OSSForks/tensorflow
1
0
Fork 0
mirror of https://github.com/zebrajr/tensorflow.git synced 2026-01-15 12:15:41 +00:00
Code Issues Actions 5 Packages Projects Releases Wiki Activity

Commit Graph

  • 4272c47e39 Automated Code Change A. Unique TensorFlower 2025-12-23 22:28:14 -08:00
  • 5e56a937de Automated Code Change A. Unique TensorFlower 2025-12-23 20:38:31 -08:00
  • ac8d8c396d Reverts 93ae7c23ea A. Unique TensorFlower 2025-12-23 20:15:41 -08:00
  • 7d0d534756 Automated Code Change A. Unique TensorFlower 2025-12-23 20:01:30 -08:00
  • 53988344aa Automated Code Change A. Unique TensorFlower 2025-12-23 19:29:49 -08:00
  • 93ae7c23ea [ReplicaGroupV3][Refactor][6/n] Update rest of spmd/ dir to use CollectiveDeviceListBase in place of vector<vector<int>> and reduce cognitive complexity in GetDefaultCollectiveOpsCreator. Bill Varcho 2025-12-23 18:12:34 -08:00
  • 0cebad6654 Add Shape to TriangularSolveThunk buffer_uses Maxim Ermilov 2025-12-23 15:52:49 -08:00
  • 989c58d0c7 Reverts 69c656e60b A. Unique TensorFlower 2025-12-23 15:06:05 -08:00
  • 13f6a371ab Add Shape to ConvolutionReorderThunk buffer_uses Maxim Ermilov 2025-12-23 15:02:25 -08:00
  • c16ae6c919 Add proto serialization for SendThunk Maxim Ermilov 2025-12-23 14:23:10 -08:00
  • a554adf928 add missing dependencies Maxim Ermilov 2025-12-23 13:56:20 -08:00
  • b39d0a9077 disable //third_party/tensorflow/compiler/xla/service/gpu:determinism_test_h100 for now Maxim Ermilov 2025-12-23 13:07:55 -08:00
  • 69c656e60b Add TPU performance counters to XSpace. Praneeth Mandala 2025-12-23 12:30:28 -08:00
  • 1b61eec06b This is an internal change A. Unique TensorFlower 2025-12-23 12:25:40 -08:00
  • db900e578c [XLA:CPU] Refactor grouped convolution handling in YNN emitter. Volodymyr Kysenko 2025-12-23 12:17:33 -08:00
  • 5bd50f92b6 Remove redundant replica group copy in CollectiveThunk. Henning Becker 2025-12-23 12:03:23 -08:00
  • 76210ebaa1 [ASAN] Initialize Thunk::PrepareParams with all members in tests. Alex Pivovarov 2025-12-23 11:10:05 -08:00
  • 4e413608be switch from deprecated TF_CHECK_OK Maxim Ermilov 2025-12-23 11:08:22 -08:00
  • b689237cba Adds a new test that MemorySpacePropagation currently fails on. Mehrdad Khani 2025-12-23 10:57:16 -08:00
  • 2cc7d2ac89 Reverts 05c8cf7c13 Jing Pu 2025-12-23 09:31:53 -08:00
  • ee2c4c4987 Automated Code Change A. Unique TensorFlower 2025-12-23 07:38:25 -08:00
  • 96f26c9527 Automated Code Change A. Unique TensorFlower 2025-12-23 07:29:42 -08:00
  • 2505aa5841 Automated Code Change A. Unique TensorFlower 2025-12-23 07:29:14 -08:00
  • 0c5029c218 Automated Code Change A. Unique TensorFlower 2025-12-23 07:11:07 -08:00
  • dcdba180c5 Automated Code Change A. Unique TensorFlower 2025-12-23 04:52:24 -08:00
  • f42e8ace05 Automated Code Change A. Unique TensorFlower 2025-12-23 04:07:29 -08:00
  • 4a49b39e2b Automated Code Change A. Unique TensorFlower 2025-12-23 03:53:31 -08:00
  • 8f7c36563c Automated Code Change A. Unique TensorFlower 2025-12-23 03:03:58 -08:00
  • 44359ffdde [ReplicaGroupV3][Refactor][5/n] Update partitioner code to use CollectiveDeviceList (V1 replica group) in place of vector<vector<int>>. Bill Varcho 2025-12-23 02:38:35 -08:00
  • bc1b277391 Automated Code Change A. Unique TensorFlower 2025-12-23 02:00:37 -08:00
  • 6e5c77d370 Automated Code Change A. Unique TensorFlower 2025-12-23 01:59:33 -08:00
  • 6ba4ee05fd Automated Code Change A. Unique TensorFlower 2025-12-23 01:57:08 -08:00
  • 6a87077928 Automated Code Change A. Unique TensorFlower 2025-12-23 01:47:10 -08:00
  • 55e5af9a8e Automated Code Change A. Unique TensorFlower 2025-12-23 01:47:07 -08:00
  • 38b33b3d64 Automated Code Change A. Unique TensorFlower 2025-12-23 01:44:09 -08:00
