Add contour kwargs to plot_decision_regions.

.github/workflows/python-package-conda.yml

@@ -1,6 +1,10 @@
-name: Python Package using Conda
+name: Tests
 
-on: [push]
+on:
+  push:
+    branches: master
+  pull_request:
+    branches: master
 
 jobs:
   build-linux:
@@ -30,9 +34,15 @@ jobs:
         flake8 . --count --exit-zero --max-complexity=10 --max-line-length=127 --statistics
     - name: Test with pytest
       run: |
-        conda install numpy scipy scikit-learn pandas tensorflow joblib pytest
-        conda install imageio scikit-image
-        conda install -c conda-forge dlib
+        conda config --add channels conda-forge
+        conda update numpy scipy scikit-learn pandas -y -q
+        conda install tensorflow joblib pytest -y -q
+        conda install imageio scikit-image -y -q
+        conda install dlib -y -q
         pip install markdown
         pip install -e  .
+        python -c "import numpy; print('NumPy:', numpy.__version__)"
+        python -c "import scipy; print('SciPy:', scipy.__version__)"
+        python -c "import sklearn; print('Scikit-learn:', sklearn.__version__)"
+        python -c "import pandas; print('Pandas:', pandas.__version__)"
         pytest -sv

docs/sources/CHANGELOG.md

@@ -17,9 +17,10 @@ The CHANGELOG for the current development version is available at
 
 
 
-##### New Features
+##### New Features and Enhancements
 
 - The `mlxtend.evaluate.bootstrap_point632_score` now supports `fit_params`. ([#861](https://github.com/rasbt/mlxtend/pull/861))
+- The `mlxtend/plotting/decision_regions.py` function now has a `contourf_kwargs` for matplotlib to change the look of the decision boundaries if desired. ([#881](https://github.com/rasbt/mlxtend/pull/881) via [[pbloem](https://github.com/pbloem)])
 
 ##### Changes
 

docs/sources/user_guide/plotting/plot_decision_regions.ipynb

@@ -1055,7 +1055,7 @@
      "text": [
       "## plot_decision_regions\n",
       "\n",
-      "*plot_decision_regions(X, y, clf, feature_index=None, filler_feature_values=None, filler_feature_ranges=None, ax=None, X_highlight=None, zoom_factor=1.0, legend=1, hide_spines=True, markers='s^oxv<>', colors='#1f77b4,#ff7f0e,#3ca02c,#d62728,#9467bd,#8c564b,#e377c2,#7f7f7f,#bcbd22,#17becf', scatter_kwargs=None, contourf_kwargs=None, scatter_highlight_kwargs=None)*\n",
+      "*plot_decision_regions(X, y, clf, feature_index=None, filler_feature_values=None, filler_feature_ranges=None, ax=None, X_highlight=None, zoom_factor=1.0, legend=1, hide_spines=True, markers='s^oxv<>', colors='#1f77b4,#ff7f0e,#3ca02c,#d62728,#9467bd,#8c564b,#e377c2,#7f7f7f,#bcbd22,#17becf', scatter_kwargs=None, contourf_kwargs=None, contour_kwargs=None, scatter_highlight_kwargs=None)*\n",
       "\n",
       "Plot decision regions of a classifier.\n",
       "\n",
@@ -1072,69 +1072,90 @@
       "\n",
       "    Feature Matrix.\n",
       "\n",
+      "\n",
       "- `y` : array-like, shape = [n_samples]\n",
       "\n",
       "    True class labels.\n",
       "\n",
+      "\n",
       "- `clf` : Classifier object.\n",
       "\n",
       "    Must have a .predict method.\n",
       "\n",
+      "\n",
       "- `feature_index` : array-like (default: (0,) for 1D, (0, 1) otherwise)\n",
       "\n",
       "    Feature indices to use for plotting. The first index in\n",
       "    `feature_index` will be on the x-axis, the second index will be\n",
       "    on the y-axis.\n",
       "\n",
+      "\n",
       "- `filler_feature_values` : dict (default: None)\n",
       "\n",
       "    Only needed for number features > 2. Dictionary of feature\n",
       "    index-value pairs for the features not being plotted.\n",
       "\n",
+      "\n",
       "- `filler_feature_ranges` : dict (default: None)\n",
       "\n",
       "    Only needed for number features > 2. Dictionary of feature\n",
       "    index-value pairs for the features not being plotted. Will use the\n",
       "    ranges provided to select training samples for plotting.\n",
       "\n",
+      "\n",
       "- `ax` : matplotlib.axes.Axes (default: None)\n",
       "\n",
       "    An existing matplotlib Axes. Creates\n",
       "    one if ax=None.\n",
       "\n",
+      "\n",
       "- `X_highlight` : array-like, shape = [n_samples, n_features] (default: None)\n",
       "\n",
       "    An array with data points that are used to highlight samples in `X`.\n",
       "\n",
+      "\n",
       "- `zoom_factor` : float (default: 1.0)\n",
       "\n",
       "    Controls the scale of the x- and y-axis of the decision plot.\n",
       "\n",
+      "\n",
       "- `hide_spines` : bool (default: True)\n",
       "\n",
       "    Hide axis spines if True.\n",
       "\n",
+      "\n",
       "- `legend` : int (default: 1)\n",
       "\n",
       "    Integer to specify the legend location.\n",
       "    No legend if legend is 0.\n",
       "\n",
+      "\n",
       "- `markers` : str (default: 's^oxv<>')\n",
       "\n",
       "    Scatterplot markers.\n",
       "\n",
+      "\n",
       "- `colors` : str (default: 'red,blue,limegreen,gray,cyan')\n",
       "\n",
       "    Comma separated list of colors.\n",
       "\n",
+      "\n",
       "- `scatter_kwargs` : dict (default: None)\n",
       "\n",
       "    Keyword arguments for underlying matplotlib scatter function.\n",
       "\n",
+      "\n",
       "- `contourf_kwargs` : dict (default: None)\n",
       "\n",
       "    Keyword arguments for underlying matplotlib contourf function.\n",
       "\n",
+      "\n",
+      "- `contour_kwargs` : dict (default: None)\n",
+      "\n",
+      "    Keyword arguments for underlying matplotlib contour function\n",
+      "    (which draws the lines between decision regions).\n",
+      "\n",
+      "\n",
       "- `scatter_highlight_kwargs` : dict (default: None)\n",
       "\n",
       "    Keyword arguments for underlying matplotlib scatter function.\n",
@@ -1157,12 +1178,19 @@
     "with open('../../api_modules/mlxtend.plotting/plot_decision_regions.md', 'r') as f:\n",
     "    print(f.read())"
    ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {},
+   "outputs": [],
+   "source": []
   }
  ],
  "metadata": {
   "anaconda-cloud": {},
   "kernelspec": {
-   "display_name": "Python 3",
+   "display_name": "Python 3 (ipykernel)",
    "language": "python",
    "name": "python3"
   },
@@ -1176,7 +1204,7 @@
    "name": "python",
    "nbconvert_exporter": "python",
    "pygments_lexer": "ipython3",
-   "version": "3.8.8"
+   "version": "3.9.6"
   },
   "toc": {
    "nav_menu": {},

