update pacakges, NumPy > 2.0

Author: pplonski

.github/workflows/run-tests.yml

@@ -4,14 +4,12 @@ on: [ push,pull_request ]
 
 jobs:
   build:
-
     runs-on: ${{ matrix.os }}
     strategy:
+      fail-fast: false
       matrix:
         os: [ ubuntu-latest ]
-        python-version: [ '3.10']
-        #os: [ ubuntu-latest, macos-latest, windows-latest ]
-        #python-version: [ '3.8', '3.9', '3.10', '3.11' ]
+        python-version: [ "3.9", "3.12" ]
 
     steps:
       - name: Install OS Dependencies
@@ -29,21 +27,19 @@ jobs:
         if: matrix.os == 'windows-latest'
         run: |
           choco install graphviz
-      - uses: actions/checkout@v2
+      - uses: actions/checkout@v4
       - name: Set up Python ${{ matrix.python-version }}
-        uses: actions/setup-python@v2
+        uses: actions/setup-python@v5
         with:
           python-version: ${{ matrix.python-version }}
       - name: Install Python Dependencies
         run: |
           python -m pip install --upgrade pip
-          pip install --upgrade setuptools
+          pip install --upgrade setuptools wheel
           pip install -U importlib-metadata>=1.7.0
           pip install -U -r requirements.txt
           pip install -U -r requirements_dev.txt
-          pip install ipython
-          python setup.py install
+          pip install -e .
       - name: Test with pytest
         run: |
           pytest tests --cov=supervised/
-    continue-on-error: true

.github/workflows/smoke-pypi-install-train.yml

@@ -0,0 +1,47 @@
+name: Smoke test PyPI install + train
+
+on:
+  push:
+  pull_request:
+  workflow_dispatch:
+
+jobs:
+  smoke:
+    name: Smoke (${{ matrix.python-version }}, ${{ matrix.os }})
+    runs-on: ${{ matrix.os }}
+    strategy:
+      fail-fast: false
+      matrix:
+        os: [ubuntu-latest, windows-latest, macos-13, macos-14]
+        python-version: ["3.9", "3.12"]
+
+    steps:
+      - name: Set up Python
+        uses: actions/setup-python@v5
+        with:
+          python-version: ${{ matrix.python-version }}
+
+      - name: Install Graphviz
+        if: matrix.os == 'ubuntu-latest'
+        run: |
+          sudo apt-get update
+          sudo apt-get -y install graphviz
+
+      - name: Install Graphviz
+        if: startsWith(matrix.os, 'macos')
+        run: |
+          brew install graphviz
+
+      - name: Install Graphviz
+        if: matrix.os == 'windows-latest'
+        run: |
+          choco install graphviz
+
+      - name: Install package from PyPI
+        run: |
+          python -m pip install --upgrade pip
+          pip install -U mljar-supervised
+
+      - name: Smoke train
+        run: |
+          python -c "from sklearn.datasets import make_classification; from supervised import AutoML; X,y=make_classification(n_samples=120,n_features=8,n_informative=4,random_state=42); automl=AutoML(mode='Explain',algorithms=['Baseline','Linear','Decision Tree'],train_ensemble=False,stack_models=False,explain_level=0,total_time_limit=30,results_path='automl_smoke'); automl.fit(X,y); preds=automl.predict(X[:5]); print(preds)"

README.md

@@ -474,6 +474,10 @@ For details please check [mljar-supervised docs](https://supervised.mljar.com).
 
 # Installation  
 
+Compatibility:
+- Python `>=3.9`
+- NumPy `>=2.0,<3`
+
 From PyPi repository:
 
 ```
@@ -496,15 +500,15 @@ python setup.py install
 Installation for development
 ```
 git clone https://github.com/mljar/mljar-supervised.git
-virtualenv venv --python=python3.6
+virtualenv venv --python=python3.9
 source venv/bin/activate
 pip install -r requirements.txt
 pip install -r requirements_dev.txt
 ```
 
 Running in the docker:
 ```
-FROM python:3.7-slim-buster
+FROM python:3.9-slim-buster
 RUN apt-get update && apt-get -y update
 RUN apt-get install -y build-essential python3-pip python3-dev
 RUN pip3 -q install pip --upgrade

requirements.txt

@@ -1,16 +1,16 @@
-numpy>=1.19.5,<2
-pandas>=2.0.0
-scipy>=1.6.1
+numpy>=2.0.0,<3.0.0
+pandas>=2.2.2
+scipy>=1.13.1
 scikit-learn>=1.5.0
-xgboost>=2.0.0
-lightgbm>=3.0.0
-catboost>=0.24.4
+xgboost>=2.1.0
+lightgbm>=4.3.0
+catboost>=1.2.8
 joblib>=1.0.1
 tabulate>=0.8.7
-matplotlib>=3.2.2
+matplotlib>=3.8.4
 dtreeviz>=2.2.2
-shap>=0.42.1
-seaborn>=0.11.1
+shap>=0.46.0
+seaborn>=0.13.2
 optuna-integration>=3.6.0
 mljar-scikit-plot>=0.3.11
 markdown

