### Example 6 - Setup data for whitening demonstration

Source: https://rasbt.github.io/mlxtend/user_guide/feature_extraction/PrincipalComponentAnalysis/

Imports and loads the wine dataset, splitting into training and test sets with stratification for the whitening example.

```python
from sklearn.preprocessing import StandardScaler
from sklearn.model_selection import train_test_split
from mlxtend.data import wine_data

X, y = wine_data()
X_train, X_test, y_train, y_test = train_test_split(X, y, random_state=123, test_size=0.3, stratify=y)
```

--------------------------------

### Example 5 - Setup data and train-test split

Source: https://rasbt.github.io/mlxtend/user_guide/feature_extraction/PrincipalComponentAnalysis/

Imports required libraries and loads the wine dataset, splitting into training and test sets with stratification.

```python
from sklearn.pipeline import make_pipeline
from sklearn.neighbors import KNeighborsClassifier
from sklearn.preprocessing import StandardScaler
from sklearn.model_selection import train_test_split
from mlxtend.data import wine_data

X, y = wine_data()
X_train, X_test, y_train, y_test = train_test_split(X, y, random_state=123, test_size=0.3, stratify=y)
```

--------------------------------

### Example 1 - Scaling a Pandas DataFrame: Setup and initial arrays

Source: https://rasbt.github.io/mlxtend/user_guide/preprocessing/shuffle_arrays_unison/

Initializes NumPy arrays X and y and prints them. This is the first part of the example, showing the data before shuffling.

```python
import numpy as np
from mlxtend.preprocessing import shuffle_arrays_unison
X = np.array([[1, 2, 3], [4, 5, 6], [7, 8, 9]])
y = np.array([1, 2, 3])
print('X:\n%s' % X)
print('y:\n%s' % y)
```

```text
X:
[[1 2 3]
 [4 5 6]
 [7 8 9]]
y:
[1 2 3]
```

--------------------------------

### Setup for zoom factor examples

Source: https://rasbt.github.io/mlxtend/user_guide/plotting/plot_decision_regions/

Loads the Iris dataset and trains an SVM classifier, which is used in the subsequent zoom factor examples. The output shows the fitted SVC object.

```python
from mlxtend.plotting import plot_decision_regions
import matplotlib.pyplot as plt
from sklearn import datasets
from sklearn.svm import SVC

# Loading some example data
iris = datasets.load_iris()
X = iris.data[:, [0, 2]]
y = iris.target

# Training a classifier
svm = SVC(C=0.5, kernel='linear')
svm.fit(X, y)
```

```text
SVC(C=0.5, kernel='linear')
```

--------------------------------

### Install mlxtend from source distribution with uv

Source: https://rasbt.github.io/mlxtend/installation/

Use this command if you encounter problems with the default binary distribution from PyPI. It installs mlxtend from the source distribution.

```bash
uv add --no-binary-package mlxtend mlxtend
```

--------------------------------

### Initialize dataset for GridSearch example

Source: https://rasbt.github.io/mlxtend/user_guide/feature_selection/SequentialFeatureSelector/

Sets up the Iris dataset for Example 8, splitting into training and test sets with test_size=0.2 and random_state=123.

```python
# Initialize the dataset

from sklearn.neighbors import KNeighborsClassifier
from sklearn.datasets import load_iris
from sklearn.model_selection import train_test_split

iris = load_iris()
X, y = iris.data, iris.target
X_train, X_test, y_train, y_test = train_test_split(
         X, y, test_size=0.2, random_state=123)
```

--------------------------------

### Install latest development version of mlxtend from GitHub

Source: https://rasbt.github.io/mlxtend/installation/

Installs the latest development version of mlxtend directly from the GitHub repository.

```bash
uv add "mlxtend @ git+https://github.com/rasbt/mlxtend.git"
```

