Fix Compatibility issue

.github/workflows/test-pr.yml

@@ -15,7 +15,7 @@ jobs:
     strategy:
       fail-fast: false
       matrix:
-        python-version: ["3.9", "3.10", "3.11", "3.12"]
+        python-version: ["3.10", "3.11", "3.12"]
         os: [ubuntu-latest, macOS-latest, windows-latest]
     steps:
     - uses: actions/checkout@v2

.pre-commit-config.yaml

@@ -6,7 +6,7 @@ repos:
       - id: end-of-file-fixer
       - id: trailing-whitespace
   - repo: https://github.com/psf/black
-    rev: 24.2.0
+    rev: 26.3.0
     hooks:
       - id: black
   - repo: https://github.com/pycqa/flake8

benchmarks/consumer_complaints/scripts/data.py

@@ -1,5 +1,6 @@
 #!/usr/bin/env python3
 """Script to share common functions for data manipulation."""
+
 import csv
 from typing import TextIO
 

benchmarks/consumer_complaints/scripts/metrics.py

@@ -1,5 +1,6 @@
 #!/usr/bin/env python3
 """Script to compute metrics."""
+
 import argparse
 import sys
 from argparse import Namespace

benchmarks/consumer_complaints/scripts/predict.py

@@ -1,5 +1,6 @@
 #!/usr/bin/env python3
 """Script to predict with flat approach."""
+
 import argparse
 import pickle
 import sys

benchmarks/consumer_complaints/scripts/split_data.py

@@ -1,5 +1,6 @@
 #!/usr/bin/env python3
 """Script to split train and test data."""
+
 import argparse
 import sys
 from argparse import Namespace

benchmarks/consumer_complaints/scripts/statistics.py

@@ -1,5 +1,6 @@
 #!/usr/bin/env python3
 """Script to store basic statistical information."""
+
 import argparse
 import os
 import sys

benchmarks/consumer_complaints/scripts/train.py

@@ -1,5 +1,6 @@
 #!/usr/bin/env python3
 """Script to train with flat or hierarchical approaches."""
+
 import argparse
 import pickle
 import sys

benchmarks/consumer_complaints/scripts/tune.py

@@ -1,5 +1,6 @@
 #!/usr/bin/env python3
 """Script to perform hyper-parameter tuning for flat and hierarchical approaches."""
+
 import hashlib
 import json
 import logging
@@ -170,7 +171,7 @@ def load_trial(cfg: DictConfig) -> List[float]:
     md5 = compute_md5(hyperparameters)
     filename = f"{cfg.output_dir}/{md5}.sav"
     if os.path.exists(filename):
-        (_, scores) = pickle.load(open(filename, "rb"))
+        _, scores = pickle.load(open(filename, "rb"))
         log.info(f"Loaded trial with F-scores {scores}")
         return scores
     else:

benchmarks/consumer_complaints/scripts/tune_table.py

@@ -1,5 +1,6 @@
 #!/usr/bin/env python3
 """Script to create table with tuning results for flat and hierarchical approaches."""
+
 import argparse
 import glob
 import pickle

docs/examples/plot_binary_policies.py

@@ -51,6 +51,7 @@
 
    Mathematical definition on the different policies is given at :ref:`Training Policies`.
 """
+
 from sklearn.ensemble import RandomForestClassifier
 
 from hiclass import LocalClassifierPerNode

docs/examples/plot_calibration.py

@@ -137,6 +137,7 @@
 In the code below, isotonic regression is used to calibrate the model.
 
 """
+
 from sklearn.ensemble import RandomForestClassifier
 
 from hiclass import LocalClassifierPerNode

docs/examples/plot_empty_levels.py

@@ -11,6 +11,7 @@
 .. figure:: ../algorithms/local_classifier_per_node.svg
    :align: center
 """
+
 import numpy as np
 from sklearn.linear_model import LogisticRegression
 

docs/examples/plot_hello_hiclass.py

@@ -6,6 +6,7 @@
 
 A minimalist example showing how to use HiClass to train and predict.
 """
+
 from sklearn.ensemble import RandomForestClassifier
 
 from hiclass import LocalClassifierPerNode

docs/examples/plot_model_persistence.py

@@ -8,6 +8,7 @@
 Pickle can be used to easily store machine learning models on disk.
 In this example, we demonstrate how to use pickle to store and load trained classifiers.
 """
+
 import pickle
 
 from sklearn.linear_model import LogisticRegression

docs/examples/plot_parallel_training.py

@@ -15,6 +15,7 @@
 .. [1] https://www.ray.io/
 .. [2] https://www.kaggle.com/datasets/kashnitsky/hierarchical-text-classification
 """
+
 import sys
 from os import cpu_count
 from sklearn.feature_extraction.text import CountVectorizer, TfidfTransformer

docs/examples/plot_pipeline.py

@@ -7,6 +7,7 @@
 HiClass can be adopted in scikit-learn pipelines, and fully supports sparse matrices as input.
 This example desmonstrates the use of both of these features.
 """
+
 from sklearn.feature_extraction.text import CountVectorizer, TfidfTransformer
 from sklearn.linear_model import LogisticRegression
 from sklearn.pipeline import Pipeline

hiclass/ConstantClassifier.py

@@ -1,9 +1,10 @@
 """Constant classifier if there is only one class in the training set."""
 
