matplotlib error in when saving pdf

Question

I have the following python code:

import numpy as np
import sklearn as sk
import pandas as pd
import scipy as sp
import matplotlib.pyplot as plt
import sys
from sklearn.decomposition import PCA
from sklearn.svm import OneClassSVM



np.random.seed(0)

x1 = np.random.normal((1,1), 0.1, (200, 2))
x2 = np.random.normal((1,0), .1, (200, 2))
x3 = np.random.normal((0,1), .1, (200, 2))
x4 = np.array([[0.5, 0.5],
               [0.5, 1],
               [1, 0.5]])


X = np.vstack([x1, x2, x3])#, x4])
X = sk.preprocessing.scale(X)

X_new = np.vstack([x1, x2, x3, x4])
X_new = sk.preprocessing.scale(X_new)



n, p = X_new.shape

anomaly_index = np.array(range(n-3, n))
normal_index = np.array(range(n-3))

#plt.scatter(X_new[normal_index,0], X_new[normal_index,1])
#plt.scatter(X_new[anomaly_index,0], X_new[anomaly_index,1], marker='*', c='r')
#plt.show()


gammas = [0.5]#, 0.7, 0.9]
nus = [0.01]#, 0.3, 0.8]
nrow = len(gammas)
ncol = len(nus)


j = 0
for gamma in gammas:
    for nu in nus:
        j += 1

        svm = OneClassSVM(kernel='rbf', degree=2, 
                          gamma=gamma, coef0=0.0, 
                          tol=0.001, 
                          nu=nu, shrinking=True, 
                          cache_size=200, 
                          verbose=False, 
                          max_iter=-1, random_state=None)
        svm.fit(X)

        anomaly_score =  - svm.decision_function(X_new)
        vmin = anomaly_score.min()
        vmax = anomaly_score.max()


        xx1, yy1 = np.meshgrid(np.linspace(X_new[:,0].min()-0.3, 
                                           X_new[:,0].max()+0.3, 1000), 
                               np.linspace(X_new[:,1].min()-0.3, 
                                           X_new[:,1].max()+0.3, 1000))
        Z1 = svm.decision_function(np.c_[xx1.ravel(), yy1.ravel()])
        Z1 = Z1.reshape(xx1.shape)



        plt.subplot(nrow, ncol, j)
        plt.title(r'$\gamma=$' + str(gamma) + r'   $\nu=$' + str(nu) + '')
        plt.scatter(X_new[normal_index, 0], X_new[normal_index, 1], 
                    c=anomaly_score[normal_index],  alpha=2, s=50, 
                    vmin=vmin, vmax=vmax)
        plt.scatter(X_new[anomaly_index, 0], 
                    X_new[anomaly_index, 1], marker='*',
                    c=anomaly_score[anomaly_index],  alpha=2, s=90,
                    vmin=vmin, vmax=vmax)

        plt.colorbar()
        #cb = plt.colorbar()
        #tick_locator = ticker.MaxNLocator(nbins=5)
        #cb.locator = tick_locator
        #cb.update_ticks()

        plt.contourf( xx1, yy1, Z1, cmap=plt.cm.Blues,
                      levels=np.linspace(Z1.min(), 0.3, 7), alpha=0.1)
        plt.xlim(X_new[:,0].min()-0.3, X_new[:, 0].max()+0.3)
        plt.ylim(X_new[:,1].min()-0.3, X_new[:, 1].max()+0.3)
        plt.xlabel(r'$x_1$', size=20)
        plt.ylabel(r'$x_2$', size=20)
        plt.locator_params(nbins=4)
        plt.tight_layout()
#plt.savefig('one_class_svm_3_clusters_grid.pdf')
plt.show()

This works fine but if I uncomment the plt.savefig I receive the following error:

Process Python[/home/donbeo/Documents/pythoncode/fault_detection/one_class_svm/oc_svm_3_clusters.py] finished
Python 3.4.0 (default, Jun 19 2015, 14:20:21) 
[GCC 4.8.2] on linux
Type "help", "copyright", "credits" or "license" for more information.
>>> >>> >>> >>> >>> 

Process Python[/home/donbeo/Documents/pythoncode/fault_detection/one_class_svm/oc_svm_3_clusters.py] finished
Python 3.4.0 (default, Jun 19 2015, 14:20:21) 
[GCC 4.8.2] on linux
Type "help", "copyright", "credits" or "license" for more information.
>>> >>> >>> >>> Traceback (most recent call last):
  File "/usr/local/lib/python3.4/dist-packages/matplotlib/colors.py", line 355, in to_rgba
    'number in rbga sequence outside 0-1 range')
ValueError: number in rbga sequence outside 0-1 range

During handling of the above exception, another exception occurred:

