_validation or global_skip_validation
    [214](https://file+.vscode-resource.vscode-cdn.net/Users/vnijs/gh/pyrsm/examples/model/~/miniconda/envs/msba/lib/python3.12/site-packages/sklearn/utils/_param_validation.py:214)         )
    [215](https://file+.vscode-resource.vscode-cdn.net/Users/vnijs/gh/pyrsm/examples/model/~/miniconda/envs/msba/lib/python3.12/site-packages/sklearn/utils/_param_validation.py:215)     ):
--> [216](https://file+.vscode-resource.vscode-cdn.net/Users/vnijs/gh/pyrsm/examples/model/~/miniconda/envs/msba/lib/python3.12/site-packages/sklearn/utils/_param_validation.py:216)         return func(*args, **kwargs)
    [217](https://file+.vscode-resource.vscode-cdn.net/Users/vnijs/gh/pyrsm/examples/model/~/miniconda/envs/msba/lib/python3.12/site-packages/sklearn/utils/_param_validation.py:217) except InvalidParameterError as e:
    [218](https://file+.vscode-resource.vscode-cdn.net/Users/vnijs/gh/pyrsm/examples/model/~/miniconda/envs/msba/lib/python3.12/site-packages/sklearn/utils/_param_validation.py:218)     # When the function is just a wrapper around an estimator, we allow
    [219](https://file+.vscode-resource.vscode-cdn.net/Users/vnijs/gh/pyrsm/examples/model/~/miniconda/envs/msba/lib/python3.12/site-packages/sklearn/utils/_param_validation.py:219)     # the function to delegate validation to the estimator, but we replace
    [220](https://file+.vscode-resource.vscode-cdn.net/Users/vnijs/gh/pyrsm/examples/model/~/miniconda/envs/msba/lib/python3.12/site-packages/sklearn/utils/_param_validation.py:220)     # the name of the estimator by the name of the function in the error
    [221](https://file+.vscode-resource.vscode-cdn.net/Users/vnijs/gh/pyrsm/examples/model/~/miniconda/envs/msba/lib/python3.12/site-packages/sklearn/utils/_param_validation.py:221)     # message to avoid confusion.
    [222](https://file+.vscode-resource.vscode-cdn.net/Users/vnijs/gh/pyrsm/examples/model/~/miniconda/envs/msba/lib/python3.12/site-packages/sklearn/utils/_param_validation.py:222)     msg = re.sub(
    [223](https://file+.vscode-resource.vscode-cdn.net/Users/vnijs/gh/pyrsm/examples/model/~/miniconda/envs/msba/lib/python3.12/site-packages/sklearn/utils/_param_validation.py:223)         r"parameter of \w+ must be",
    [224](https://file+.vscode-resource.vscode-cdn.net/Users/vnijs/gh/pyrsm/examples/model/~/miniconda/envs/msba/lib/python3.12/site-packages/sklearn/utils/_param_validation.py:224)         f"parameter of {func.__qualname__} must be",
    [225](https://file+.vscode-resource.vscode-cdn.net/Users/vnijs/gh/pyrsm/examples/model/~/miniconda/envs/msba/lib/python3.12/site-packages/sklearn/utils/_param_validation.py:225)         str(e),
    [226](https://file+.vscode-resource.vscode-cdn.net/Users/vnijs/gh/pyrsm/examples/model/~/miniconda/envs/msba/lib/python3.12/site-packages/sklearn/utils/_param_validation.py:226)     )