--------------------------------

### Example git log output

Source: https://rasbt.github.io/mlxtend/CONTRIBUTING/

Example output of 'git log' showing commits from newest to oldest.

```bash
commit 046e3af8a9127df8eac879454f029937c8a31c41
Author: rasbt <mail@sebastianraschka.com>
Date:   Tue Nov 24 03:46:37 2015 -0500

    fixed setup.py

commit c3c00f6ba0e8f48bbe1c9081b8ae3817e57ecc5c
Author: rasbt <mail@sebastianraschka.com>
Date:   Tue Nov 24 03:04:39 2015 -0500

        documented feature x

commit d87934fe8726c46f0b166d6290a3bf38915d6e75
Author: rasbt <mail@sebastianraschka.com>
Date:   Tue Nov 24 02:44:45 2015 -0500

        added support for feature x
```

--------------------------------

### Install wheel and source distribution in isolated environments

Source: https://rasbt.github.io/mlxtend/CONTRIBUTING/

Installs the distribution files in isolated uv environments to verify they work. The filenames change with each version.

```bash
$ uv run --isolated --with ./dist/mlxtend-*.tar.gz python -c 'import mlxtend; print(mlxtend.__version__)'
$ uv run --isolated --with ./dist/mlxtend-*-py3-none-any.whl python -c 'import mlxtend; print(mlxtend.__version__)'
```

--------------------------------

### Import libraries and set up DummyClassifier for GroupTimeSeriesSplit examples

Source: https://rasbt.github.io/mlxtend/user_guide/evaluate/GroupTimeSeriesSplit/

Imports required libraries including numpy, pandas, sklearn's DummyClassifier and cross_val_score, and mlxtend's GroupTimeSeriesSplit, plot_splits, print_cv_info, and print_split_info. This setup is used in the following examples.

```python
import numpy as np
import pandas as pd
from sklearn.dummy import DummyClassifier
from sklearn.model_selection import cross_val_score

from mlxtend.evaluate.time_series import (
    GroupTimeSeriesSplit,
    plot_splits,
    print_cv_info,
    print_split_info,
)
```

--------------------------------

### Install from PyPI test server and check import

Source: https://rasbt.github.io/mlxtend/CONTRIBUTING/

Installs mlxtend from the PyPI test server and checks that it imports, using the test index and the real PyPI as fallback.

```bash
$ uv run --isolated --default-index https://test.pypi.org/simple/ --index https://pypi.org/simple --with mlxtend python -c 'import mlxtend; print(mlxtend.__version__)'
```

--------------------------------

### Example output of git fetch upstream

Source: https://rasbt.github.io/mlxtend/CONTRIBUTING/

Example output showing the fetch progress and the new branch upstream/master.

```bash
remote: Counting objects: xx, done.
remote: Compressing objects: 100% (xx/xx), done.
remote: Total xx (delta xx), reused xx (delta x)
Unpacking objects: 100% (xx/xx), done.
From https://github.com/rasbt/mlxtend
 * [new branch]      master     -> upstream/master
```

--------------------------------

### Start interactive rebase to squash commits

Source: https://rasbt.github.io/mlxtend/CONTRIBUTING/

Start an interactive rebase for the last 3 commits to squash them into one.

```bash
$ git rebase -i HEAD~3
```

--------------------------------

### Example output of git remote -v

Source: https://rasbt.github.io/mlxtend/CONTRIBUTING/

Example output showing the origin remote pointing to your fork. This is the expected output before adding the upstream remote.

```bash
origin  https://github.com/<your username>/mlxtend.git (fetch)
origin  https://github.com/<your username>/mlxtend.git (push)
```

--------------------------------

### Serve documentation locally with mkdocs serve

Source: https://rasbt.github.io/mlxtend/CONTRIBUTING/

Run from the mlxtend/docs directory to preview the documentation locally. The command is executed from the docs directory.

```bash
~/github/mlxtend/docs$ mkdocs serve
```

--------------------------------

### Setup for ROC curve with StackingCVClassifier

Source: https://rasbt.github.io/mlxtend/user_guide/classifier/StackingClassifier/

