Skip to content

/ onnxruntime

New pull request New
72 Open 4,432 Closed
72 Open 4,432 Closed
Author
Filter by author
Label
Filter by label
Use alt + click/return to exclude labels.
Projects
Filter by project
Milestones
Filter by milestone
Reviews
Filter by reviews
No reviews Review required Approved review Changes requested
Assignee
Filter by who’s assigned
Sort
Sort by
Newest Oldest Most commented Least commented Recently updated Least recently updated
Most reactions
👍 👎 😄 🎉 😕 ❤️ 🚀 👀
Train BERT Using BFloat16 on A100
component:training-core
#6090 opened Dec 9, 2020 by iK1D • Review required
1
Reduction kernel optimization
component:operator component:training-core
#6088 opened Dec 9, 2020 by edgchen1 • Review required
1
Update AMD transpose to match CUDA transpose.
#6086 opened Dec 8, 2020 by jessebenson • Approved
1
Enable running the mnist_training sample without cuda
#6085 opened Dec 8, 2020 by georgen117 • Review required
Second round of cherry-pick
#6083 opened Dec 8, 2020 by duli2012 • Approved
adding fp16 support for topk cuda kernel
#6082 opened Dec 8, 2020 by duli2012 • Review required
2
Add skip test reason for onnx model zoo models and tier 2 models
#6081 opened Dec 8, 2020 by ryanlai2 • Review required
If building inbox, hook up winrt_activation_handler for WinML Tests
#6074 opened Dec 8, 2020 by ryanlai2 • Review required
[WIP] add support for pipeline parallelism for non_differentiable edge
#6071 opened Dec 8, 2020 by xzhu1900
(WIP) Liqun/speech model loop to scan
#6070 opened Dec 7, 2020 by liqunfu • Review required
Enable varied input type for custom op
#6066 opened Dec 7, 2020 by RandySheriffH • Review required
7
Dnnl training
#6045 opened Dec 4, 2020 by georgen117 • Review required
6
Enable ONNX backend test of SequenceProto input/output
#6043 opened Dec 4, 2020 by jcwchen • Review required
3
Revert Windows GPU python 3.8 and python 3.9 changes in the release branch
release:1.6 triage:approved
#6035 opened Dec 4, 2020 by snnn • Review required
6
Support double for operators Where, LpNormalisation (CPU)
#6034 opened Dec 4, 2020 by xadupre • Review required
Support double for operators Log, Reciprocal, Sum (CPU)
#6032 opened Dec 4, 2020 by xadupre • Review required
Upgrade packages to avoid docker image security vulnerability
#6029 opened Dec 4, 2020 by jingyanwangms • Draft
1
Fix multi-tensor LAMB reduction to be deterministic
#6028 opened Dec 3, 2020 by suffiank • Review required
11
Liqun/move nightly pl to linux multi gpu v100
component:build component:training-frontend
#6024 opened Dec 3, 2020 by liqunfu • Review required
5
Expose model_with_training_graph_path to Python ORTTrainer
#6001 opened Dec 2, 2020 by ryotatomioka • Changes requested
8
Add hipFFT include directory (transitional step) before ROCm.
#5992 opened Dec 1, 2020 by memmett • Review required
2
Update orttraining-linux-gpu-e2e-test-nightly-pipeline.yml for Azure Pipelines
#5986 opened Dec 1, 2020 by liqunfu • Review required
Upgrade build.py to assert for python 3.6+
#5981 opened Dec 1, 2020 by WilliamTambellini • Review required
1
Add ReduceL2Grad and ClipGrad
component:training-core
#5970 opened Nov 30, 2020 by iK1D • Review required
7
Fix DEBUG_NODE_INPUTS_OUTPUTS test by putting it in a separate process, clean up unused test_main.cc files.
#5949 opened Nov 26, 2020 by edgchen1 • Review required
1
ProTip! What’s not been updated in a month: updated:<2020-11-09.
You can’t perform that action at this time.