Skip to content

Pull requests: google/XNNPACK

New pull request New
56 Open 3,160 Closed
56 Open 3,160 Closed
Author
Filter by author
Label
Filter by label
Use alt + click/return to exclude labels
or ⇧ + click/return for logical OR
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
👍 👎 😄 🎉 😕 ❤️ 🚀 👀

Pull requests list

Move post-operation structs into separate file and lib
#3340 opened Aug 15, 2022 by copybara-service bot Loading…
Add support for fused operators to some JIT generators and also microkernel tests (not generated)
#3338 opened Aug 15, 2022 by copybara-service bot Loading…
BF16 GEMM microkernels
#3326 opened Aug 8, 2022 by copybara-service bot Loading…
WIP pipe fused params through subgraph and operators
#3269 opened Jul 22, 2022 by copybara-service bot Loading…
Variable size transpose ukernels are treated as 1d ukernels. They are inherently 2d ukernels, element size contains one dimensions. When the input or output dimension is strided, then adding element size to input pointer in the outer loop will no longer point to the next element.
#3250 opened Jul 18, 2022 by copybara-service bot Loading…
SpaceToDepth xnnpack delegate
#3249 opened Jul 18, 2022 by copybara-service bot Loading…
Add avx & avx2 transpose microkernel generator
#3247 opened Jul 15, 2022 by copybara-service bot Loading…
Add Space to depth to XNNPACK
#3190 opened Jun 22, 2022 by copybara-service bot Loading…
FP16 MobileNet V1 sparse model
#3187 opened Jun 21, 2022 by copybara-service bot Loading…
Dispatch x24 transpose
#3180 opened Jun 21, 2022 by copybara-service bot Loading…
Depth to space nhwc uses transpose.
#3163 opened Jun 16, 2022 by copybara-service bot Loading…
Transpose normalization now handles input and output strides.
#3162 opened Jun 16, 2022 by copybara-service bot Loading…
Variable size transpose ukernels no longer assume that inner most input and output strides are equal to the element size.
#3161 opened Jun 16, 2022 by copybara-service bot Loading…
Add vaddc AArch64 x4, x12, x16 assembly
#3104 opened Jun 7, 2022 by copybara-service bot Loading…
Depth to space nchw2nhwc uses transpose
#3088 opened Jun 2, 2022 by copybara-service bot Loading…
Support TRANSPOSE in XNNPack delegate
#3010 opened Apr 29, 2022 by copybara-service bot Loading…
Internal config change
#2810 opened Mar 29, 2022 by copybara-service bot Loading…
Remove incorrect check on dimensions for split node's split dimension input tensor
#2759 opened Mar 16, 2022 by copybara-service bot Loading…
Internal config change.
#2718 opened Mar 9, 2022 by copybara-service bot Loading…
Rename quantized concatenation tests to include quantization in test name
#2709 opened Mar 8, 2022 by copybara-service bot Loading…
[XNNPACK] Support SPLIT operator with 2 outputs
#2696 opened Mar 3, 2022 by copybara-service bot Loading…
Add XNNPACK delegate per operator profiling to TFLite
#2694 opened Mar 3, 2022 by copybara-service bot Loading…
Internal config change
#2588 opened Feb 10, 2022 by copybara-service bot Loading…
Make SSE2 microkernels consistent with neon zip microkernels.
#2457 opened Jan 28, 2022 by copybara-service bot Loading…
Split up aarch32 instruction encoding tests to avoid stack frame size errors in certain build configs
#2394 opened Jan 21, 2022 by copybara-service bot Loading…
ProTip! What’s not been updated in a month: updated:<2022-07-16.