[Numpy-discussion] Catching and dealing with floating point errors

Skipper Seabold jsseabold@gmail....
Mon Nov 8 14:52:18 CST 2010


On Mon, Nov 8, 2010 at 3:42 PM, Bruce Southey <bsouthey@gmail.com> wrote:
> On 11/08/2010 02:17 PM, Skipper Seabold wrote:
>> On Mon, Nov 8, 2010 at 3:14 PM, Skipper Seabold<jsseabold@gmail.com>  wrote:
>>> I am doing some optimizations on random samples.  In a small number of
>>> cases, the objective is not well-defined for a given sample (it's not
>>> possible to tell beforehand and hopefully won't happen much in
>>> practice).  What is the most numpythonic way to handle this?  It
>>> doesn't look like I can use np.seterrcall in this case (without
>>> ignoring its actual intent).  Here's a toy example of the method I
>>> have come up with.
>>>
>>> import numpy as np
>>>
>>> def reset_seterr(d):
>>>     """
>>>     Helper function to reset FP error-handling to user's original settings
>>>     """
>>>     for action in [i+'='+"'"+d[i]+"'" for i in d]:
>>>         exec(action)
>>>     np.seterr(over=over, divide=divide, invalid=invalid, under=under)
>>>
>> It just occurred to me that this is unsafe.  Better options for
>> resetting seterr?
>>
>>> def log_random_sample(X):
>>>     """
>>>     Toy example to catch a FP error, re-sample, and return objective
>>>     """
>>>     d = np.seterr() # get original values to reset
>>>     np.seterr('raise') # set to raise on fp error in order to catch
>>>     try:
>>>         ret = np.log(X)
>>>         reset_seterr(d)
>>>         return ret
>>>     except:
>>>         lb,ub = -1,1  # includes bad domain to test recursion
>>>         X = np.random.uniform(lb,ub)
>>>         reset_seterr(d)
>>>         return log_random_sample(X)
>>>
>>> lb,ub = 0,0
>>> orig_setting = np.seterr()
>>> X = np.random.uniform(lb,ub)
>>> log_random_sample(X)
>>> assert(orig_setting == np.seterr())
>>>
>>> This seems to work, but I'm not sure it's as transparent as it could
>>> be.  If it is, then maybe it will be useful to others.
>>>
>>> Skipper
>>>
>> _______________________________________________
>> NumPy-Discussion mailing list
>> NumPy-Discussion@scipy.org
>> http://mail.scipy.org/mailman/listinfo/numpy-discussion
> What do you mean by 'floating point error'?
> For example, log of zero is not what I would consider a 'floating point
> error'.
>
> In this case, if you are after a log distribution, then you should be
> ensuring that the lower bound to the np.random.uniform() is always
> greater than zero. That is, if lb <= zero then you *know* you have a
> problem at the very start.
>
>

Just a toy example to get a similar error.  I call x <= 0 on purpose here.

> Bruce
>
>
> _______________________________________________
> NumPy-Discussion mailing list
> NumPy-Discussion@scipy.org
> http://mail.scipy.org/mailman/listinfo/numpy-discussion
>


More information about the NumPy-Discussion mailing list