Loads iris data, binarizes the target, splits into train/test, and initializes base classifiers and a StackingClassifier wrapped in OneVsRestClassifier. This setup is used for both predict_proba and decision_function ROC examples.

```python
from sklearn import model_selection
from sklearn.linear_model import LogisticRegression
from sklearn.neighbors import KNeighborsClassifier
from sklearn.svm import SVC
from sklearn.ensemble import RandomForestClassifier
from mlxtend.classifier import StackingCVClassifier
from sklearn.metrics import roc_curve, auc
import numpy as np
from sklearn.model_selection import train_test_split
from sklearn import datasets
from sklearn.preprocessing import label_binarize
from sklearn.multiclass import OneVsRestClassifier


iris = datasets.load_iris()
X, y = iris.data[:, [0, 1]], iris.target


# Binarize the output
y = label_binarize(y, classes=[0, 1, 2])
n_classes = y.shape[1]



RANDOM_SEED = 42


X_train, X_test, y_train, y_test = train_test_split(
    X, y, test_size=0.33, random_state=RANDOM_SEED)

clf1 =  LogisticRegression()
clf2 = RandomForestClassifier(random_state=RANDOM_SEED)
clf3 = SVC(random_state=RANDOM_SEED)
lr = LogisticRegression()


sclf = StackingClassifier(classifiers=[clf1, clf2, clf3],
                          meta_classifier=lr)


# Learn to predict each class against the other
classifier = OneVsRestClassifier(sclf)
```

--------------------------------

### Run mlxtend from a local checkout

Source: https://rasbt.github.io/mlxtend/installation/

Clone your fork of the mlxtend repository, set up the development environment, and run mlxtend from your local checkout.

```bash
git clone https://github.com/<your_username>/mlxtend.git
cd mlxtend
uv sync --group dev
uv run python -c "import mlxtend; print(mlxtend.__version__)"
```

--------------------------------

### Example 5 - Build and evaluate PCA pipeline

Source: https://rasbt.github.io/mlxtend/user_guide/feature_extraction/PrincipalComponentAnalysis/

Creates a pipeline with StandardScaler, PrincipalComponentAnalysis (n_components=3), and KNeighborsClassifier (n_neighbors=5), then fits and evaluates on training and test sets.

```python
pipe_pca = make_pipeline(StandardScaler(),
                         PrincipalComponentAnalysis(n_components=3),
                         KNeighborsClassifier(n_neighbors=5))

pipe_pca.fit(X_train, y_train)


print('Transf. training accyracy: %.2f%%' % (pipe_pca.score(X_train, y_train)*100))
print('Transf. test accyracy: %.2f%%' % (pipe_pca.score(X_test, y_test)*100))
```

```text
Transf. training accyracy: 96.77%
Transf. test accyracy: 96.30%
```

--------------------------------

### Custom legend labels with plot_decision_regions

Source: https://rasbt.github.io/mlxtend/user_guide/plotting/plot_decision_regions/

Shows how to set custom legend labels by getting the handles and labels from the axis object returned by plot_decision_regions and passing them to ax.legend. The first block shows the core pattern; the second block is a complete example using the Iris dataset and an SVM classifier.

```python
ax = plot_decision_regions(X, y, clf=svm, legend=0)

handles, labels = ax.get_legend_handles_labels()
ax.legend(handles, 
          ['class 0', 'class 1', 'class 2'], 
           framealpha=0.3, scatterpoints=1)
```

```python
from mlxtend.plotting import plot_decision_regions
import matplotlib.pyplot as plt
from sklearn import datasets
from sklearn.svm import SVC

# Loading some example data
iris = datasets.load_iris()
X = iris.data[:, [0, 2]]
y = iris.target

# Training a classifier
svm = SVC(C=0.5, kernel='linear')
svm.fit(X, y)


# Plotting decision regions
ax = plot_decision_regions(X, y, clf=svm, legend=0)

# Adding axes annotations
plt.xlabel('sepal length [cm]')
plt.ylabel('petal length [cm]')
plt.title('SVM on Iris')

handles, labels = ax.get_legend_handles_labels()
ax.legend(handles, 
          ['class square', 'class triangle', 'class circle'], 
           framealpha=0.3, scatterpoints=1)

plt.show()
```