 import numpy as np
+from sklearn.base import BaseEstimator
 
 
-class ConstantClassifier:
+class ConstantClassifier(BaseEstimator):
     """A classifier that always returns the only label seen during fit."""
 
     def fit(self, X, y, sample_weight=None):

hiclass/LocalClassifierPerLevel.py

@@ -431,7 +431,7 @@ def _fit_classifier(self, level, separator):
             md5 = hashlib.md5(str(level).encode("utf-8")).hexdigest()
             filename = f"{self.tmp_dir}/{md5}.sav"
             if exists(filename):
-                (_, classifier) = pickle.load(open(filename, "rb"))
+                _, classifier = pickle.load(open(filename, "rb"))
                 self.logger_.info(
                     f"Loaded trained model for local classifier {level} from file {filename}"
                 )

hiclass/LocalClassifierPerNode.py

@@ -404,7 +404,7 @@ def _fit_classifier(self, node):
             md5 = hashlib.md5(node.encode("utf-8")).hexdigest()
             filename = f"{self.tmp_dir}/{md5}.sav"
             if exists(filename):
-                (_, classifier) = pickle.load(open(filename, "rb"))
+                _, classifier = pickle.load(open(filename, "rb"))
                 self.logger_.info(
                     f"Loaded trained model for local classifier {node.split(self.separator_)[-1]} from file {filename}"
                 )

hiclass/LocalClassifierPerParentNode.py

@@ -372,7 +372,7 @@ def _fit_classifier(self, node):
             md5 = hashlib.md5(node.encode("utf-8")).hexdigest()
             filename = f"{self.tmp_dir}/{md5}.sav"
             if exists(filename):
-                (_, classifier) = pickle.load(open(filename, "rb"))
+                _, classifier = pickle.load(open(filename, "rb"))
                 self.logger_.info(
                     f"Loaded trained model for local classifier {node.split(self.separator_)[-1]} from file {filename}"
                 )

hiclass/_calibration/BinaryCalibrator.py

@@ -1,8 +1,9 @@
 import abc
 import numpy as np
+from sklearn.base import BaseEstimator
 
 
-class _BinaryCalibrator(abc.ABC):
+class _BinaryCalibrator(abc.ABC, BaseEstimator):
     @abc.abstractmethod
     def fit(
         self, y: np.ndarray, scores: np.ndarray, X: np.ndarray = None

setup.cfg

@@ -14,7 +14,7 @@ exclude = **/__init__.py,  docs/source/conf.py
 ;file.py: error
 
 [requires]
-python_version = ">=3.9,<3.13"
+python_version = ">=3.10,<3.13"
 
 # See the docstring in versioneer.py for instructions. Note that you must
 # re-run 'versioneer.py setup' after changing this section, and commit the

setup.py

@@ -23,7 +23,7 @@
 URL_ISSUES = "https://github.com/scikit-learn-contrib/hiclass/issues"
 EMAIL = "fabio.malchermiranda@hpi.de, Niklas.Koehnecke@student.hpi.uni-potsdam.de"
 AUTHOR = "Fabio Malcher Miranda, Niklas Koehnecke"
-REQUIRES_PYTHON = ">=3.9,<3.13"
+REQUIRES_PYTHON = ">=3.10,<3.13"
 KEYWORDS = ["hierarchical classification"]
 DACS_SOFTWARE = "https://gitlab.com/dacs-hpi"
 # What packages are required for this module to be executed?
@@ -39,7 +39,7 @@
         "pydocstyle",
         "pytest-cov==3.0.0",
         "pyfakefs==5.3.5",
-        "black==24.2.0",
+        "black==26.3.0",
         "pre-commit==2.20.0",
         "ray",
     ],
@@ -152,7 +152,6 @@ def run(self):
         "Operating System :: Unix",
         "Operating System :: MacOS",
         "Programming Language :: Python :: 3",
-        "Programming Language :: Python :: 3.9",
         "Programming Language :: Python :: 3.10",
         "Programming Language :: Python :: 3.11",
         "Programming Language :: Python :: 3.12",

tests/test_LocalClassifiers.py

@@ -124,7 +124,7 @@ def test_tmp_dir(classifier):
             filename = "0cc175b9c0f1b6a831c399e269772661.sav"
             expected_name = "a"
         assert patcher.fs.exists(filename)
-        (name, classifier) = pickle.load(open(filename, "rb"))
+        name, classifier = pickle.load(open(filename, "rb"))
         assert expected_name == name
         check_is_fitted(classifier)
         clf.fit(x, y)

tests/test_metrics.py

@@ -22,7 +22,6 @@
     adaptive_calibration_error,
 )
 
-
 # TODO: add tests for 3D dataframe (not sure if it's possible to have 3D dataframes)
 
 

versioneer.py

@@ -417,9 +417,7 @@ def run_command(commands, args, cwd=None, verbose=False, hide_stderr=False, env=
     return stdout, p.returncode
 
 
-LONG_VERSION_PY[
-    "git"
-] = r'''
+LONG_VERSION_PY["git"] = r'''
 # This file helps to compute a version number in source trees obtained from
 # git-archive tarball (such as those provided by githubs download-from-tag
 # feature). Distribution tarballs (built by setup.py sdist) and build