math.erfc OverflowError

[geremy condra]

In Python3.2, calling math.erfc with a value in [-27.2, -30) raises
an OverflowError: math range error. This is inconsistent with the
erfc function from scipy (scipy.special.erfc) as well as with the C99
function by the same name, both of which return 2. I suspect that
this is the result of the cutoff for the use of the continuing fraction
approximation of erfc beginning when abs(x) > 30, but I'm not sure.
Is this desired behavior or should I file a bug report?

Geremy Condra

 

[Steven D'Aprano]

In Python3.2, calling math.erfc with a value in [-27.2, -30) raises an
OverflowError: math range error. This is inconsistent with the erfc
function from scipy (scipy.special.erfc) as well as with the C99
function by the same name, both of which return 2. I suspect that this
is the result of the cutoff for the use of the continuing fraction
approximation of erfc beginning when abs(x) > 30, but I'm not sure. Is
this desired behavior or should I file a bug report?

Geremy, I know that as a general rule, whenever a person finds what seems
to be a bug in Python (or the standard library) they should suspect their
own mistake rather than a bug, but I think this is taking caution to
extremes. The domain of erfc is the set of all real numbers, and the
result is bounded by [0, 2]. So asking if OverflowError is desired
behaviour is a bit like the fellow who bought a new car and then asked if
it's supposed to catch fire when starting the engine...

 

[geremy condra]

In Python3.2, calling math.erfc with a value in [-27.2, -30) raises an
OverflowError: math range error. This is inconsistent with the erfc
function from scipy (scipy.special.erfc) as well as with the C99
function by the same name, both of which return 2. I suspect that this
is the result of the cutoff for the use of the continuing fraction
approximation of erfc beginning when abs(x) > 30, but I'm not sure. Is
this desired behavior or should I file a bug report?

Geremy, I know that as a general rule, whenever a person finds what seems
to be a bug in Python (or the standard library) they should suspect their
own mistake rather than a bug, but I think this is taking caution to
extremes. The domain of erfc is the set of all real numbers, and the
result is bounded by [0, 2]. So asking if OverflowError is desired
behaviour is a bit like the fellow who bought a new car and then asked if
it's supposed to catch fire when starting the engine...

Well, that was more or less my feeling on it, but as a general rule
my opinion on math in python doesn't seem to be shared, so I
thought I'd ask.

Geremy Condra