  • 863879c12c Automated Code Change A. Unique TensorFlower 2025-12-23 01:40:03 -08:00
  • 6051ff167a Automated Code Change A. Unique TensorFlower 2025-12-23 01:39:36 -08:00
  • 117cd46dcb Automated Code Change A. Unique TensorFlower 2025-12-23 01:39:34 -08:00
  • 1dc6721ee5 Automated Code Change A. Unique TensorFlower 2025-12-23 01:36:39 -08:00
  • d612103dcf Automated Code Change A. Unique TensorFlower 2025-12-23 01:35:35 -08:00
  • 2c2340a46b Automated Code Change A. Unique TensorFlower 2025-12-23 01:35:33 -08:00
  • 1c22acd932 Automated Code Change A. Unique TensorFlower 2025-12-23 01:30:42 -08:00
  • 07f7911434 Automated Code Change A. Unique TensorFlower 2025-12-23 01:30:26 -08:00
  • c313fc596d Update GraphDef version to 2450. A. Unique TensorFlower 2025-12-23 01:04:06 -08:00
  • c19720a481 compat: Update forward compatibility horizon to 2025-12-23 A. Unique TensorFlower 2025-12-23 01:03:58 -08:00
  • ccbe08ca2e Merge pull request #105471 from Redempt1onzzZZ:master TensorFlower Gardener 2025-12-22 23:40:27 -08:00
  • 458b16c38a Automated Code Change A. Unique TensorFlower 2025-12-22 23:10:04 -08:00
  • 544a1362cf Automated Code Change A. Unique TensorFlower 2025-12-22 22:45:50 -08:00
  • 05c8cf7c13 respect IGNORE bad indices policy A. Unique TensorFlower 2025-12-22 22:39:43 -08:00
  • 87b04bd84f Merge pull request #105392 from tensorflow:dependabot/docker/tensorflow/tools/tf_sig_build_dockerfiles/ubuntu-104ae83 TensorFlower Gardener 2025-12-22 22:16:57 -08:00
  • eed068a5ea Automated Code Change A. Unique TensorFlower 2025-12-22 21:16:50 -08:00
  • 86c82562e1 This is an internal change A. Unique TensorFlower 2025-12-22 17:19:54 -08:00
  • 6160600087 Remove ml_build_arm64 Dockerfile Quoc Truong 2025-12-22 17:19:50 -08:00
  • dd10786acd Migrate conditional_test to PjRt. Niklas Vangerow 2025-12-22 15:58:02 -08:00
  • 69ea2a9308 Allow prefetching an hlo value if its use is colored in alternate memory even if if loop optimizer has decided otherwise. Subhankar Shah 2025-12-22 13:47:19 -08:00
  • 21d80205a6 Add PJRT_Buffer_DonateWithControlDependency to the PJRT C API. Haibo Huang 2025-12-22 13:32:14 -08:00
  • 3480eee02b Add HloModuleFromXlaComputation to HloRunnerAgnosticTestBase. Niklas Vangerow 2025-12-22 12:54:52 -08:00
  • b7650e843b Add proto serialization for CollectivePermuteStartThunk Maxim Ermilov 2025-12-22 12:21:12 -08:00
  • 678058948b [Autotuner] Limit CuDNN tests to CuDNN autotuner backend. Dirk Hornung 2025-12-22 12:05:54 -08:00
  • 1fa15367ad Add proto serialization for RaggedAllToAllStartThunk Maxim Ermilov 2025-12-22 11:28:27 -08:00
  • fec780d7fe Set FC's keep_num_dims to false when output dims is different from input dims after quantization. Byungchul Kim 2025-12-22 10:38:48 -08:00
  • 9ca49fcfa5 Limit CublasDot deterministic test to Cublas autotuning backend. Dirk Hornung 2025-12-22 10:08:37 -08:00
  • 573bbe2b41 Migrates builder.create<Op>() => Op::create() in tablegen files A. Unique TensorFlower 2025-12-22 09:45:57 -08:00
  • 3cec0d7b92 [XLA:GPU] Clean up RaggedAllToAllStartThunk rendezvous helpers. Oleg Shyshkov 2025-12-22 08:59:00 -08:00
  • af38f913d0 Automated Code Change A. Unique TensorFlower 2025-12-22 07:27:26 -08:00
  • 3ea706cab3 Add --xla_gpu_experimental_autotune_backends to allow for selecting backends. This change for the new autotuner. The new autotuner with its Triton backend competes with cuDNN fusions leading to flaky tests. Also some tests disable some autotuning paths via --xla_gpu_cudnn_gemm_fusion_level or --xla_gpu_cublas_fallback which are not fully compatible with the new autotuner. Other tests rely on the order of the backends, which would be resolved by adding a backend selection mechanism. Dirk Hornung 2025-12-22 07:04:01 -08:00
