Thanks. I've filled [an issue](https://github.com/usnistgov/fipy/issues/531)

> On Nov 9, 2017, at 9:39 PM, James Pringle <[email protected]> wrote:
> 
> update -- as of scipy 0.19, linalg.splu had a depreciated option "drop_tol" 
> which did nothing: 
> https://docs.scipy.org/doc/scipy-0.19.1/reference/generated/generated/scipy.sparse.linalg.splu.html
> 
> As of version 1.0, the keyword seems to have been dropped entirely, and 
> should be removed from linearLUSolver.py. It had not done anything, and so 
> can be removed. 
> 
> This is the only place this keyword exists. 
> 
> Jamie
> 
> On Thu, Nov 9, 2017 at 6:16 PM, James Pringle <[email protected]> wrote:
> Dear all -- 
> 
>     I just installed fipy from the conda channel with a fresh install of 
> anaconda. It is based on python 2.7.14. When I run with the scipy solver it 
> crashes with 
> 
>   File 
> "/home/pringle/anaconda3/envs/MYFIPYENV27/lib/python2.7/site-packages/fipy/solvers/scipy/linearLUSolver.py",
>  line 64, in _solve_
>     permc_spec=3)
> TypeError: splu() got an unexpected keyword argument 'drop_tol'
> 
> splu is loaded from scipy.sparse.linalg, and indeed it does not have the 
> drop_tol keyword. When I remove the drop_tol keyword, it appears to run as 
> intended, and produce good solutions for my problem. 
> 
> However, I am not sure that is the optimal solution...
> 
> Jamie Pringle
> 
> _______________________________________________
> fipy mailing list
> [email protected]
> http://www.ctcms.nist.gov/fipy
>  [ NIST internal ONLY: https://email.nist.gov/mailman/listinfo/fipy ]


_______________________________________________
fipy mailing list
[email protected]
http://www.ctcms.nist.gov/fipy
  [ NIST internal ONLY: https://email.nist.gov/mailman/listinfo/fipy ]

Reply via email to