[pull] master from tensorflow:master#4058
Merged
pull[bot] merged 43 commits intoosmaelo:masterfrom Aug 6, 2020
Merged
Conversation
…lumn, IdentityColumn and WeightCategoricalColumn. PiperOrigin-RevId: 325090178 Change-Id: I427fae37945eeed690b99bc959e7757dea7086ed
PiperOrigin-RevId: 325090279 Change-Id: I69802041abfd269e705b8a4fe98e0ddb94fdc772
PiperOrigin-RevId: 325091138 Change-Id: I61c6188bd9985858eefe3c8fe8bf833991e20f29
PiperOrigin-RevId: 325094476 Change-Id: I6e0ed20c868dd80f2f3d8735ecd126108bf20a02
Moreover explicit check for whether XLA is linked in is not needed, as we already check for cpu_compiler/gpu_compiler linkage. PiperOrigin-RevId: 325095702 Change-Id: I82220c7bb15520cf3683fd944fef57d7f3f426e2
Const op does not allocate its output on every execution - rather it allocates a persistent tensor once, with default AllocatorAttributes, and reuses this everywhere. Since Const does not invoke any of the typical allocation functions on the OpKernelContext, such as `allocate_output` and `allocate_temp`, it is not compatible with the ScopedAllocatorOptimizer design. PiperOrigin-RevId: 325098062 Change-Id: Iae92870b2123a6368c2c91af9817a2e663fcbf47
PiperOrigin-RevId: 325098417 Change-Id: I29833a917181f5c8b21f2dfb55489ab2fd20e3b5
PiperOrigin-RevId: 325101084 Change-Id: I9f655ef9d3a9e71443d42dfc5c4e0597c987cae3
PiperOrigin-RevId: 325101427 Change-Id: I73831dac9a2a70575ad620f25207326afe3f76b0
PiperOrigin-RevId: 325112371 Change-Id: I96a704a62d5acf3305e0328eba621455bb4290bb
PiperOrigin-RevId: 325112494 Change-Id: Ib0e54e6084ef216b5d696d62682afd7680422f7f
PiperOrigin-RevId: 325112498 Change-Id: Ibd4641faf6871ebd58477f9fe945cd2e0ac641f8
PiperOrigin-RevId: 325113003 Change-Id: Ida571463512f3e2c834a25338ba7653b57e3cad9
PiperOrigin-RevId: 325116169 Change-Id: Ib96e80016131c3b0d5741f5dffbe0362ef21d330
PiperOrigin-RevId: 325117492 Change-Id: I9320d0ba23267d36020186dd823c3bb7dadd152e
…_DELEGATE_NO_GL is defined. PiperOrigin-RevId: 325119498 Change-Id: I8c8c6bdbc51a4c6fbf69ba0b4683371a07ccd4f5
… alternate mem. PiperOrigin-RevId: 325120226 Change-Id: I5f7416b3467e17613f10c99871b6326b58988ba8
…oupInterface` and `core_index` and makes CacheEntry less transparent and move application specific logics outside of cache. PiperOrigin-RevId: 325120711 Change-Id: I36f03ff585d80dc09a4d468c1b86e01a73e15859
Tested using grad(exp(x)). PiperOrigin-RevId: 325121430 Change-Id: Ib293198698aff00f8f8d8b376a8da62d4c2a8038
PiperOrigin-RevId: 325121569 Change-Id: I3bee9c46f6d0622781b3fe684a790fab061d6699
For partial replication partial tiling, add a trailing dimension to tile_assignment, representing the subgroups of replication. PiperOrigin-RevId: 325123131 Change-Id: I0e86b492e286d66d329dd9719faa0d47027f89b1
timespec= isn't compatible with Python 3.5. PiperOrigin-RevId: 325124472 Change-Id: Id0a88eb60d0483d3f29e1c0e64677a637ec5a2b4
Updates LLVM usage to match [acb66b9111ba](llvm/llvm-project@acb66b9111ba) PiperOrigin-RevId: 325125249 Change-Id: I149d91bf019a0f39a71a1309fe4df4d1c16ba079
… strict. * Most kernels have been ported over to use the new EvalTensor API. The exceptions are kernels in arc_mli, ethos-u, xtensa_hifi and xtensa_hifimini_staging. * The kernel tests were previously changed to where they would fail unless they used EvalTensors * This change enforces the fact that kernels should no longer be using full TfLiteTensors in the Eval functions. Calling GetTensor from a kernel's Eval will now crash due to dereferencing a nullptr. PiperOrigin-RevId: 325126736 Change-Id: I1a1c9af7dcca5765461794a60ef626387fa82638
PiperOrigin-RevId: 325128806 Change-Id: I610db52800e55887e93b45b145718e61ceedbb79
PiperOrigin-RevId: 325131400 Change-Id: Ic600f9c21f2080ab2fa62e771f85bd7fcabaded3
PiperOrigin-RevId: 325132234 Change-Id: I705a857042f4f42906723c3b608c2ff621a17f81
…al::IdForHost(). This also refactors TpuCoreLocationExternal::chip_coordinates() and downstream plumbing to be more compact. PiperOrigin-RevId: 325135180 Change-Id: Ic350d3d64b01fced4d6d90551f71e97b271e0f2b
…for TPU embeddings, behind an optimization parameter setting because it hurts performance. PiperOrigin-RevId: 325137678 Change-Id: Iee7331beb6f76cdf93a7ffd33f1b8cd083b624a5
PiperOrigin-RevId: 325140153 Change-Id: Id3c4821ce4ac1893bf3f1140e94e056f7fd87293
PiperOrigin-RevId: 325140951 Change-Id: Ib855a6b696b6b3cf31808b533c16ce1814c720e7
PiperOrigin-RevId: 325147676 Change-Id: I22d2e5de747c4efb278c9d2b7f2de17e968db9b4
PiperOrigin-RevId: 325148493 Change-Id: I36727dec4e4217bd085d77033db8cac3c1c89b1a
PiperOrigin-RevId: 325150437 Change-Id: If1b31f931895d0c656e3a56f96e7e176ef511029
PiperOrigin-RevId: 325151668 Change-Id: I1a0ac5d58e8237cf47785034086c7cdc240ba116
…lassification in TFLite Model Maker. PiperOrigin-RevId: 325152885 Change-Id: I7b1d4745b4bdcefb3308f28a4b26f9f400866cc3
If XLA support is already enabled, there is no need to set the environment variable (which is only a trigger to enable this support) PiperOrigin-RevId: 325154534 Change-Id: Iab9bf804050d0c87ad8b1b01290fec2879fd5eb9
… and size one in the concatenate dimension PiperOrigin-RevId: 325159041 Change-Id: Iba80992e57c173039911bd609a646715e1450961
PiperOrigin-RevId: 325159511 Change-Id: I05e6e8a192d054b8db56c4e00d500c95b832ad92
PiperOrigin-RevId: 325160099 Change-Id: I6fcd935790b361f8afdb085ed36f973f24cd7dbd
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
20 participants
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
See Commits and Changes for more details.
Created by
pull[bot]. Want to support this open source service? Please star it : )