File ~/miniconda/envs/msba/lib/python3.12/site-packages/sklearn/inspection/_partial_dependence.py:682, in partial_dependence(estimator, X, features, sample_weight, categorical_features, feature_names, response_method, percentiles, grid_resolution, method, kind)
    [676](https://file+.vscode-resource.vscode-cdn.net/Users/vnijs/gh/pyrsm/examples/model/~/miniconda/envs/msba/lib/python3.12/site-packages/sklearn/inspection/_partial_dependence.py:676)     averaged_predictions = _partial_dependence_recursion(
    [677](https://file+.vscode-resource.vscode-cdn.net/Users/vnijs/gh/pyrsm/examples/model/~/miniconda/envs/msba/lib/python3.12/site-packages/sklearn/inspection/_partial_dependence.py:677)         estimator, grid, features_indices
    [678](https://file+.vscode-resource.vscode-cdn.net/Users/vnijs/gh/pyrsm/examples/model/~/miniconda/envs/msba/lib/python3.12/site-packages/sklearn/inspection/_partial_dependence.py:678)     )
    [680](https://file+.vscode-resource.vscode-cdn.net/Users/vnijs/gh/pyrsm/examples/model/~/miniconda/envs/msba/lib/python3.12/site-packages/sklearn/inspection/_partial_dependence.py:680) # reshape averaged_predictions to
    [681](https://file+.vscode-resource.vscode-cdn.net/Users/vnijs/gh/pyrsm/examples/model/~/miniconda/envs/msba/lib/python3.12/site-packages/sklearn/inspection/_partial_dependence.py:681) # (n_outputs, n_values_feature_0, n_values_feature_1, ...)
--> [682](https://file+.vscode-resource.vscode-cdn.net/Users/vnijs/gh/pyrsm/examples/model/~/miniconda/envs/msba/lib/python3.12/site-packages/sklearn/inspection/_partial_dependence.py:682) averaged_predictions = averaged_predictions.reshape(
    [683](https://file+.vscode-resource.vscode-cdn.net/Users/vnijs/gh/pyrsm/examples/model/~/miniconda/envs/msba/lib/python3.12/site-packages/sklearn/inspection/_partial_dependence.py:683)     -1, *[val.shape[0] for val in values]
    [684](https://file+.vscode-resource.vscode-cdn.net/Users/vnijs/gh/pyrsm/examples/model/~/miniconda/envs/msba/lib/python3.12/site-packages/sklearn/inspection/_partial_dependence.py:684) )
    [685](https://file+.vscode-resource.vscode-cdn.net/Users/vnijs/gh/pyrsm/examples/model/~/miniconda/envs/msba/lib/python3.12/site-packages/sklearn/inspection/_partial_dependence.py:685) pdp_results = Bunch(grid_values=values)
    [687](https://file+.vscode-resource.vscode-cdn.net/Users/vnijs/gh/pyrsm/examples/model/~/miniconda/envs/msba/lib/python3.12/site-packages/sklearn/inspection/_partial_dependence.py:687) if kind == "average":

ValueError: cannot reshape array of size 1 into shape (2)

Versions

System:
    python: 3.12.7 | packaged by Anaconda, Inc. | (main, Oct  4 2024, 08:22:19) [Clang 14.0.6 ]
executable: /Users/vnijs/miniconda/envs/msba/bin/python
   machine: macOS-14.2.1-arm64-arm-64bit

Python dependencies:
      sklearn: 1.6.1
          pip: 24.3.1
   setuptools: 75.1.0
        numpy: 2.2.1
        scipy: 1.15.1
       Cython: None
       pandas: 2.2.3
   matplotlib: 3.10.0
       joblib: 1.4.2
threadpoolctl: 3.5.0

Built with OpenMP: True

threadpoolctl info:
       user_api: openmp
   internal_api: openmp
    num_threads: 16
         prefix: libomp
       filepath: /Users/vnijs/miniconda/envs/msba/lib/python3.12/site-packages/sklearn/.dylibs/libomp.dylib
        version: None
@vnijs vnijs added Bug Needs Triage Issue requires triage labels Jan 20, 2025
@lesteve
Copy link
Member

lesteve commented Jan 20, 2025

Thanks for the issue, I confirm your snippet works with 1.5.2 and fails with 1.6.1, so this seems like a regression in 1.6.

This looked somewhat similar to #30271 but is slightly different ...

@lesteve lesteve added Regression and removed Needs Triage Issue requires triage labels Jan 20, 2025
@lesteve
Copy link
Member

lesteve commented Jan 20, 2025

Hmmm actually I misremembered and thought the PR fixing the issue was merged in time for 1.6.1 but it is only in main #26202.

Right now on main your snippet gives the following warning:

/home/lesteve/dev/scikit-learn/sklearn/inspection/_partial_dependence.py:709: FutureWarning: The column 1 contains integer data. Partial dependence plots are not supported for integer data: this can lead to implicit rounding with NumPy arrays or even errors with newer pandas versions. Please convert numerical featuresto floating point dtypes ahead of time to avoid problems. This will raise ValueError in scikit-learn 1.9.
  warnings.warn(

@vnijs if you wish you can you try installing the development version following these instructions.

@lesteve
Copy link
Member

lesteve commented Jan 23, 2025

@glemaitre when you have the time, is this worth thinking about a 1.6.2 release for this, or do you think this can wait for 1.7?

@vnijs
Copy link
Author

vnijs commented Feb 19, 2025

@lesteve Thanks for following up. If I am the only one reporting then maybe this is not so urgent. I'm using a docker container with my 150 students and have fixed usage to 1.5.2 for now.

@MarcBresson MarcBresson mentioned this issue Mar 26, 2025
Open
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Assignees
No one assigned
Labels
Bug Regression
Projects
None yet
Milestone
No milestone
Development

No branches or pull requests
2 participants
@lesteve @vnijs