<html><body>
Actually, the behavior right now isn’t that of `default` but that of `initializer` or `start`.<div><br></div><div>This was discussed further down in the PR but to reiterate: `np.sum([10], initializer=5)` becomes `15`.</div><div><br></div><div>Also, `np.min([5], initializer=0)` becomes `0`, so it isn’t really the default value, it’s the initial value among which the reduction is performed.</div><div><br></div><div>This was the reason to call it initializer in the first place. I like `initial` and `initial_value` as well, and `start` also makes sense but isn’t descriptive enough.</div><div><br></div><div>
Hameer<br>
<div class="">Sent from <a href="https://www.helloastro.com">Astro</a> for Mac</div><br><blockquote class="hm_quoted_text" style="padding-left:8px;margin:0;border-left:1px solid rgb(185,185,185);color:rgb(100,100,100)">
<div>On Mar 26, 2018 at 12:06, Sebastian Berg <<a href="mailto:sebastian@sipsolutions.net">sebastian@sipsolutions.net</a>> wrote:</div><p>
<br><span style="white-space:pre-wrap" class="hm_plaintext">Initializer or this sounds fine to me. As an other data point which I<br>think has been mentioned before, `sum` uses start and min/max use<br>default. `start` does not work, unless we also change the code to<br>always use the identity if given (currently that is not the case), in<br>which case it might be nice. However, "start" seems a bit like solving<br>a different issue in any case.<br><br>Anyway, mostly noise. I really like adding this, the only thing worth<br>discussing a bit is the name :).<br><br>- Sebastian</span></p><div><br><br>On Mon, 2018-03-26 at 05:57 -0400, Hameer Abbasi wrote:<br><blockquote type="cite">It calls it `initializer` - See <a href="https://docs.python.org/3.5/library/f">https://docs.python.org/3.5/library/f</a><br>unctools.html#functools.reduce<br><br>Sent from Astro for Mac<br><br><blockquote type="cite">On Mar 26, 2018 at 09:54, Eric Wieser <<a href="mailto:wieser.eric%2Bnumpy@gmail.com">wieser.eric+numpy@gmail.com</a>><br>wrote:<br><br>It turns out I mispoke - functools.reduce calls the argument<br>`initial`<br><br>On Mon, 26 Mar 2018 at 00:17 Stephan Hoyer <<a href="mailto:shoyer@gmail.com">shoyer@gmail.com</a>><br>wrote:<br><blockquote type="cite">This looks like a very logical addition to the reduce interface.<br>It has my support!<br><br>I would have preferred the more descriptive name "initial_value",<br>but consistency with functools.reduce makes a compelling case for<br>"initializer".<br><br>On Sun, Mar 25, 2018 at 1:15 PM Eric Wieser <wieser.eric+numpy@gm<br><a href="http://ail.com">ail.com</a>> wrote:<br><blockquote type="cite">To reiterate my comments in the issue - I'm in favor of this.<br><br>It seems seem especially valuable for identity-less functions<br>(`min`, `max`, `lcm`), and the argument name is consistent with<br>`functools.reduce`. too.<br><br>The only argument I can see against merging this would be<br>`kwarg`-creep of `reduce`, and I think this has enough use<br>cases to justify that.<br><br>I'd like to merge in a few days, if no one else has any<br>opinions.<br><br>Eric<br><br>On Fri, 16 Mar 2018 at 10:13 Hameer Abbasi <einstein.edison@gma<br><a href="http://il.com">il.com</a>> wrote:<br><blockquote type="cite">Hello, everyone. I’ve submitted a PR to add a initializer<br>kwarg to ufunc.reduce. This is useful in a few cases, e.g.,<br>it allows one to supply a “default†value for identity-less<br>ufunc reductions, and specify an initial value for reductions<br>such as sum (other than zero.)<br><br>Please feel free to review or leave feedback, (although I<br>think Eric and Marten have picked it apart pretty well).<br><br><a href="https://github.com/numpy/numpy/pull/10635">https://github.com/numpy/numpy/pull/10635</a><br><br>Thanks,<br><br>Hameer<br>Sent from Astro for Mac<br><br>_______________________________________________<br>NumPy-Discussion mailing list<br><a href="mailto:NumPy-Discussion@python.org">NumPy-Discussion@python.org</a><br><a href="https://mail.python.org/mailman/listinfo/numpy-discussion">https://mail.python.org/mailman/listinfo/numpy-discussion</a><br></blockquote><br>_______________________________________________<br>NumPy-Discussion mailing list<br><a href="mailto:NumPy-Discussion@python.org">NumPy-Discussion@python.org</a><br><a href="https://mail.python.org/mailman/listinfo/numpy-discussion">https://mail.python.org/mailman/listinfo/numpy-discussion</a><br></blockquote><br>_______________________________________________<br>NumPy-Discussion mailing list<br><a href="mailto:NumPy-Discussion@python.org">NumPy-Discussion@python.org</a><br><a href="https://mail.python.org/mailman/listinfo/numpy-discussion">https://mail.python.org/mailman/listinfo/numpy-discussion</a><br></blockquote><br>_______________________________________________<br>NumPy-Discussion mailing list<br><a href="mailto:NumPy-Discussion@python.org">NumPy-Discussion@python.org</a><br><a href="https://mail.python.org/mailman/listinfo/numpy-discussion">https://mail.python.org/mailman/listinfo/numpy-discussion</a><br></blockquote><br>_______________________________________________<br>NumPy-Discussion mailing list<br><a href="mailto:NumPy-Discussion@python.org">NumPy-Discussion@python.org</a><br><a href="https://mail.python.org/mailman/listinfo/numpy-discussion">https://mail.python.org/mailman/listinfo/numpy-discussion</a><br></blockquote></div><br><span style="white-space:pre-wrap" class="hm_plaintext"><div>_______________________________________________<br>NumPy-Discussion mailing list<br><a href="mailto:NumPy-Discussion@python.org">NumPy-Discussion@python.org</a><br><a href="https://mail.python.org/mailman/listinfo/numpy-discussion">https://mail.python.org/mailman/listinfo/numpy-discussion</a><br></div></span>
</blockquote>
</div></body></html>