Differences

This shows you the differences between two versions of the page.

--- code:demo2d [2015/09/28 14:36]
asa created
+++ code:demo2d [2015/09/28 14:51]
asa
@@ Line 1: / Line 1: @@
-Displaying the decision boundary of a classifier
+===== Displaying the decision boundary of a classifier =====
-=================================================
+Below is Python code for displaying the decision boundary of a classifier.
+To use it:
+<file python using_demo2d.py>
+import demo2d
+from sklearn import svm
+demo2d.get_data()
+classifier = svm.LinearSVC()
+demo2d.decision_surface(classifier)
+</file>
+<file python demo2d.py>
+import numpy
+from numpy import arange
+import matplotlib
+from matplotlib import pylab
+pylab.rcParams['contour.negative_linestyle'] = 'solid'
+"""
+demo2d:  display decision boundaries and contours of the decision function
+of a classifier on two dimensional data.
+USAGE::
+  first you need to generate some data; you need to call
+  demo2d.generate_data()
+  data is generated by clicking '1' or '2' at positions on the figure
+  where you want your data points to be.
+  click 'q' when you're done.
+  demo2d.decision_surface(classifier) then plots the decision boundary and
+  contours of the discriminant function of the given classifier on the data
+  that was generated.
+  demo2d.decision_surface can be called several times using different classifiers.
+"""
+X = []
+Y = []
+#plotStr = ['or', 'ob']
+plotStr = ['or', '+b']
+xmin = -1
+xmax = 1
+ymin = -1
+ymax = 1
+def pick(event) :
+    global X
+    global Y
+    key_to_class = {'1' : 0, '2' : 1}
+    if event.key == 'q' :
+        if len(X) == 0 : return
+        print 'done creating data.  close this window and use the decisionSurface function'
+        pylab.disconnect(binding_id)
+    if event.key =='1' or event.key == '2' :
+        if event.inaxes is not None:
+            print 'data coords', event.xdata, event.ydata
+            X.append([event.xdata, event.ydata])
+            Y.append(key_to_class[event.key])
+            pylab.plot([event.xdata], [event.ydata],
+                       plotStr[int(event.key) - 1])
+            pylab.draw()
+def get_data(**args) :
+    pylab.subplot(111)
+    pylab.plot([xmin,xmin,xmax,xmax], [ymin,ymax,ymin,ymax], '.k')
+    pylab.title("press the numbers 1 or 2 to generate data points and 'q' to quit")
+    global binding_id
+    binding_id = pylab.connect('key_press_event', pick)
+    pylab.show()
+def scatter(X, Y, **args) :
+    markersize = 5
+    if 'markersize' in args :
+        markersize = args['markersize']
+    for i in range(len(X)) :
+        pylab.plot(X[i][0], X[i][1], plotStr[Y[i]], markersize=markersize)
+def decision_surface(classifier, fileName = None, **args) :
+    global X
+    global Y
+    classifier.fit(X, Y)
+    numContours = 3
+    if 'numContours' in args :
+        numContours = args['numContours']
+    title = None
+    if 'title' in args :
+        title = args['title']
+    markersize=5
+    fontsize = 'medium'
+    if 'markersize' in args :
+        markersize = args['markersize']
+    if 'fontsize' in args :
+        fontsize = args['fontsize']
+    contourFontsize = 10
+    if 'contourFontsize' in args :
+        contourFontsize = args['contourFontsize']
+    showColorbar = False
+    if 'showColorbar' in args :
+        showColorbar = args['showColorbar']
+    show = True
+    if fileName is not None :
+        show = False
+    if 'show' in args :
+        show = args['show']
+    # setting up the grid
+    delta = 0.01
+    if 'delta' in args :
+        delta = args['delta']
+    x = arange(xmin, xmax, delta)
+    y = arange(ymin, ymax, delta)
+    Z = numpy.zeros((len(x), len(y)), numpy.float)
+    gridX = numpy.zeros((len(x) *len(y), 2), numpy.float)
+    n = 0
+    for i in range(len(x)) :
+        for j in range(len(y)) :
+            gridX[n][0] = x[i]
+            gridX[n][1] = y[j]
+            n += 1
+    results = classifier.decision_function(gridX)
+    n = 0
+    for i in range(len(x)) :
+        for j in range(len(y)) :
+            Z[i][j] = results[n]
+            n += 1
+    #pylab.figure()
+    im = pylab.imshow(numpy.transpose(Z),
+                      interpolation='bilinear', origin='lower',
+                      cmap=pylab.cm.gray, extent=(xmin,xmax,ymin,ymax) )
+    if numContours == 1 :
+        C = pylab.contour(numpy.transpose(Z),
+                          [0],
+                          origin='lower',
+                          linewidths=(3),
+                          colors = 'black',
+                          extent=(xmin,xmax,ymin,ymax))
+    elif numContours == 3 :
+        C = pylab.contour(numpy.transpose(Z),
+                          [-1,0,1],
+                          origin='lower',
+                          linewidths=(1,3,1),
+                          colors = 'black',
+                          extent=(xmin,xmax,ymin,ymax))
+    else :
+        C = pylab.contour(numpy.transpose(Z),
+                          numContours,
+                          origin='lower',
+                          linewidths=2,
+                          extent=(xmin,xmax,ymin,ymax))
+    pylab.clabel(C,
+                 inline=1,
+                 fmt='%1.1f',
+                 fontsize=contourFontsize)
+    # plot the data
+    scatter(X, Y, markersize=markersize)
+    xticklabels = pylab.getp(pylab.gca(), 'xticklabels')
+    yticklabels = pylab.getp(pylab.gca(), 'yticklabels')
+    pylab.setp(xticklabels, fontsize=fontsize)
+    pylab.setp(yticklabels, fontsize=fontsize)
+    if title is not None :
+        pylab.title(title, fontsize=fontsize)
+    if showColorbar :
+        pylab.colorbar(im)
+    # colormap:
+    pylab.hot()
+    if fileName is not None :
+        pylab.savefig(fileName)
+    if show :
+        pylab.show()
+</file>

CS545 fall 2016

User Tools

Site Tools

Differences

Page Tools