Problem in the MSAM

[Pruthvi3796]

I am want to do the msam like below.

from spectral import *
import gdal
d=ev.load_ENVI_spec_lib("D:\data\speclib.hdr")
img=gdal.Open("D:\data\sub_66")
M=img.ReadAsArray()
M=np.swapaxes(M,0,2)
n1=d[0]
n2=d[1]
n1=np.transpose(n1)
n1=n1[:425]
wavelength=n2['wavelength']
wavelength=np.float64(wavelength)
class1=msam(M,n1)

But it gives error: AssertionError: Matrix dimensions are not aligned. where M=>(210 row,275 column,425 band) shape and n1=(425 bandreflectance,16 spectra)

[Pruthvi3796]

Also clear that the classification through msam is possible?How?

[tboggs]

The last dimension of M and n1 must be the same but they are not in your case. The data arg should have shape (n_rows, n_cols, n_bands) and members should have shape (n_spectra, n_bands).

[Pruthvi3796]

Thanks for reply.I corrected it.But how to do classification using msam?Please Provide example if possible.

[tboggs]

SAM and MSAM do not do classification. They perform spectral unmixing. The returned array indicates the fractional abundance of each endmember for each pixel in the data argument.

[Pruthvi3796]

but it includes in the classification in the documentation.

On Sat, Apr 15, 2017 at 4:09 AM, Thomas Boggs notifications@github.com wrote:

SAM and MSAM do not do classification. They perform spectral unmixing. The returned array indicates the fractional abundance of each endmember for each pixel in the data argument.

— You are receiving this because you authored the thread. Reply to this email directly, view it on GitHub https://github.com/spectralpython/spectral/issues/64#issuecomment-294249904, or mute the thread https://github.com/notifications/unsubscribe-auth/AZtrF8Y48FNlUOJ8m8N26POLoKMA2nxhks5rv_WSgaJpZM4M6TN2 .

-- (http://aau.in)

[tboggs]

That's for organizational purposes. Those two algorithms don't select a class. They give an estimate of how much of each class is present in a pixel. If you really want to use them for classification, you can pick the class with the highest abundance or use the abundance vectors in a subsequent classifier.