setup.py

@@ -21,13 +21,13 @@
     packages=find_packages(exclude=["*.tests", "*.tests.*", "tests.*", "tests"]),
     install_requires=open("requirements.txt").readlines(),
     include_package_data=True,
-    python_requires='>=3.8',
+    python_requires='>=3.9',
     classifiers=[
         "Programming Language :: Python",
-        "Programming Language :: Python :: 3.8",
         "Programming Language :: Python :: 3.9",
         "Programming Language :: Python :: 3.10",
         "Programming Language :: Python :: 3.11",
+        "Programming Language :: Python :: 3.12",
     ],
     keywords=[
         "automated machine learning",

supervised/algorithms/xgboost.py

@@ -78,6 +78,7 @@ def __init__(self, params):
         self.early_stopping_rounds = additional.get("early_stopping_rounds", 50)
         self.learner_params = {
             "tree_method": "hist",
+            "device": "cpu",
             "booster": "gbtree",
             "objective": self.params.get("objective"),
             "eval_metric": self.params.get("eval_metric"),
@@ -166,7 +167,7 @@ def fit(
         dtrain = xgb.DMatrix(
             X.values if isinstance(X, pd.DataFrame) else X,
             label=y,
-            missing=np.NaN,
+            missing=np.nan,
             weight=sample_weight,
         )
 
@@ -176,7 +177,7 @@ def fit(
                 if isinstance(X_validation, pd.DataFrame)
                 else X_validation,
                 label=y_validation,
-                missing=np.NaN,
+                missing=np.nan,
                 weight=sample_weight_validation,
             )
         else:
@@ -264,7 +265,7 @@ def predict(self, X):
             raise XgbAlgorithmException("Xgboost model is None")
 
         dtrain = xgb.DMatrix(
-            X.values if isinstance(X, pd.DataFrame) else X, missing=np.NaN
+            X.values if isinstance(X, pd.DataFrame) else X, missing=np.nan
         )
         # xgboost > 2.0.0 version
         if hasattr(self.model, "best_iteration"):

supervised/base_automl.py

@@ -805,7 +805,7 @@ def _build_dataframe(self, X, y=None, sample_weight=None, sensitive_features=Non
             sensitive_features.reset_index(drop=True, inplace=True)
 
             for col in sensitive_features.columns:
-                if not sensitive_features[col].dtype.name in ["category", "object"]:
+                if pd.api.types.is_numeric_dtype(sensitive_features[col].dtype):
                     self.verbose_print("Sensitive features should be categorical")
                     self.verbose_print(
                         f"Apply automatic binarization for feature {col}"

supervised/preprocessing/preprocessing_categorical.py

@@ -58,13 +58,15 @@ def transform(self, X):
                 lbl = LabelEncoder()
                 lbl.from_json(lbl_params)
                 transformed_values = lbl.transform(X.loc[:, column])
+                col_dtype = X.loc[:, column].dtype
                 # check for pandas FutureWarning: Setting an item
                 # of incompatible dtype is deprecated and will raise
                 # in a future error of pandas.
-                if transformed_values.dtype != X.loc[:, column].dtype and \
-                    (X.loc[:, column].dtype == bool or X.loc[:, column].dtype == int):
+                if transformed_values.dtype != col_dtype and \
+                    (col_dtype == bool or col_dtype == int):
                     X = X.astype({column: transformed_values.dtype})
-                if isinstance(X[column].dtype, pd.CategoricalDtype):
+                # pandas string/categorical extension dtypes reject integer arrays.
+                if isinstance(col_dtype, pd.CategoricalDtype) or pd.api.types.is_string_dtype(col_dtype):
                     X[column] = X[column].astype('object')
                 X.loc[:, column] = transformed_values
 

supervised/utils/automl_plots.py

@@ -23,7 +23,7 @@
 import warnings
 import matplotlib.pyplot as plt
 
-warnings.filterwarnings("ignore", category=np.VisibleDeprecationWarning)
+warnings.filterwarnings("ignore", category=DeprecationWarning)
 
 class AutoMLPlots:
     # Original filename definitions
@@ -366,4 +366,3 @@ def models_correlation(results_path, models):
              # Ensure plot is closed if error occurred during saving/closing
             plt.close("all")
 
-

supervised/utils/leaderboard_plots.py

@@ -14,7 +14,7 @@
 
 import matplotlib.pyplot as plt
 
-warnings.filterwarnings("ignore", category=np.VisibleDeprecationWarning)
+warnings.filterwarnings("ignore", category=DeprecationWarning)
 
 
 markers = {