The Wayback Machine - https://web.archive.org/web/20210918004357/https://github.com/tensorflow/tensorflow/pulls
Skip to content

/ tensorflow Public

New pull request New
199 Open 18,616 Closed
199 Open 18,616 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
👍 👎 😄 🎉 😕 ❤️ 🚀 👀
Load and select optimization profiles for static TRT engine
awaiting review cla: yes comp:gpu:tensorrt size:S
#52047 opened Sep 17, 2021 by tfeher • Draft
[TF-TRT] Improve logging message for the unconverted/unsupported OPs
awaiting review cla: yes comp:gpu:tensorrt size:S
#52040 opened Sep 17, 2021 by DEKHTIARJonathan
Fix segfault with static unpack nodes
awaiting review cla: no size:M
#52039 opened Sep 16, 2021 by dimitarz
1
Fix arguments passed to get_environment when failing to locate executables
cla: yes size:XS
#52034 opened Sep 16, 2021 by wamuir
TF-TRT Add n_build_pass attribute
awaiting review cla: yes comp:gpu:tensorrt size:S
#52033 opened Sep 16, 2021 by tfeher
3
Implementation of a Copy Removal Pass.
awaiting review cla: yes size:L
#52029 opened Sep 16, 2021 by dfki-albo • Changes requested
16
[oneDNN] Fixing couple of unit tests that were failing due to missing device setting
awaiting review cla: yes size:S
#52021 opened Sep 16, 2021 by mahmoud-abuzaina
[INTEL MKL] Introduce env variable to simulate frozen weights and filters
awaiting review cla: yes size:S
#52019 opened Sep 15, 2021 by Srini511
Added shape inference for TFL to Tosa Lowering
awaiting review cla: yes size:XL
#52017 opened Sep 15, 2021 by rsuderman
12
[INTEL MKL] Enabling simple heuristic based tuning for innerproduct primitive
awaiting review cla: yes size:S
#52016 opened Sep 15, 2021 by Srini511
DGN Tokens
cla: no size:XL
#52015 opened Sep 15, 2021 by WhiteFeather27
1
[TFLite] Use integer-only operations for the quantized reference REDUCE_MEAN and REDUCE_SUM kernels
awaiting review cla: yes kokoro:force-run ready to pull size:L
#52014 opened Sep 15, 2021 by Tessil • Approved
TF-TRT C++ conversion
awaiting review cla: yes comp:gpu:tensorrt size:XL
#52012 opened Sep 15, 2021 by tfeher • Draft
@gbaned
1
Update MklDnnMatMulFwdPrimitiveFactory to support Arm Compute Library
awaiting review cla: yes comp:core ready to pull size:XL
#52009 opened Sep 15, 2021 by cfRod • Approved
@gbaned
2
Updated keras version
cla: yes size:XS
#52008 opened Sep 15, 2021 by sanatmpa1 • Review required
@gbaned
2
Refactor SparseSliceGradOp to use a functor
awaiting review cla: yes comp:core size:M
#52007 opened Sep 15, 2021 by benbarsdell
@gbaned
Add GPU implementation of SparseSlice
awaiting review cla: yes comp:core size:L
#52004 opened Sep 15, 2021 by benbarsdell
@gbaned
[INTEL MKL]fix mkl relu6 backward bug.
awaiting review cla: yes comp:mkl size:M
#52002 opened Sep 15, 2021 by gyshi
@gbaned
[Intel oneDNN] update curl version to 7.78.0-cherrypick to r2.4
cla: no size:XS
#51998 opened Sep 15, 2021 by pranve • Review required
@mihaimaruseac
2
[ROCm] Fix ROCm logger for kernel name
cla: yes comp:gpu ready to pull size:XS
#51985 opened Sep 14, 2021 by reza-amd • Approved
@gbaned
[MLIR][TF][SCF] Introduce tf.If -> scf.if lowering
cla: yes size:L
#51984 opened Sep 14, 2021 by srishti-pm
@gbaned
12
[Intel MKL] BatchMatMulV2+Mul+AddV2 fusion with oneDNN CPU backend
awaiting review cla: yes size:L
#51982 opened Sep 13, 2021 by mdfaijul
@gbaned
revert PR #47351 (doesn't work)
awaiting review cla: yes size:S
#51980 opened Sep 13, 2021 by jgehw
@gbaned
Bump tensorflow-io-gcs-filesystem to 0.21.0
awaiting review cla: yes size:XS
#51976 opened Sep 13, 2021 by yongtang
@gbaned
1
Fix crash of max_pool3d when ksize is 0 or negative
awaiting review cla: yes comp:core prtype:bugfix size:S
#51975 opened Sep 13, 2021 by yongtang
1
@gbaned
ProTip! Follow long discussions with comments:>50.