  • 4d0edd395f Refactor std::optional comparison in ReshapeSharding tests Kanish Anand 2025-12-22 07:00:54 -08:00
  • 12502acbf5 Remove unnecessary if_gpu_is_configured from Triton tests. Henning Becker 2025-12-22 06:18:09 -08:00
  • 2f90852c17 [XLA:GPU] Remove TF_ prefix from RETURN_IF_ERROR and ASSIGN_OR_RETURN macros. Oleg Shyshkov 2025-12-22 04:50:10 -08:00
  • d0b7f40548 [tosa] : fixing dynamic batch handling in FullyConnected legalization (#106638) deeptanshusekhri 2025-12-22 12:10:39 +00:00
  • f5b102299e [Autotuner] Log autotuner config in readable json format. When debugging the autotuner we often want to know the values of the AutotuneConfig. Dirk Hornung 2025-12-22 02:41:59 -08:00
  • 23dd865ee5 Remove redundant TENSORFLOW_USE_ROCM define. Henning Becker 2025-12-22 02:22:47 -08:00
  • dfc5b243ca Automated Code Change A. Unique TensorFlower 2025-12-22 01:45:07 -08:00
  • 79af5068fd [Autotuner] Avoid compiling all configurations if we only return the first one. This happens when we want to select the first configuration that successfuly compiles. E.g. for determinism. Dirk Hornung 2025-12-22 01:09:10 -08:00
  • d48869043b compat: Update forward compatibility horizon to 2025-12-22 A. Unique TensorFlower 2025-12-22 01:04:16 -08:00
  • 14b51dd700 Update GraphDef version to 2449. A. Unique TensorFlower 2025-12-22 01:04:04 -08:00
  • 85172d7831 [XLA:GPU] Shard the gpu_compiler_test. The _h100 test regularly causes timeouts. Dirk Hornung 2025-12-22 01:02:40 -08:00
  • 37da2f6658 Automated Code Change A. Unique TensorFlower 2025-12-22 00:39:25 -08:00
  • ec8a966f0d Automated Code Change A. Unique TensorFlower 2025-12-22 00:27:53 -08:00
  • 2e5b1e44fc Automated Code Change A. Unique TensorFlower 2025-12-22 00:27:24 -08:00
  • 53c2f78993 Automated Code Change A. Unique TensorFlower 2025-12-22 00:24:57 -08:00
  • 3c7c52e730 Automated Code Change A. Unique TensorFlower 2025-12-22 00:17:58 -08:00
  • 6e5d62bf3e Increase shards for fusion_emitter_device_test to speed up the test. Dirk Hornung 2025-12-21 23:54:13 -08:00
  • bb8c750b2f Automated Code Change A. Unique TensorFlower 2025-12-21 23:37:50 -08:00
  • b1d2538541 [Autotuner] Initialize random input values for buffer checks. If values are initialized to 0 buffer checker will fail to detect backends with wrong results. Dirk Hornung 2025-12-21 23:34:56 -08:00
  • 7b0d71c54e Automated Code Change A. Unique TensorFlower 2025-12-21 23:26:00 -08:00
  • 64337a1e3c Automated Code Change A. Unique TensorFlower 2025-12-21 23:25:05 -08:00
  • 6165d577f9 Automated Code Change A. Unique TensorFlower 2025-12-21 23:17:28 -08:00
  • 2b621d61f9 Automated Code Change A. Unique TensorFlower 2025-12-21 23:17:07 -08:00
  • f4e53263b1 Automated Code Change A. Unique TensorFlower 2025-12-21 23:16:33 -08:00
  • e3e3bc1946 Reverts c549ee47f8 A. Unique TensorFlower 2025-12-21 18:04:49 -08:00
  • 7733c4c03d Use StartDetachedThread instead of SchedClosure to dispatch atom program compilation Junwhan Ahn 2025-12-21 17:25:22 -08:00
  • ce61030c67 Automated Code Change A. Unique TensorFlower 2025-12-21 13:16:18 -08:00
  • 0cfba6c852 Automated Code Change A. Unique TensorFlower 2025-12-21 07:27:23 -08:00
  • f356a762f3 Automated Code Change A. Unique TensorFlower 2025-12-21 07:27:16 -08:00
  • 630698a3af Automated Code Change A. Unique TensorFlower 2025-12-21 07:25:04 -08:00
  • f253afed70 Automated Code Change A. Unique TensorFlower 2025-12-21 07:16:50 -08:00
  • 5042531aa8 Moving definitions to cpp file, match function definition declaration order Kanish Anand 2025-12-21 04:52:21 -08:00
  • a27a856d1c Automated Code Change A. Unique TensorFlower 2025-12-21 02:50:18 -08:00
  • ffb02301df Update GraphDef version to 2448. A. Unique TensorFlower 2025-12-21 01:04:06 -08:00