1

I am using a confusion matrix to measure the performance of my classifier. This example would work fine for me (its from here), but I get the whole time TypeError: Invalid dimensions for image data 

from numpy import * import matplotlib.pyplot as plt from pylab import * conf_arr = [[50.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0], [3.0, 26.0, 0.0, 0.0, 0.0, 1.0, 0.0], [0.0, 0.0, 10.0, 0.0, 0.0, 0.0, 0.0], [4.0, 1.0, 0.0, 5.0, 0.0, 0.0, 0.0], [3.0, 0.0, 1.0, 0.0, 6.0, 0.0, 0.0], [0.0, 0.0, 0.0, 0.0, 0.0, 47.0, 0.0], [2.0, 0.0, 0.0, 0.0, 0.0, 0.0, 8.0]] norm_conf = [] for i in conf_arr: a = 0 tmp_arr = [] a = sum(i,0) for j in i: tmp_arr.append(float(j)/float(a)) norm_conf.append(tmp_arr) plt.clf() fig = plt.figure() ax = fig.add_subplot(111) res = ax.imshow(array(norm_conf), cmap=cm.jet, interpolation='nearest') cb = fig.colorbar(res) savefig("confmat.png", format="png")

I am new to python and matplotlib. Any help?

Matplot version is 1.1.1. and here is the full traceback:

after res =... I get 

TypeError Traceback (most recent call last) C:\Python27\lib\site-packages\SimpleCV\Shell\Shell.pyc in <module>() ----> 1 res = ax.imshow(array(norm_conf), cmap=cm.jet, interpolation='nearest') C:\Python27\lib\site-packages\matplotlib\axes.pyc in imshow(self, X, cmap, norm, aspect, interpolation, alpha, vmin, vmax, origin, extent, shape, filternorm, filter rad, imlim, resample, url, **kwargs) 6794 filterrad=filterrad, resample=resample, **kwargs) 6795 -> 6796 im.set_data(X) 6797 im.set_alpha(alpha) 6798 self._set_artist_props(im) C:\Python27\lib\site-packages\matplotlib\image.pyc in set_data(self, A) 409 if (self._A.ndim not in (2, 3) or 410 (self._A.ndim == 3 and self._A.shape[-1] not in (3, 4))): --> 411 raise TypeError("Invalid dimensions for image data") 412 413 self._imcache =None TypeError: Invalid dimensions for image data SimpleCV:105> cb = fig.colorbar(res)

For print norm_conf I get now results: [[1.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0], [1.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0], [1.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0],...]]. I corrected the indentation problem. But my confusion .png is pretty distorted. Further how should I proceed to label the squares in the matrix?

Improve this question
0

1 Answer 1

Reset to default
2

This works for me just fine (matplotlib 1.1.1rc). Originally I wanted you to confirm your matplotlib version and post the entire traceback -- by "traceback" I mean the few lines before the TypeError line which show what caused the error -- and that's still a good idea, but I think I see what the problem might be.

This is the error you'd get if norm_conf somehow weren't being filled (i.e. norm_conf = []):

Traceback (most recent call last): File "mdim2.py", line 19, in <module> res = ax.imshow(array(norm_conf), cmap=cm.jet, interpolation='nearest') File "/usr/lib/pymodules/python2.7/matplotlib/axes.py", line 6796, in imshow im.set_data(X) File "/usr/lib/pymodules/python2.7/matplotlib/image.py", line 411, in set_data raise TypeError("Invalid dimensions for image data") TypeError: Invalid dimensions for image data

Your code looks like it might have some indentation problems, which often happens when using mixed tabs and spaces. So I'd recommend (1) trying python -tt yourprogramname.py to see if there are whitespace errors, and (2) making sure that you're using 4-space tabs throughout.

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

3 Comments

@user1665514: okay, so it looks like we're seeing the same error. Can you run python -tt yourprogramname.py to check for whitespace problems?
@user1665514: you could also add print norm_conf before plt.clf() to see what its value is.
@user1665514: yep, then that's the problem. And the only reason I can think of that norm_conf isn't being filled with a value is because that loop isn't being executed due to whitespace problems.

Start asking to get answers

Find the answer to your question by asking.

Ask question

Explore related questions

See similar questions with these tags.