--------------------------------

### API example for tokenizer_emoticons

Source: https://rasbt.github.io/mlxtend/user_guide/text/tokenizer/

API documentation example showing the usage of tokenizer_emoticons. The output is a list of emoticons.

```python
>>> tokenizer_emoticons('</a>This :) is :( a test :-)!')
[':)', ':(', ':-)']
```

--------------------------------

### Build source distribution and wheel with uv build

Source: https://rasbt.github.io/mlxtend/CONTRIBUTING/

Creates the source distribution and wheel in the ./dist directory.

```bash
$ uv build
```

--------------------------------

### Example output of git merge upstream/master

Source: https://rasbt.github.io/mlxtend/CONTRIBUTING/

Example output showing a fast-forward merge with changed files.

```bash
Updating xxx...xxx
Fast-forward
SOME FILE1                    |    12 +++++++
SOME FILE2                    |    10 +++++++
2 files changed, 22 insertions(+),
```

--------------------------------

### API example for tokenizer_words_and_emoticons

Source: https://rasbt.github.io/mlxtend/user_guide/text/tokenizer/

API documentation example showing the usage of tokenizer_words_and_emoticons. The output is a list of lowercase words and emoticons.

```python
>>> tokenizer_words_and_emoticons('</a>This :) is :( a test :-)!')
['this', 'is', 'a', 'test', ':)', ':(', ':-)']
```

--------------------------------

### Bootstrap example using mlxtend.evaluate.bootstrap

Source: https://rasbt.github.io/mlxtend/api_subpackages/mlxtend.evaluate/

Shows how to compute the bootstrap statistic, standard error, and 95% confidence interval for the mean of a random sample. Requires numpy and mlxtend.evaluate.bootstrap. The output shows the mean, standard error, and confidence interval.

```python
>>> from mlxtend.evaluate import bootstrap
>>> rng = np.random.RandomState(123)
>>> x = rng.normal(loc=5., size=100)
>>> original, std_err, ci_bounds = bootstrap(x,
...                                          num_rounds=1000,
...                                          func=np.mean,
...                                          ci=0.95,
...                                          seed=123)
>>> print('Mean: %.2f, SE: +/- %.2f, CI95: [%.2f, %.2f]' % (original,
...                                                         std_err,
...                                                         ci_bounds[0],
...                                                         ci_bounds[1]))
Mean: 5.03, SE: +/- 0.11, CI95: [4.80, 5.26]
>>>
```

--------------------------------

### Build static HTML files with mkdocs build --clean

Source: https://rasbt.github.io/mlxtend/CONTRIBUTING/

Builds the static HTML files of the mlxtend documentation after checking locally.

```bash
~/github/mlxtend/docs$ mkdocs build --clean
```

--------------------------------

### tokenizer_emoticons usage example

Source: https://rasbt.github.io/mlxtend/api_subpackages/mlxtend.text/

Shows how to extract emoticons from a text string using tokenizer_emoticons. The example returns a list of emoticons found in the input text.

```python
>>> tokenizer_emoticons('</a>This :) is :( a test :-)!')
    [':)', ':(', ':-)']
```

--------------------------------

### Build API documentation with make_api.py

Source: https://rasbt.github.io/mlxtend/CONTRIBUTING/

Run from the mlxtend/docs directory to generate API documentation into the sources/api_modules and sources/api_subpackes directories.

```python
~/github/mlxtend/docs$ python make_api.py
```

--------------------------------

### List current remote repositories

Source: https://rasbt.github.io/mlxtend/CONTRIBUTING/

Run this command to see the configured remote repositories for your fork. The output shows the origin remote pointing to your fork.

```bash
$ git remote -v
```

--------------------------------

### HTTPError 403 Forbidden example

Source: https://rasbt.github.io/mlxtend/CONTRIBUTING/

Example error message that may appear if twine is outdated. Ensure twine is up to date in the uv development environment.

```text
HTTPError: 403 Forbidden from https://upload.pypi.org/legacy/
```

--------------------------------

### Set up StackingCVRegressor and GridSearchCV

Source: https://rasbt.github.io/mlxtend/user_guide/regressor/StackingCVRegressor/

