|
From: | Daniel J Sebald |
Subject: | Re: svds test failure |
Date: | Tue, 31 Jul 2012 14:23:44 -0500 |
User-agent: | Mozilla/5.0 (X11; U; Linux x86_64; en-US; rv:1.9.2.24) Gecko/20111108 Fedora/3.1.16-1.fc14 Thunderbird/3.1.16 |
On 07/31/2012 02:00 PM, Daniel J Sebald wrote:
Two things to note above. One is that the result isn't accurate to within 2*eps and the test fails. The second is that the results are *not* deterministic. Therefore, it is random that this test could pass or fail. I suggest bumping the tolerance up to 10*eps or maybe 100*eps... or 1e-10 if you like, same as the rest. It is simply trying to test the results are fairly accurate, right?
And if that change is made, the code feeding in random numbers with 'opts' could probably be removed. My guess is that test is passing fine. But, if you do want to test that 'opts' is in fact working, just pick some numbers to pass in rather than use the random number generator.
Dan
[Prev in Thread] | Current Thread | [Next in Thread] |