[Numpy-discussion] BigInteger equivalent in numpy
Sun Jun 7 05:05:36 CDT 2009
(Please do not send twice your message to numpy/scipy ML, thank you)
> int64 and float seem to work for the stderr calculation. Now, the
> calculation of the p-value causes an underflow.
> File "C:\Python26\lib\site-packages\scipy\stats\distributions.py", line 2829,in _cdf
> return special.stdtr(df, x)
> FloatingPointError: underflow encountered in stdtr
> It seems that the error occurs in scipy.special.stdtr(df, x) if df =
> array() and x = array([61.88071696]).
The error seems to happen in the cephes library (the library we use for
many functions in scipy.special, including this one). I don't know
exactly what triggers it (your df is quite high, though, and the cdf is
already quite past the 1-eps at your point).
More information about the Numpy-discussion