You need to install bigfloat module first. See here for instructions <http://bigfloat.readthedocs.org/en/latest/#installation>.
Once it is done, use bigfloat.exp instead of numpy.exp. In [31]: v = numpy.linspace(0, 100, 10) In [32]: from bigfloat import precision, exp In [33]: with precision(100000): ....: j = [exp(i/1.3806488e-23) for i in v] Cheers, Anand Reddy Pandikunta (@ChillarAnand) www.avilpage.com On Wed, Mar 30, 2016 at 12:46 PM, Santosh Chiniwar <santosh.ch...@gmail.com> wrote: > Thanks for kind reply. I am new to this python coding. Could you please > tell me how use bigfloat to fix this overflow problem > > I am getting overflow for following values > > v = numpy.linspace(0, 100, 10) > j= numpy.exp(v/1.3806488e-23) > > > > I have tried to use In [30]: from bigfloat import * > > --------------------------------------------------------------------------- > > ImportError Traceback (most recent call last) > > <ipython-input-30-45ecbac7174b> in <module>() > > ----> 1 from bigfloat import * > > > ImportError: No module named 'bigfloat' > > > > > > > *Santosh P. Chiniwar* > > *----------------------------------Research Scholar, Thin Film Measurement > Lab * > *Department of Electronics Engineering* > *Chang Gung University, Taiwan (ROC)* > *259 Wenhua 1st Rd., Guishan Dist., Taoyuan, * > *33302, **Taiwan (ROC), **Tel.: +886-3211-8800* > *-ext.535 **Mob: +886-905667324* > > The first principle is that you must not fool yourself-and you are the > easiest person to fool..... Richard Feynman > > > > On 29 March 2016 at 21:34, Gora Mohanty <g...@mimirtech.com> wrote: > > > On 29 March 2016 at 17:14, Santosh Chiniwar <santosh.ch...@gmail.com> > > wrote: > > > Dear BangPypers, > > > Please find my code and I get Overflow warning. I am unable to > continue > > > further. Any support or help is appreciated. > > > > Somewhere you are getting too large a number for the range of type in > > the array that you are using: probably because you are dividing by the > > very small Boltzmann constant. > > > > You either need to use a type with a bigger range, say bigfloat for > > arbitrary precision floating-point arithmetic, or if the errors only > > occur for values that can be ignored, see > > http://docs.scipy.org/doc/numpy/reference/generated/numpy.seterr.html > > for how to ignore errors. > > > > Incidentally, all this information is available from an easy Google > search. > > > > Regards, > > Gora > > _______________________________________________ > > BangPypers mailing list > > BangPypers@python.org > > https://mail.python.org/mailman/listinfo/bangpypers > > > _______________________________________________ > BangPypers mailing list > BangPypers@python.org > https://mail.python.org/mailman/listinfo/bangpypers > _______________________________________________ BangPypers mailing list BangPypers@python.org https://mail.python.org/mailman/listinfo/bangpypers
