- 25 8月, 2020 19 次提交
-
-
由 A. Unique TensorFlower 提交于
PiperOrigin-RevId: 328181751 Change-Id: Icc9c27218a9024e64b5a95a1cd35d83c46ab876b
-
由 A. Unique TensorFlower 提交于
PiperOrigin-RevId: 328181545 Change-Id: I4e5d52e775a060fe686ee623986b423787171e81
-
由 Stephan Hoyer 提交于
__array_module__ is an experimental protocol for "duck array" compatibility that for indicating how to find a "numpy compatible" module. The hope is to make it easier to write generic code that works across a range of array libraries. A full example, which should work equally work for TF-NumPy as for JAX, can be found at https://github.com/google/jax/pull/4076. More examples, and motivation for this protocol can be found at https://numpy.org/neps/nep-0037-array-module.html. This design has not yet been finalized in NumPy, so at present it requires using the experimental numpy_dispatch module: https://github.com/seberg/numpy-dispatch. Unlike NumPy's __array_ufunc__ and __array_function__ protocols, __array_module__ by design has no backwards compatibility consequences. The protocol only controls the behavior of numpy_dispatch.get_array_module() or numpy.get_array_module() -- it does not change any existing NumPy functions. PiperOrigin-RevId: 328181308 Change-Id: I98b648a59709ed3bc295aaf97f471a1ff7ae1f05
-
由 A. Unique TensorFlower 提交于
PiperOrigin-RevId: 328180923 Change-Id: I7c1032442a881346bda0c09412cc089fe54f7724
-
由 Katherine Wu 提交于
Force long-running model revive tests to run in separate shards by putting them in a single Test class. PiperOrigin-RevId: 328180778 Change-Id: Id0c7d39de91b7ca87c9a7ee874c41ee29f505970
-
由 A. Unique TensorFlower 提交于
Because jax.jit is annotated on top-level functions, and that backends are initialised only after GoogleInit, we cannot have the client when calling the C++ jit. PiperOrigin-RevId: 328177723 Change-Id: I462385e4a687461bf41c0d3e2875d4736285c42d
-
由 Tim Shen 提交于
PiperOrigin-RevId: 328177436 Change-Id: I76f5bbbc4ce6a235e9e5cfd5df32802d207be849
-
由 Mehdi Amini 提交于
This is part of an effort to remove entirely the global constructor registration in favor of explicit registry in order to get more manageable build dependencies. PiperOrigin-RevId: 328176806 Change-Id: I2dd02eef77d439fcaf1f79f2fd537b8021aa2e2d
-
由 TensorFlower Gardener 提交于
PiperOrigin-RevId: 328175218 Change-Id: I3e4aca437f68bc748595eff537799b3a809c5c24
-
由 Mehdi Amini 提交于
Update third_party/tensorflow/compiler/mlir/tensorflow/utils/... to not depend on the global Dialect Registry (NFC) PiperOrigin-RevId: 328171679 Change-Id: I3a4e40a04cb14b9c3d53239b31d3d642bb97daac
-
由 Blake Hechtman 提交于
PiperOrigin-RevId: 328171480 Change-Id: I1adbe658c9e3f435d4a42c2627cbbfef297e02f3
-
由 Mehdi Amini 提交于
PiperOrigin-RevId: 328168585 Change-Id: I702964790826638b194eb7c5b21f98492e90c727
-
由 Mehdi Amini 提交于
PiperOrigin-RevId: 328164449 Change-Id: I2421d154e6c2f702def7d4274a4df51877f436aa
-
由 Mehdi Amini 提交于
This is part of a series of patches intended to make MLIR Dialects explictly registered and remove the "always_link" global-constructor based registration. The registration is mainly for tools that need to parse a "TF" MLIR input. PiperOrigin-RevId: 328161577 Change-Id: I533930a804d2a22b76d0ef36b5970b80500542ea
-
由 Scott Zhu 提交于
1. Update to use v2 public API from their v1 version. 2. Use the Gfile API to read/write file content for save_model code. PiperOrigin-RevId: 328159642 Change-Id: I38373fb16449ab8d19f15f4e22ad99d3c598266b
-
由 Ken Franko 提交于
Remove const type from vectors. PiperOrigin-RevId: 328157979 Change-Id: I8df58b0b23831b842c04c3243290ca61ecf7f4aa
-
由 Andrew Audibert 提交于
PiperOrigin-RevId: 328157361 Change-Id: I57675a6822cdd56c662aa877077d4688aa6a4411
-
由 A. Unique TensorFlower 提交于
PiperOrigin-RevId: 328150132 Change-Id: If1468b900e95398106fcca29d0263ebb8869731e
-
由 A. Unique TensorFlower 提交于
PiperOrigin-RevId: 328149666 Change-Id: I0c5561152383f12126ab9568c0facc4c3043c6a3
-
- 24 8月, 2020 21 次提交
-
-
由 Benjamin Kramer 提交于
