Home » Oreps » TensorFlow » All Releases » pending_release Release
tensorflow/tensorflow
An Open Source Machine Learning Framework for Everyone
Watch
6
Commit Activity
Directory Browser
Velocity History
ReleasesLatest Pending
Basic Details Issues Addressed Top Contributors Directory Browser
TensorFlow Latest Pending

TensorFlow: Latest Pending Release

Release date:
Pending release in a future version
Previous version:
v2.18.0-rc0 (released September 26, 2024)
Magnitude:
11,536 Diff Delta
Contributors:
78 total committers
Data confidence:
---
Commits:
293 impacting commits

293 Commits in this Release

Ordered by the degree to which they evolved the repo in this version.

Add Dispatch API to LiteRT runtime with a Qualcomm implementation
Authored October 8, 2024
Add TensorBuffer support for the LiteRT runtime
Authored October 5, 2024
[HLO Componentization] Create hlo/parser sub-component (Phase II).
Authored October 7, 2024
PR #16585: Add support for float8_e4m3 and float8_e3m4 types
Authored October 2, 2024
Reverts cc97551632b6f9c21923a6cfa34371d676c239e1
Authored September 27, 2024
CC wrapper for lrt tensors and basic lrt -> qnn tensor mapping
Authored October 4, 2024
Simplify management of status codes
Authored October 1, 2024
[XLA:GPU] Support 0D Tensors in the Generic Triton Emitter.
Authored October 7, 2024
Remove more unnecessary forward declarations from stream_executor
Authored October 1, 2024
[XLA:GPU] Remove the use of GpuTimer::ReturnRandomDurationsForTesting() in determinism_test.cc.
Authored October 7, 2024
[HLO Componentization] Create hlo/builder sub-component (Phase II).
Authored October 1, 2024
copy the legalize tf passes under lite/stablehlo
Authored September 30, 2024
[XLA:TPU] Reenable fragmentation aware loop optimizer. Only restore the original heap state for early forced prefetches that were not allocated. This was a bug in the original change to enable fragmentation aware loop optimizer. It was also a bug in the earlier loop optimizer which was causing some prefetches to be pushed later instead of earlier.
Authored October 4, 2024
CC wrapper for lrt op and basic lrt -> qnn op mapping
Authored October 4, 2024
Make rocm files use tsl DsoLoader functions instead of the stream_executor wrappers.
Authored October 2, 2024
Remove cuda_only_cc_library
Authored September 27, 2024
Move `tsl/protobuf/error_codes.proto` to `xla/tsl/protobuf`
Authored October 4, 2024
Pass down device_type in tf2xla api fallback path
Authored October 7, 2024
Calculate different flops for different nvidia gpu.
Authored September 27, 2024
Add explicit includes to fix Kokoro compile issues.
Authored September 30, 2024
Support multiple floating point types in client library test base
Authored September 30, 2024
Reverts 1b240fae66605d39e14d4459450664c0ac20e97f
Authored September 30, 2024
[XLA:GPU] Generalize the Reduce-Precision in MHLO to also work on Tensors and use it in a Triton emitter.
Authored October 2, 2024
Finish prototype implementation for qnn compiler plugin. Currently only knows how to handle simple cases where there winds up being one partition with one float 2x2 mul op.
Authored October 4, 2024
[XLA:GPU] Unify DimVar, RangeVar and RTVar.
Authored October 2, 2024
[HLO Componentization] Create hlo/parser sub-component (Phase I).
Authored October 1, 2024
[XLA:GPU] Introduce the `FusionBlockLevelRewriter` pass.
Authored October 2, 2024
Move CreateDeviceDescription to CudaExecutor and RocmExecutor and add tests
Authored September 30, 2024
Pull generic dynamic loading utils into core api. This will need to be reused for loading plugin .so from driver.
Authored October 5, 2024
[XLA:GPU] Use metadata to print and parse indexing maps.
Authored September 30, 2024
#sdy import `CallOp`s with `backend_configs` to `NamedComputationOp`s.
Authored October 7, 2024
[XLA:GPU] propagate the algorithm flag of dot op to cublasGemm custom call.
Authored September 27, 2024
[IFRT] Add Client::GetAllDevices()
Authored October 5, 2024
Associates names with individual input sharding combinations (rather than strategies).
Authored October 2, 2024
re-enabling some tests that had previously been failing
Authored October 7, 2024
Fork `xla::ExecuteOptions` into `xla::ifrt::ExecuteOptions`
Authored September 30, 2024
Delete xla_client.execute_with_python_values.
Authored October 2, 2024
#sdy remove size one axes from all shardings and meshes in the module.
Authored September 27, 2024
#sdy export NamedComputationOps with backend_configs to CallOps.
Authored October 7, 2024
Remove unneeded dso_loader.h include from ROCM files.
Authored October 1, 2024

Browse Other Releases

Latest Pending Unreleased 😎
v2.18.0-rc0 Released September 26, 2024
0 Ξ”
v2.17.0 Released July 9, 2024
3 Ξ”
v2.17.0-rc1 Released June 28, 2024
0 Ξ”
v2.17.0-rc0 Released June 12, 2024
0 Ξ”
v2.16.2 Released June 25, 2024
0 Ξ”
v2.16.1 Released March 6, 2024
13,257 Ξ”
v2.16.0-rc0 Released March 8, 2024
409 Ξ”
v2.15.1 Released March 8, 2024
18 Ξ”
v2.15.0 Released November 10, 2023
9 Ξ”
v2.15.0-rc1 Released November 2, 2023
7,949 Ξ”

Top Contributors in Latest Pending

tensorflower-gardener
beckerhe
sdasgup3
LukeBoyer
apivovarov
klucke
dimitar-asenov
pifon2a
bartchr808
loislo

Directory Browser for Latest Pending

We haven't yet finished calculating and confirming the files and directories changed in this release. Please check back soon.

GitClear uses cookies to ensure you get the best experience on our website. Learn more