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.

Sign up for GitHub

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

Cherry for release 1.12.0 final #12218

Merged
merged 15 commits into from
Jul 19, 2022
Merged

Cherry for release 1.12.0 final #12218

Changes from 1 commit
Commits
Show all changes
15 commits
Select commit Hold shift + click to select a range
7d82a5a
support optimizer opt for deepspeed 0.5.9
zhijxu-MS Apr 27, 2022
f181593
resolve comments
Jun 22, 2022
f61cee1
resolve comments
zhijxu-MS Jun 23, 2022
12c9628
FP16_Optimizer Support for more Deepspeed Versions (#12046)
centwang Jun 30, 2022
bbbbaec
Fix unused function warning for decodeMIDR(). (#12069)
edgchen1 Jul 6, 2022
4b223e4
pin protobuf version to be compatible with onnx (#12132)
askhade Jul 8, 2022
158ef0f
RoiAlign CPU EP add warning for max mode with samples != 1 (#12136)
fdwr Jul 12, 2022
9db1f6b
include coreml_provider_factory.h in macos build instead of coreml_ex…
carsonswope Jul 12, 2022
775adb2
List 3.10 as supported python version and remove 3.6 (#12141)
RandySheriffH Jul 12, 2022
4de96ad
Use updated symbolic_helper.check_training_mode (#11900)
jingyanwangms Jul 13, 2022
e9b71a8
Fix GH issue 12151 by using inverse perms for updating DQ axis attrib…
skottmckay Jul 13, 2022
f91cee3
fixing positions for beam search gpt2 (#12156)
viboga Jul 14, 2022
0ae704f
remove wrong placed libs (#12201)
leqiao-1 Jul 18, 2022
3f6a276
Add file mapping for windows platform. (#12183)
caoting-dotcom Jul 18, 2022
45b8e11
Fix bug where onnxruntime_USE_NCCL flag would default to ON (#12195)
seanmurr1 Jul 18, 2022
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
Prev Previous commit
Fix bug where onnxruntime_USE_NCCL flag would default to ON (#12195) 
Fix bug where onnxruntime_USE_NCCL flag would default to ON, causing ORT to not build properly. New functionality: flag is ON when training is enabled and NCCL is not disabled. Flag is OFF otherwise
  • Loading branch information
@seanmurr1 @RandyShuai
seanmurr1 authored and RandyShuai committed Jul 18, 2022
commit 45b8e11f1af3d0636d5e57753ca35a744ccd98df
2 changes: 1 addition & 1 deletion tools/ci_build/build.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -875,7 +875,7 @@ def generate_build_tree(
"-Donnxruntime_ENABLE_TRAINING_TORCH_INTEROP=" + ("ON" if args.enable_training_torch_interop else "OFF"),
# Enable advanced computations such as AVX for some traininig related ops.
"-Donnxruntime_ENABLE_CPU_FP16_OPS=" + ("ON" if args.enable_training else "OFF"),
"-Donnxruntime_USE_NCCL=" + ("OFF" if args.disable_nccl else "ON"),
"-Donnxruntime_USE_NCCL=" + ("ON" if args.enable_training and not args.disable_nccl else "OFF"),
"-Donnxruntime_BUILD_BENCHMARKS=" + ("ON" if args.build_micro_benchmarks else "OFF"),
"-Donnxruntime_USE_ROCM=" + ("ON" if args.use_rocm else "OFF"),
"-DOnnxruntime_GCOV_COVERAGE=" + ("ON" if args.code_coverage else "OFF"),
Expand Down