Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Merge master into basic-certainty propagation #5

Open
wants to merge 23 commits into
base: basic-certainty-propagation
Choose a base branch
from

Conversation

MelleKoning
Copy link

Improved merge without the clutter.

manyoso and others added 13 commits February 23, 2019 22:49
* misc changes to cudnn backend

- replace all cudaMemcpyAsync used for loading weights with cudaMemcpy as  source (in CPU memory) could be deleted before the async version of the function actually does the copy.
- minor naming/style changes.
- add comment explaining what the policy map layer does and how the layout conversion from CHW to HWC works.

* fix typo in comment

* clang-format

* address review comment

*  cudnn-fp16 SE layer fix

- fall back to non-fused path when sizes are not supported
…eelaChessZero#741)

* move cuda info display earlier

* warn if cudnn older than 7.3.1 is used with rtx cards

* update cudnn to 7.4.2
* Add const to locals

* Update random.cc

Fix linux build errors

* Update syzygy.cc

fixed linux build break.

* Add const to locals

* Update random.cc

Fix linux build errors

* Update syzygy.cc

fixed linux build break.

* more consts

more consts

* Revert "Update random.cc"

This reverts commit 644a2fd.

* Revert "Revert "Update random.cc""

This reverts commit 7f0ba73.

* a few more consts

* Update loader.cc

fixed build break
Do VerboseMoveStats on every info update.
* Quick patch for negative gamma

Still wrong, but at least it's consistently wrong.

* Fold BN to weights and biases

* Fix handling of biases

* Fold batch norm in network_legacy.

No need to do duplicate work in the backends.

* Use biases in OpenCL

* Cleanup and small fixes

Remove unused functions. Fix blas. Disable tensorflow backend.

* clang-format
…tion to FpuValue). (LeelaChessZero#750)

* Add AtRoot versions of FpuStrategy and FpuValue (and convert FpuReduction to FpuValue).

* Move root checks to Params and default to same
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

Successfully merging this pull request may close these issues.