| commit | 75dc228289e6f422bf46e0afb853e0d1ff16939e | [log] [tgz] |
|---|---|---|
| author | Brian Osman <brianosman@google.com> | Mon Apr 02 17:05:18 2018 -0400 |
| committer | Skia Commit-Bot <skia-commit-bot@chromium.org> | Mon Apr 02 21:16:23 2018 +0000 |
| tree | fe90187b235fbe63a18fe10c4364a972a46842ce | |
| parent | ad0c4f19b56341b1623726b8f67e7a8a45d996ca [diff] |
Don't snap nonlinear TF parameters after regression It's great to have "nice" values, and this actually moves error in both directions (depending on the profile), but there's no principled reason to be doing this. Change-Id: I9952b13cb877df9fc020443cdd8512e641609356 Reviewed-on: https://skia-review.googlesource.com/117920 Commit-Queue: Mike Klein <mtklein@chromium.org> Reviewed-by: Mike Klein <mtklein@chromium.org>