mlxtend/classifier/tests/test_ensemble_vote_classifier.py

@@ -20,7 +20,7 @@
 from mlxtend.data import iris_data
 from mlxtend.utils import assert_raises
 
-from distutils.version import LooseVersion as Version
+from packaging.version import Version
 from sklearn import __version__ as sklearn_version
 
 X, y = iris_data()
@@ -198,7 +198,7 @@ def test_EnsembleVoteClassifier_gridsearch():
     params = {'logisticregression__C': [1.0, 100.0],
               'randomforestclassifier__n_estimators': [20, 200]}
 
-    if Version(sklearn_version) < '0.24.1':
+    if Version(sklearn_version) < Version('0.24.1'):
         grid = GridSearchCV(estimator=eclf, param_grid=params, cv=5, iid=False)
     else:
         grid = GridSearchCV(estimator=eclf, param_grid=params, cv=5)
@@ -225,7 +225,7 @@ def test_EnsembleVoteClassifier_gridsearch_enumerate_names():
               'randomforestclassifier__n_estimators': [5, 20],
               'voting': ['hard', 'soft']}
 
-    if Version(sklearn_version) < '0.24.1':
+    if Version(sklearn_version) < Version('0.24.1'):
         grid = GridSearchCV(estimator=eclf, param_grid=params, cv=5, iid=False)
     else:
         grid = GridSearchCV(estimator=eclf, param_grid=params, cv=5)
@@ -261,7 +261,7 @@ def test_classifier_gridsearch():
 
     params = {'clfs': [[clf1, clf1, clf1], [clf2, clf3]]}
 
-    if Version(sklearn_version) < '0.24.1':
+    if Version(sklearn_version) < Version('0.24.1'):
         grid = GridSearchCV(estimator=eclf,
                             param_grid=params,
                             iid=False,

mlxtend/classifier/tests/test_stacking_classifier.py

@@ -5,7 +5,7 @@
 # License: BSD 3 clause
 
 import random
-from distutils.version import LooseVersion as Version
+from packaging.version import Version
 
 import numpy as np
 import pytest

mlxtend/classifier/tests/test_stacking_cv_classifier.py

@@ -6,7 +6,7 @@
 # License: BSD 3 clause
 
 import random
-from distutils.version import LooseVersion as Version
+from packaging.version import Version
 
 import numpy as np
 import pandas as pd
@@ -206,7 +206,7 @@ def test_gridsearch():
     params = {'meta_classifier__C': [1.0, 100.0],
               'randomforestclassifier__n_estimators': [20, 200]}
 
-    if Version(sklearn_version) < '0.24.1':
+    if Version(sklearn_version) < Version('0.24.1'):
         grid = GridSearchCV(estimator=sclf, param_grid=params, cv=5, iid=False)
     else:
         grid = GridSearchCV(estimator=sclf, param_grid=params, cv=5)
@@ -233,7 +233,7 @@ def test_gridsearch_enumerate_names():
               'randomforestclassifier-2__n_estimators': [5, 20],
               'use_probas': [True, False]}
 
