[SciPy-User] Extracting best fit from sparse.linalg.lsqr

[srean]

Hi,

 I want to obtain the best fit that lsqr has found, but without multiplying
A*x. It probably represents this vector somewhere internally but does not
return it. I do not want to incur the avoidable cost of an extra mat x vec.

 If you are familiar with the code, could you point me to what exactly
should I return, if I were to modify the code.

I tried to figure it out from
https://github.com/scipy/scipy/blob/a582c5a367dee4955ecc490b77232992f9d01eb7/scipy/sparse/linalg/isolve/lsqr.pybut
too sleep starved to see the light.

Thanks a lot
  -- srean
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.scipy.org/pipermail/scipy-user/attachments/20111016/d680a778/attachment.html>