From 665c1c469e624db106a97c08486a43001555c036 Mon Sep 17 00:00:00 2001
From: Vytautas Jancauskas <unaudio@gmail.com>
Date: Sun, 17 Jun 2012 20:55:37 +0300
Subject: [PATCH 1/3] Added parallel coordinates plotting

---
 pandas/tools/plotting.py | 47 ++++++++++++++++++++++++++++++++++++++++
 1 file changed, 47 insertions(+)

diff --git a/pandas/tools/plotting.py b/pandas/tools/plotting.py
index 515434edda6b0..284c9bfb54c88 100644
--- a/pandas/tools/plotting.py
+++ b/pandas/tools/plotting.py
@@ -178,6 +178,53 @@ def random_color(column):
     ax.grid()
     return ax
 
+def parallel_coordinates(data, class_column, cols=None, ax=None, **kwds):
+    """Parallel coordinates plotting.
+
+    Parameters:
+    -----------
+    data: A DataFrame containing data to be plotted
+    class_column: Column name containing class names
+    cols: A list of column names to use, optional
+    ax: matplotlib axis object, optional
+    kwds: A list of keywords for matplotlib plot method
+
+    Returns:
+    --------
+    ax: matplotlib axis object
+    """
+    import matplotlib.pyplot as plt
+    import random
+    def random_color(column):
+        random.seed(column)
+        return [random.random() for _ in range(3)]
+    n = len(data)
+    classes = set(data[class_column])
+    class_col = data[class_column]
+    if cols == None:
+        columns = [data[col] for col in data.columns if (col != class_column)]
+    else:
+        columns = [data[col] for col in cols]
+    used_legends = set([])
+    x = range(len(columns))
+    if ax == None:
+        ax = plt.gca()
+    for i in range(n):
+        row = [columns[c][i] for c in range(len(columns))]
+        y = row
+        label = None
+        if str(class_col[i]) not in used_legends:
+            label = str(class_col[i])
+            used_legends.add(label)
+        ax.plot(x, y, color=random_color(class_col[i]), label=label, **kwds)
+    for i, col in enumerate(columns):
+        ax.axvline(i, linewidth=1, color='black')
+    ax.set_xticks(range(len(columns)))
+    ax.set_xticklabels([col for col in data.columns if col != class_column])
+    ax.legend(loc='upper right')
+    ax.grid()
+    return ax
+
 def lag_plot(series, ax=None, **kwds):
     """Lag plot for time series.
 

From b74a7cee67524016333597491798964068ee286f Mon Sep 17 00:00:00 2001
From: Vytautas Jancauskas <unaudio@gmail.com>
Date: Sun, 17 Jun 2012 21:02:03 +0300
Subject: [PATCH 2/3] Added tests for parallel coordinates plotting

---
 pandas/tests/test_graphics.py | 8 ++++++++
 1 file changed, 8 insertions(+)

diff --git a/pandas/tests/test_graphics.py b/pandas/tests/test_graphics.py
index 8326445f28cb0..8c1b512622513 100644
--- a/pandas/tests/test_graphics.py
+++ b/pandas/tests/test_graphics.py
@@ -271,6 +271,14 @@ def test_andrews_curves(self):
         df = read_csv(path)
         _check_plot_works(andrews_curves, df, 'Name')
 
+    @slow
+    def test_parallel_coordinates(self):
+        from pandas import read_csv
+        from pandas.tools.plotting import parallel_coordinates
+        path = os.path.join(curpath(), 'data/iris.csv')
+        df = read_csv(path)
+        _check_plot_works(parallel_coordinates, df, 'Name')
+
     @slow
     def test_plot_int_columns(self):
         df = DataFrame(np.random.randn(100, 4)).cumsum()

From f09b6a653f5362cabd527c737dcdb575e33ddb5b Mon Sep 17 00:00:00 2001
From: Vytautas Jancauskas <unaudio@gmail.com>
Date: Sun, 17 Jun 2012 21:14:23 +0300
Subject: [PATCH 3/3] Added parallel coordinates documentation

---
 doc/source/visualization.rst | 21 +++++++++++++++++++++
 1 file changed, 21 insertions(+)

diff --git a/doc/source/visualization.rst b/doc/source/visualization.rst
index 34bd803516468..dc89d99749d39 100644
--- a/doc/source/visualization.rst
+++ b/doc/source/visualization.rst
@@ -284,6 +284,27 @@ of the same class will usually be closer together and form larger structures.
    @savefig andrews_curves.png width=6in
    andrews_curves(data, 'Name')
 
+Parallel Coordinates
+~~~~~~~~~~~~~~~~~~~~
+
+Parallel coordinates is a plotting technique for plotting multivariate data.
+It allows one to see clusters in data and to estimate other statistics visually.
+Using parallel coordinates points are represented as connected line segments.
+Each vertical line represents one attribute. One set of connected line segments
+represents one data point. Points that tend to cluster will appear closer together.
+
+.. ipython:: python
+
+   from pandas import read_csv
+   from pandas.tools.plotting import parallel_coordinates
+
+   data = read_csv('data/iris.data')
+
+   plt.figure()
+
+   @savefig parallel_coordinates.png width=6in
+   parallel_coordinates(data, 'Name')
+
 Lag Plot
 ~~~~~~~~