New Feature: Scatter Histogram

docs/mkdocs.yml

@@ -123,6 +123,7 @@ nav:
     - user_guide/plotting/plot_linear_regression.md
     - user_guide/plotting/plot_sequential_feature_selection.md
     - user_guide/plotting/scatterplotmatrix.md
+    - user_guide/plotting/scatter_hist.md
     - user_guide/plotting/stacked_barplot.md
   - preprocessing:
     - user_guide/preprocessing/CopyTransformer.md

docs/sources/CHANGELOG.md

@@ -16,6 +16,7 @@ The CHANGELOG for the current development version is available at
 ##### New Features
 
 - Adds a second "balanced accuracy" interpretation ("balanced") to `accuracy_score` in addition to the existing "average" option to compute the scikit-learn-style balanced accuracy. ([#764](https://github.com/rasbt/mlxtend/pull/764))
+- Adds new `scatter_hist` function to `mlxtend.plotting` for generating a scattered histogram. ([#757](https://github.com/rasbt/mlxtend/issues/757) via [Maitreyee Mhasaka](https://github.com/Maitreyee1))
 
 ##### Changes
 
@@ -834,4 +835,4 @@ imput arrays via `transform` and `fit_transform`
 
 ### Version 0.1.1 (2014-08-13)
 
-- Simplified code for ColumnSelector.
+- Simplified code for ColumnSelector.

docs/sources/USER_GUIDE_INDEX.md

@@ -95,6 +95,7 @@
 - [plot_linear_regression](user_guide/plotting/plot_linear_regression.md)
 - [plot_sequential_feature_selection](user_guide/plotting/plot_sequential_feature_selection.md)
 - [scatterplotmatrix](user_guide/plotting/scatterplotmatrix.md)
+- [scatter_hist](user_guide/plotting/scatter_hist.md)
 - [stacked_barplot](user_guide/plotting/stacked_barplot.md)
 
 ## `preprocessing`

mlxtend/plotting/__init__.py

@@ -19,8 +19,7 @@
 from .ecdf import ecdf
 from .scatterplotmatrix import scatterplotmatrix
 from .pca_correlation_graph import plot_pca_correlation_graph
-
-
+from .scatter_hist import scatter_hist
 __all__ = ["plot_learning_curves",
            "plot_decision_regions",
            "plot_confusion_matrix",
@@ -34,4 +33,5 @@
            "checkerboard_plot",
            "ecdf",
            "scatterplotmatrix",
-           "plot_pca_correlation_graph"]
+           "plot_pca_correlation_graph",
+           "scatter_hist"]

mlxtend/plotting/scatter_hist.py

@@ -0,0 +1,79 @@
+# Sebastian Raschka 2014-2020
+# mlxtend Machine Learning Library Extensions
+# Author: Sebastian Raschka <sebastianraschka.com>
+#
+# License: BSD 3 clause
+import pandas as pd
+import numpy as np
+import matplotlib.pyplot as plt
+
+
+def scatter_hist(x, y, xlabel=None, ylabel=None, figsize=(5, 5)):
+    """
+    Scatter plot and individual feature histograms along axes.
+
+    Parameters
+    ----------
+    x : 1D array-like or Pandas Series
+        X-axis values.
+
+    y : 1D array-like or Pandas Series
+        Y-axis values.
+
+    xlabel : str (default: `None`)
+        Label for the X-axis values. If `x` is a pandas Series,
+        and `xlabel` is `None`, the label is inferred automatically.
+
+    ylabel : str (default: `None`)
+        Label for the X-axis values. If `y` is a pandas Series,
+        and `ylabel` is `None`, the label is inferred automatically.
+
+    figsize : tuple (default: `(5, 5)`)
+        Matplotlib figure size.
+
+    Returns
+    ---------
+    plot : Matplotlib Figure object
+
+    """
+    left, width = 0.1, 0.65
+    bottom, height = 0.1, 0.65
+    spacing = 0.001
+    rect_scatter = [left, bottom, width, height]
+    rect_histx = [left, bottom + height + spacing, width, 0.2]
+    rect_histy = [left + width + spacing, bottom, 0.2, height]
+
+    if hasattr(x, 'values'):
+        x_values = x.values
+        if xlabel is None:
+            xlabel = x.name
+    else:
+        x_values = x
+
+    if hasattr(y, 'values'):
+        y_values = y.values
+        if ylabel is None:
+            ylabel = y.name
+    else:
+        y_values = y
+
+
+    fig = plt.figure(figsize=figsize)
+    ax = fig.add_axes(rect_scatter)
+
+    if xlabel is not None:
+        ax.set_xlabel(xlabel)
+    if ylabel is not None:
+        ax.set_ylabel(ylabel)
+
+    ax_histx = fig.add_axes(rect_histx, sharex=ax)
+    ax_histy = fig.add_axes(rect_histy, sharey=ax)
+    ax_histx.tick_params(axis="x", labelbottom=False)
+    ax_histy.tick_params(axis="y", labelleft=False)
+    ax_histx.axis("off")
+    ax_histy.axis("off")
+    ax_histx.hist(x_values, edgecolor='white', bins='auto')
+    ax_histy.hist(y_values, edgecolor='white', 
+                  orientation='horizontal', bins='auto')
+    plot = ax.scatter(x_values, y_values)
+    return plot

mlxtend/plotting/tests/test_scatter_hist.py

@@ -0,0 +1,19 @@
+from mlxtend.data import iris_data
+from mlxtend.plotting import scatter_hist
+import pandas as pd
+
+
+X, y = iris_data()
+df = pd.DataFrame(X)
+df.columns = (['sepal length [cm]',
+               'sepal width [cm]',
+               'petal length [cm]',
+               'petal width [cm]'])
+
+
+def test_pass_data_as_dataframe():
+    scatter_hist(df["sepal length [cm]"], df["sepal width [cm]"])
+
+
+def test_pass_data_as_numpy_array():
+    scatter_hist(X[:, 0], X[:, 1])