Github user naftaliharris commented on the pull request:
https://github.com/apache/spark/pull/1057#issuecomment-53973576
@mateiz oh yeah, no problem. Thanks again for the fixes!
---
If your project is set up for it, you can reply to this email and have your
reply appear on GitHub as w
Github user naftaliharris closed the pull request at:
https://github.com/apache/spark/pull/1057
---
If your project is set up for it, you can reply to this email and have your
reply appear on GitHub as well. If your project does not have this feature
enabled and wishes so, or if the feature
Github user mateiz commented on the pull request:
https://github.com/apache/spark/pull/1057#issuecomment-53972967
Hey @naftaliharris, might closing this pull request now that this has been
fixed in other PRs?
---
If your project is set up for it, you can reply to this email and have
Github user naftaliharris commented on the pull request:
https://github.com/apache/spark/pull/1057#issuecomment-50520177
Yeah, that fixes the overflow errors! The second probability calculation
has some slight numerical issues, (see in-line comment), but not a huge deal.
---
If your
Github user mateiz commented on the pull request:
https://github.com/apache/spark/pull/1057#issuecomment-50514187
That makes sense. I believe this was fixed just recently in here though:
https://github.com/apache/spark/pull/1493. Does the fix look okay to you?
---
If your project is
Github user naftaliharris commented on the pull request:
https://github.com/apache/spark/pull/1057#issuecomment-50508305
My point actually is that you get OverflowError's if margin is too large,
for example:
```python
>>> from math import exp
>>> margin = -1000
>>> prob
Github user mengxr commented on the pull request:
https://github.com/apache/spark/pull/1057#issuecomment-50433121
I think we should keep it as it is now and add support for setting
thresholds.
---
If your project is set up for it, you can reply to this email and have your
reply appea
Github user mateiz commented on the pull request:
https://github.com/apache/spark/pull/1057#issuecomment-50429778
Actually the Python version of this code seems like it needs to be brought
in line with the threshold API added to Java / Scala. After that, the change
won't work any more
Github user mateiz commented on the pull request:
https://github.com/apache/spark/pull/1057#issuecomment-50429657
@mengxr what do you think about this?
---
If your project is set up for it, you can reply to this email and have your
reply appear on GitHub as well. If your project does
Github user AmplabJenkins commented on the pull request:
https://github.com/apache/spark/pull/1057#issuecomment-45810663
Can one of the admins verify this patch?
---
If your project is set up for it, you can reply to this email and have your
reply appear on GitHub as well. If your pro
GitHub user naftaliharris opened a pull request:
https://github.com/apache/spark/pull/1057
Check if margin > 0, not if prob > 0.5
This avoids an unnecessary computation and also possible math overflow
errors.
You can merge this pull request into a Git repository by running:
$
11 matches
Mail list logo
