Closed mobeets closed 8 years ago
Something like:
[U,S,V] = svd(L, 'econ'); % L = 88x10; U = 88x10;
minSpikeVal = zeros(1,88);
baselineSpikeVal = [1 2 ...]; % whatever this value is, 1x88
spikeCountStd = [1 1 ...]; % whatever this value is, 1x88
normalizedMinSpikeVal = diag(1./spikeCountStd)*(minSpikeVal - baselineSpikeVal);
d = [1e-14 ...]; % mean latent value, 1x10, but it's basically just zeros
minLatentVal = U' * (normalizedMinSpikeVal' - d);
minimal firing point: orthogonal point svd on L from FA -> U'*(0 - mu) where mu is FA mean