Updates LLVM usage to match [bad7d6b3735d](https://github.com/llvm/llvm-project/commit/bad7d6b3735d) PiperOrigin-RevId: 328142991 Change-Id: I54a09bd386149cd5a649f5e0ca7bcc97028692c5
-
由 Jiri Simsa 提交于
PiperOrigin-RevId: 328141067 Change-Id: I154c55c9b1cc47ec069bfc663a1d3fb0f135e67d
-
由 TensorFlower Gardener 提交于
PiperOrigin-RevId: 328140122 Change-Id: Iacbb71fb3288057de3db6e0ab06e1cea20961223
-
由 TensorFlower Gardener 提交于
PiperOrigin-RevId: 328137876 Change-Id: I809b345ca9c54c352959987c904d122347436aa6
-
由 TensorFlower Gardener 提交于
PiperOrigin-RevId: 328134070 Change-Id: I3ad10cd9ee442e7f478192c3ea0e1254077c5f2e
-
由 TensorFlower Gardener 提交于
PiperOrigin-RevId: 328133653 Change-Id: I99ceafa58d99364665175222748ed1a4ae16b0e4
-
由 Rick Chao 提交于
PiperOrigin-RevId: 328129947 Change-Id: If3fc9bd3ccb4a271c60e48e052969c07174326e4
-
由 Thomas Joerg 提交于
PiperOrigin-RevId: 328128307 Change-Id: Ia9a5e9f4c20d78deee32b738ae1002d80eb935c1
-
由 Mehdi Amini 提交于
PiperOrigin-RevId: 328109152 Change-Id: Ia460e89f785e9a2aaf21538083733e7e13730299
-
由 A. Unique TensorFlower 提交于
PiperOrigin-RevId: 328100708 Change-Id: Ib79b4c72b07ce8c03bdaaac246f4bc492a116914
-
由 A. Unique TensorFlower 提交于
PiperOrigin-RevId: 328100698 Change-Id: Ia54de53b1a7abf269de44d5aa1817c14b9da1053
-
由 Mehdi Amini 提交于
Updates LLVM usage to match [b999400a4fb6](https://github.com/llvm/llvm-project/commit/b999400a4fb6) PiperOrigin-RevId: 328096769 Change-Id: I42c48819078ea634ec7a7bf4d73579c7ddda9c47
-
由 Kibeom Kim 提交于
TFE_TensorHandleToNumpy seems to be working with TFRT. PiperOrigin-RevId: 328077896 Change-Id: I0fc569e67440c00327009d87e50960734ed2bce1
-
由 A. Unique TensorFlower 提交于
Use BuiltinOpResolverWithoutDefaultDelegates instead of BuiltinOpResolver for unit tests of xnnpack delegate itself to prepare for enabling xnnpack delegate by default across all platforms in the next 2.4.0 release. PiperOrigin-RevId: 328076934 Change-Id: I69e21a6fbbe1b0e7146669ccd6481b774dcd9d2e
-
由 Mehdi Amini 提交于
Updates LLVM usage to match [f6decfa36d89](https://github.com/llvm/llvm-project/commit/f6decfa36d89) PiperOrigin-RevId: 328073633 Change-Id: I5cd74bcf36c453cf073766f910a0f8442b66cb93
-
由 Chao Mei 提交于
Use BuiltinOpResolverWithoutDefaultDelegates instead of BuiltinOpResolver for unit tests of xnnpack delegate itself to prepare for enabling xnnpack delegate by default across all platforms in the next 2.4.0 release. PiperOrigin-RevId: 328068258 Change-Id: I3459bc3e7f25d2925da65fba3e19ac2bad57fff1
-
由 Akshay Modi 提交于
PiperOrigin-RevId: 328068227 Change-Id: Ia084d946f3a0e5d071d7e8fec4263d1da26d9671
-
由 Mehdi Amini 提交于
Updates LLVM usage to match [f164534ca8e0](https://github.com/llvm/llvm-project/commit/f164534ca8e0) PiperOrigin-RevId: 328046788 Change-Id: I714164211a50e0d273ec49046c66f7e484989428
-
由 Yuanzhong Xu 提交于
1. Try to reuse the original target tiled sharding when finding compatible target from partial sharding. 2. If the HLO is a broadcast, check if data is already the same between source/target pairs. PiperOrigin-RevId: 328043490 Change-Id: I69dec53c50cb6cedf586afafc5181cd1ee29cdc6
-
由 Uday Bondhugula 提交于
Imported from GitHub PR https://github.com/tensorflow/tensorflow/pull/42508 An lmhlo.constant op on an memref that is locally allocated and with no users other than dealloc's can be deleted. Add a canonicalization pattern for this. Copybara import of the project: -- 8758c409a15f567e7cb8e1077faa020f5705c85a by Uday Bondhugula <uday@polymagelabs.com>: [MLIR] Erase dead lmhlo.constant ops An lmhlo.constant op on an memref that is locally allocated and with no other users (other than dealloc's) can be deleted. Add a canonicalization patter for this. COPYBARA_INTEGRATE_REVIEW=https://github.com/tensorflow/tensorflow/pull/42508 from polymage-labs:lhlo_constant_erase 8758c409a15f567e7cb8e1077faa020f5705c85a PiperOrigin-RevId: 328042416 Change-Id: I27f9b5b5297bbf6fe81aff589f009197b75f49eb
-
由 Eugene Burmako 提交于
MLIR is moving to require explicitly loading of Dialect before creating entities in a Dialect. PiperOrigin-RevId: 328037037 Change-Id: Ib46275b26e8f77aab0fbd0f70cd2a48844dc360c
-