The input you provide is not correct. sdp_norm is a low-level function constructing normalization pipeline action. It takes dimensionality (scalar) and output (feature) labels as parameters:
>> help sdp_norm
SDP_NORM Pipeline action: Normalize soft output to sum to one
INPUT
DIM Input dimensionality
LAB Output feature labels
OUTPUT
PP Pipeline object
DESCRIPTION
Normalize the output to sum to one.
Example use:
>> a
'Fruit set' 260 by 2 sddata, 3 classes: 'apple'(100) 'banana'(100) 'stone'(60)
>> p=sdgauss(a)
Gaussian model pipeline 2x3 3 classes, 3 components (sdp_normal)
>> p.lab'
1 apple
2 banana
3 stone
>> pm=sdp_norm(3,p.lab)
Output normalization pipeline 3x3 (sdp_norm)
>> pfull=p*pm
sequential pipeline 2x3 'Gaussian model+Output normalization'
1 Gaussian model 2x3 3 classes, 3 components (sdp_normal)
2 Output normalization 3x3 (sdp_norm)
>> out=a*pfull
'Fruit set' 260 by 3 sddata, 3 classes: 'apple'(100) 'banana'(100) 'stone'(60)
>> out.featlab'
1 apple
2 banana
3 stone
Pavel
