Skip to content

Pull requests: microsoft/onnxruntime

New pull request New
Clear current search query, filters, and sorts
260 Open 16,613 Closed
260 Open 16,613 Closed
Author
Filter by author
Loading
Label
Filter by label
Loading
Use alt + click/return to exclude labels
or ⇧ + click/return for logical OR
Projects
Filter by project
Loading
Milestones
Filter by milestone
Loading
Reviews
Filter by reviews
No reviews Review required Approved review Changes requested
Assignee
Filter by who’s assigned
Assigned to nobody Loading
Sort
Sort by
Newest Oldest Most commented Least commented Recently updated Least recently updated Best match
Most reactions
👍 👎 😄 🎉 😕 ❤️ 🚀 👀

Pull requests list

Consider initializer value of NodeArg* when comparing equivalence classes
#5591 opened Oct 26, 2020 by sriduth • Draft
@MaximKalininMS
34
[WIP] Relax If node and subgraph shape inference check
#5636 opened Oct 29, 2020 by BowenBao • Draft
3
updates to onnx test filters and validation for op registrations
#5866 opened Nov 19, 2020 by askhade Loading…
4
Memory pattern enhancements - Part 2
#5930 opened Nov 25, 2020 by codemzs • Draft
5
Add equalization algorithm and example usage quantization issues related to quantization
#6508 opened Jan 30, 2021 by deepology Loading…
6
[Work in Progress] Option to enable partition of graph after AD training issues related to ONNX Runtime training; typically submitted using template
#6958 opened Mar 9, 2021 by jupvfranco • Draft
2
Enable CUDA provider option configuration for C#
#7315 opened Apr 12, 2021 by chilo-ms Loading…
33
Fix RoiAlign CPU EP issues
#7354 opened Apr 16, 2021 by fdwr Loading…
@linnealovespie
8
Fuse SkipLayerNorm + MatMul/Bias/Matmul subgraph for BERT models.
#8354 opened Jul 12, 2021 by nkreeger • Draft
2
AMDMigraphx check ret value of HIP function call to avoid build error
#8356 opened Jul 12, 2021 by scxiao Loading…
@weixingzhang
8
Add quantized operator and tooling for BiasGelu Op.
#8666 opened Aug 9, 2021 by nkreeger • Draft
2
WIP: Quantized SkipLayerNormalization
#8691 opened Aug 11, 2021 by nkreeger • Draft
2
Improve GenerateId to reduce hash collision
#8837 opened Aug 24, 2021 by stevenlix Loading…
28
Implement better past state parsing semantics for language models
#9075 opened Sep 16, 2021 by hariharans29 Loading…
3
Add readme for docs folder redirecting users to actual documentation
#9100 opened Sep 17, 2021 by prasanthpul Loading…
2
Benckmark tool for e2e inference on GPT2 LM
#9218 opened Sep 29, 2021 by viboga Loading…
6
Use in-place object creation semantics instead of plain insert
#9320 opened Oct 10, 2021 by mosdav Loading…
2
[WIP] External Thread Pool
#9341 opened Oct 12, 2021 by RandySheriffH Loading…
2
Clean up
#9705 opened Nov 9, 2021 by lifa08 Loading…
5
[WIP] ORTTrainer's AllReduce graph builder supports models with different float types.
#9715 opened Nov 10, 2021 by ashari4 • Draft
1
adding compile flag for ort_aten
#9718 opened Nov 10, 2021 by ajindal1 Loading…
2
Use QuantizeLinear rather than dynamic quantization in quantize_qat() when weight is transposed before MatMul
#9869 opened Nov 27, 2021 by kkang-groq Loading…
5
[WIP] Include pthreadpool as a submodule for perf test.
#9953 opened Dec 7, 2021 by RandySheriffH Loading…
2
Change hard-coded DLL check to using ctypes.WinDLL
#9971 opened Dec 8, 2021 by FelixS90 Loading…
@ytaous
6
[NO REVIEW REQUIRED] POC: Integrate DeepSpeed CUDA Transformer kernels into ORT
#10024 opened Dec 13, 2021 by hariharans29 Loading…
1
ProTip! Adding no:label will show everything without a label.