Imports necessary libraries, loads the Boston dataset, creates base regressors (Lasso, Ridge) and a meta regressor (RandomForestRegressor), and sets up a StackingCVRegressor with use_features_in_secondary=True. Defines a parameter grid for GridSearchCV and fits the grid search to the data.

```python
from mlxtend.regressor import StackingCVRegressor
from sklearn.datasets import load_boston
from sklearn.linear_model import Lasso
from sklearn.linear_model import Ridge
from sklearn.ensemble import RandomForestRegressor
from sklearn.model_selection import GridSearchCV

X, y = load_boston(return_X_y=True)

ridge = Ridge(random_state=RANDOM_SEED)
lasso = Lasso(random_state=RANDOM_SEED)
rf = RandomForestRegressor(random_state=RANDOM_SEED)

stack = StackingCVRegressor(regressors=(lasso, ridge),
                            meta_regressor=rf, 
                            random_state=RANDOM_SEED,
                            use_features_in_secondary=True)

params = {'lasso__alpha': [0.1, 1.0, 10.0],
          'ridge__alpha': [0.1, 1.0, 10.0]}

grid = GridSearchCV(
    estimator=stack, 
    param_grid={
        'lasso__alpha': [x/5.0 for x in range(1, 10)],
        'ridge__alpha': [x/20.0 for x in range(1, 10)],
        'meta_regressor__n_estimators': [10, 100]
    }, 
    cv=5,
    refit=True
)

grid.fit(X, y)

print("Best: %f using %s" % (grid.best_score_, grid.best_params_))
```

--------------------------------

### Fit Kmeans and print convergence info

Source: https://rasbt.github.io/mlxtend/user_guide/cluster/Kmeans/

Initializes Kmeans with k=3, max_iter=50, random_seed=1, and print_progress=3, fits the model, and prints the number of iterations and final centroids. The output shows convergence after 2 iterations.

```python
from mlxtend.cluster import Kmeans

km = Kmeans(k=3, 
            max_iter=50, 
            random_seed=1, 
            print_progress=3)

km.fit(X)

print('Iterations until convergence:', km.iterations_)
print('Final centroids:\n', km.centroids_)
```

```text
Iteration: 2/50 | Elapsed: 00:00:00 | ETA: 00:00:00

Iterations until convergence: 2
Final centroids:
 [[-1.5947298   2.92236966]
 [ 2.06521743  0.96137409]
 [ 0.9329651   4.35420713]]
```

--------------------------------

### Example 1: Standard Accuracy

Source: https://rasbt.github.io/mlxtend/user_guide/evaluate/accuracy_score/

Computes the overall accuracy as the fraction of correct predictions over all samples. The example uses a 3-class dataset and prints the accuracy score.

```python
import numpy as np
from mlxtend.evaluate import accuracy_score


y_targ = [0, 0, 0, 1, 1, 1, 2, 2, 2]
y_pred = [1, 0, 0, 0, 1, 2, 0, 2, 2]

accuracy_score(y_targ, y_pred)
```

```text
0.5555555555555556
```

--------------------------------

### Create uv environment with development tools

Source: https://rasbt.github.io/mlxtend/CONTRIBUTING/

From the mlxtend repository root, creates a uv environment with Python 3.11 and the dev group.

```bash
$ uv sync --group dev --python 3.11
```

--------------------------------

### Generate PDF of documentation with md2pdf.py

Source: https://rasbt.github.io/mlxtend/CONTRIBUTING/

Run from the mlxtend/docs directory to generate a PDF version of the documentation.

```bash
python md2pdf.py
```

--------------------------------

### tokenizer_words_and_emoticons usage example

Source: https://rasbt.github.io/mlxtend/api_subpackages/mlxtend.text/

Shows how to convert text to lowercase words and emoticons using tokenizer_words_and_emoticons. The example returns a list of words and emoticons from the input text.

```python
>>> tokenizer_words_and_emoticons('</a>This :) is :( a test :-)!')
    ['this', 'is', 'a', 'test', ':)', ':(', ':-)']
```