I get this error when training a neural net using using Theano and Lasagne.
In particular, I get this error when taking the square or square root of a variable e.g., theano.tensor.sqr(input)
In case it helps:
I've installed python packages: mkl 2018.0.3, mkl-service 1.1.2, mkl_fft 1.0.4, mkl_random 1.0.1
Could you please execute your entire Python reproducer with environment variable MKL_VERBOSE=1 and provide the printed verbose info for the offensive DGELSD call.
DGELSD is solving least squared problem for a general matrix `A`. The parameter 4 is likely one of the dimensions of the matrix, but the requested output will determine this definitively.