-    if Version(sklearn_version) < '0.24.1':
+    if Version(sklearn_version) < Version('0.24.1'):
         grid = GridSearchCV(estimator=sclf, param_grid=params, cv=5, iid=False)
     else:
         grid = GridSearchCV(estimator=sclf, param_grid=params, cv=5)
@@ -405,7 +405,7 @@ def test_classifier_gridsearch():
 
     params = {'classifiers': [[clf1], [clf1, clf2, clf3]]}
 
-    if Version(sklearn_version) < '0.24.1':
+    if Version(sklearn_version) < Version('0.24.1'):
         grid = GridSearchCV(estimator=sclf,
                             param_grid=params,
                             cv=5,

mlxtend/evaluate/tests/test_bootstrap_point632.py

@@ -167,7 +167,6 @@ def test_scoring_proba():
                                           random_seed=123)
 
 
-
 if 'TRAVIS' in os.environ or os.environ.get('TRAVIS') == 'true':
     TRAVIS = True
 else:
@@ -178,15 +177,16 @@ def test_scoring_proba():
 else:
     APPVEYOR = False
 
+
 @pytest.mark.skipif(TRAVIS or APPVEYOR, reason="TensorFlow dependency")
 def test_keras_fitparams():
     import tensorflow as tf
-    
+
     model = tf.keras.Sequential([
-    tf.keras.layers.Dense(32, activation=tf.nn.relu),
-    tf.keras.layers.Dense(1)])
+        tf.keras.layers.Dense(32, activation=tf.nn.relu),
+        tf.keras.layers.Dense(1)])
 
     optimizer = tf.keras.optimizers.Adam()
     model.compile(loss='mean_squared_error', optimizer=optimizer)
 
-    model.fit(X, y, epochs=5)
+    model.fit(X, y, epochs=5)

mlxtend/evaluate/tests/test_combined_ftest_5x2cv.py

@@ -12,7 +12,7 @@
 from sklearn.linear_model import Ridge
 from sklearn.tree import DecisionTreeClassifier
 from sklearn.model_selection import train_test_split
-from distutils.version import LooseVersion as Version
+from packaging.version import Version
 from sklearn import __version__ as sklearn_version
 
 

mlxtend/evaluate/tests/test_holdout.py

@@ -12,7 +12,7 @@
 from mlxtend.evaluate import PredefinedHoldoutSplit
 from mlxtend.data import iris_data
 
-from distutils.version import LooseVersion as Version
+from packaging.version import Version
 from sklearn import __version__ as sklearn_version
 
 
@@ -68,7 +68,7 @@ def test_randomholdoutsplit_in_sfs():
 def test_randomholdoutsplit_in_grid():
     params = {'n_neighbors': [1, 2, 3, 4, 5]}
 
-    if Version(sklearn_version) < '0.24.1':
+    if Version(sklearn_version) < Version('0.24.1'):
         grid = GridSearchCV(KNeighborsClassifier(),
                             iid=False,
                             param_grid=params,
@@ -135,7 +135,7 @@ def test_predefinedholdoutsplit_in_sfs():
 def test_predefinedholdoutsplit_in_grid():
     params = {'n_neighbors': [1, 3, 5]}
 
-    if Version(sklearn_version) < '0.24.1':
+    if Version(sklearn_version) < Version('0.24.1'):
         grid = GridSearchCV(KNeighborsClassifier(),
                             param_grid=params,
                             iid=False,

mlxtend/evaluate/tests/test_paired_ttest_5x2cv.py

@@ -12,7 +12,7 @@
 from sklearn.linear_model import Ridge
 from sklearn.tree import DecisionTreeClassifier
 from sklearn.model_selection import train_test_split
-from distutils.version import LooseVersion as Version
+from packaging.version import Version
 from sklearn import __version__ as sklearn_version
 
 

mlxtend/evaluate/tests/test_paired_ttest_resampled.py

@@ -14,7 +14,7 @@
 from sklearn.linear_model import Ridge
 from sklearn.tree import DecisionTreeClassifier
 from sklearn.model_selection import train_test_split
-from distutils.version import LooseVersion as Version
+from packaging.version import Version
 from sklearn import __version__ as sklearn_version
 
 

mlxtend/feature_selection/tests/test_column_selector.py

@@ -14,7 +14,7 @@
 from sklearn.pipeline import make_pipeline
 from sklearn.model_selection import GridSearchCV
 
-from distutils.version import LooseVersion as Version
+from packaging.version import Version
 from sklearn import __version__ as sklearn_version
 
 

mlxtend/feature_selection/tests/test_exhaustive_feature_selector.py

@@ -6,7 +6,7 @@
 
 import numpy as np
 import pandas as pd
-from distutils.version import LooseVersion as Version
+from packaging.version import Version
 from numpy.testing import assert_almost_equal
 from mlxtend.feature_selection import ExhaustiveFeatureSelector as EFS
 from sklearn.ensemble import RandomForestClassifier