Traceback (most recent call last):
  File "<stdin>", line 1, in <module>
  File "/home/donbeo/Documents/pythoncode/fault_detection/one_class_svm/oc_svm_3_clusters.py", line 98, in <module>
    plt.savefig(plot_path + 'one_class_svm_3_clusters_grid.pdf')
  File "/usr/local/lib/python3.4/dist-packages/matplotlib/pyplot.py", line 577, in savefig
    res = fig.savefig(*args, **kwargs)
  File "/usr/local/lib/python3.4/dist-packages/matplotlib/figure.py", line 1476, in savefig
    self.canvas.print_figure(*args, **kwargs)
  File "/usr/local/lib/python3.4/dist-packages/matplotlib/backends/backend_qt5agg.py", line 161, in print_figure
    FigureCanvasAgg.print_figure(self, *args, **kwargs)
  File "/usr/local/lib/python3.4/dist-packages/matplotlib/backend_bases.py", line 2211, in print_figure
    **kwargs)
  File "/usr/local/lib/python3.4/dist-packages/matplotlib/backends/backend_pdf.py", line 2485, in print_pdf
    self.figure.draw(renderer)
  File "/usr/local/lib/python3.4/dist-packages/matplotlib/artist.py", line 59, in draw_wrapper
    draw(artist, renderer, *args, **kwargs)
  File "/usr/local/lib/python3.4/dist-packages/matplotlib/figure.py", line 1085, in draw
    func(*args)
  File "/usr/local/lib/python3.4/dist-packages/matplotlib/artist.py", line 59, in draw_wrapper
    draw(artist, renderer, *args, **kwargs)
  File "/usr/local/lib/python3.4/dist-packages/matplotlib/axes/_base.py", line 2110, in draw
    a.draw(renderer)
  File "/usr/local/lib/python3.4/dist-packages/matplotlib/artist.py", line 59, in draw_wrapper
    draw(artist, renderer, *args, **kwargs)
  File "/usr/local/lib/python3.4/dist-packages/matplotlib/collections.py", line 772, in draw
    Collection.draw(self, renderer)
  File "/usr/local/lib/python3.4/dist-packages/matplotlib/artist.py", line 59, in draw_wrapper
    draw(artist, renderer, *args, **kwargs)
  File "/usr/local/lib/python3.4/dist-packages/matplotlib/collections.py", line 320, in draw
    self._offset_position)
  File "/usr/local/lib/python3.4/dist-packages/matplotlib/backends/backend_pdf.py", line 1658, in draw_path_collection
    antialiaseds, urls, offset_position):
  File "/usr/local/lib/python3.4/dist-packages/matplotlib/backend_bases.py", line 488, in _iter_collection
    gc0.set_foreground(fg)
  File "/usr/local/lib/python3.4/dist-packages/matplotlib/backend_bases.py", line 1008, in set_foreground
    self._rgb = colors.colorConverter.to_rgba(fg)
  File "/usr/local/lib/python3.4/dist-packages/matplotlib/colors.py", line 376, in to_rgba
    'to_rgba: Invalid rgba arg "%s"\n%s' % (str(arg), exc))
ValueError: to_rgba: Invalid rgba arg "[ 0.  0.  0.  2.]"
number in rbga sequence outside 0-1 range
>>>

Can you reduce this to a minimal example? There is way too much going on in this one. — tacaswell
– tacaswell, Commented Jun 26, 2015 at 16:52

Julien Spronck · Accepted Answer · 2015-06-29 08:06:53Z

2

Look at the error message, it tells you what is wrong. The last line of your traceback shows that alpha (the fourth value) of your rgba color is set to 2, while it should be between 0 and 1.

As a small example,

x1 = np.random.normal((1,1), 0.1, (200, 2))
plt.scatter(x1[:,0], x1[:,1], alpha=2)
plt.show()

will give the same error message. Simply replace your alpha value by a number between 0 and 1 and the error goes away:

x1 = np.random.normal((1,1), 0.1, (200, 2))
plt.scatter(x1[:,0], x1[:,1], alpha=0.5)
plt.show()

In your code, change the alpha value in the following lines. Here, I replaced it by 0.5 but you can chose what you want as long as it is between 0 and 1.

plt.scatter(X_new[normal_index, 0], X_new[normal_index, 1], 
                c=anomaly_score[normal_index],  alpha=0.5, s=50, 
                vmin=vmin, vmax=vmax)
plt.scatter(X_new[anomaly_index, 0], 
                X_new[anomaly_index, 1], marker='*',
                c=anomaly_score[anomaly_index],  alpha=0.5, s=90,
                vmin=vmin, vmax=vmax)

edited Jun 29, 2015 at 8:06

answered Jun 27, 2015 at 15:12

Julien Spronck

15.5k5 gold badges50 silver badges57 bronze badges

Sign up to request clarification or add additional context in comments.

2 Comments

Donbeo Over a year ago

thanks a lot it works. It is strange the fact that I was able to save png plot with alpha=2

Julien Spronck Over a year ago

you're welcome :-) I don't know why you were able to save the png plot, strange indeed.

Collectives™ on Stack Overflow

matplotlib error in when saving pdf

1 Answer 1

2 Comments

Your Answer

Hot Network Questions

Collectives™ on Stack Overflow

1 Answer 1

2 Comments

Your Answer

Sign up or log in

Post as a guest

Related