Hi there ,
I am doing svm classification for remote sensing images usisg scikit learn.
My problem is that for image size of (2000,2000,4), the input to the
clf.predict is of size 2000x2000,4 i,e four millions rows and clf.predict
is taking forever (26 min) to give me the output.
Is there an effiecient way to tackle the problem. I use matlab normally
there is function called blkproc which process block by block. In
sckit-image there is function called view as blocks, but problem with that
is input l should be exactly divisible by block_shape provided. Is there
any other way which could take care of blocks which are at the edge
similar to blockproc in matlab.
Thanks for any input
view = view_as_blocks(l, block_shape)
------------------------------------------------------------------------------
Sponsored by Intel(R) XDK
Develop, test and display web and hybrid apps with a single code base.
Download it for free now!
http://pubads.g.doubleclick.net/gampad/clk?id=111408631&iu=/4140/ostg.clktrk
_______________________________________________
Scikit-learn-general mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/scikit-learn-general
