I'm experimenting with the examples/tutorials to get a feel for 
scikit-learn.

"Faces recognition example using eigenfaces and SVMs" example.

Windows 7, python 2.73., etc.  Installed all the packages to the latest 
versions and the example was crashing here:

 > face = np.asarray(imread(file_path)[slice_], dtype=np.float32)

in lfw.py

with an error to the effect of: can't slice a 0-dm array.

Eventually discovered that when you install PIL with pip, it does not 
install the jpeg decoder.  Only easy_install will install the decoder as 
well:

http://stackoverflow.com/questions/8891091/how-to-install-pil-package-with-jpeg-support-in-windows

I guess it makes sense that if you don't have the decoder, then the 
array from imread(file_path) will be without dimensions - but it took me 
a while to get there.

I also found that the following is necessary on Windows 7 with PIL code 
as simple as this:

 > import Image
 > im=Image.open("file.jpg")
 > im.show()

http://www.velocityreviews.com/forums/t707158-python-pil-and-vista-windows-7-show-not-working.html

That is, in ImageShow.py, for the Windows viewer, you need to substitute:

 > return "start /wait %s && PING 127.0.0.1 -n 5 > NUL && del /f %s" % 
(file, file)

for the current line #99.

Pat


------------------------------------------------------------------------------
Everyone hates slow websites. So do we.
Make your web apps faster with AppDynamics
Download AppDynamics Lite for free today:
http://p.sf.net/sfu/appdyn_d2d_mar
_______________________________________________
Scikit-learn-general mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/scikit-learn-general

Reply via email to