Similar to the case for svd (#9516), lstsq currently enforces double precision. Now that we expose the single-precision BLAS functions as gufuncs (#9980), there is no longer a need for this limitation. This is a reminder that we should allow them to be used, most easily perhaps through a new dtype argument that defaults to using double precision (for discussion, see #9980).
Similar to the case for
svd(#9516),lstsqcurrently enforces double precision. Now that we expose the single-precision BLAS functions asgufuncs(#9980), there is no longer a need for this limitation. This is a reminder that we should allow them to be used, most easily perhaps through a newdtypeargument that defaults to using double precision (for discussion, see #9980).