update to latest#9

Merged

sourcecode369 merged 473 commits into

sourcecode369:masterfrom

tensorflow:master

Jul 10, 2020

This pull request is big! We're only showing the most recent 250 commits

Commits on Jul 8, 2020

Merge pull request #40585 from yongtang:40577-tensor_scatter_nd_add-gpu
tensorflower-gardener
committed
Remove ReplicaContext.__enter__ and __exit__ methods from the docs since users are not expected to use them directly.

guptapriya
authored and
tensorflower-gardener
committed
Merge pull request #40158 from gaurav1086:fix_segmentation_fault_in_segment_graph
tensorflower-gardener
committed
Fully init TF_TString in order to mitigate potential msan false positives.

gharibian
authored and
tensorflower-gardener
committed
Rolling forward with #include of missing logging.h header.

bmzhao
authored and
tensorflower-gardener
committed
Integrate LLVM at https://github.com/llvm/llvm-project/commit/f54d0e36be6a

Stephan Herhut
authored and
tensorflower-gardener
committed
Updates to Profiler Guide

anirudh161
authored and
tensorflower-gardener
committed
[MLIR] Convert FuncOp signature with unranked types in HLO->LHLO conversion.

pifon2a
authored and
tensorflower-gardener
committed
Update GraphDef version to 456.
tensorflower-gardener
committed
compat: Update forward compatibility horizon to 2020-07-08
tensorflower-gardener
committed
[MLIR][LHLO] Convert mhlo.dynamic_reshape -> lhlo.reshape_memref_cast.

pifon2a
authored and
tensorflower-gardener
committed
Add error message for outfeed shape mismatch.
tensorflower-gardener
committed
Fix docstring
lgeiger
committed
Add rename, move, delete recursiveley
vnghia
committed
[MLIR][LHLO] Legalize CallOp that call funcs with tensor args/results.

pifon2a
authored and
tensorflower-gardener
committed
fix
ZhuBaohe
committed
[XLA:CPU] Teach dot_op_emitter how to tile&vectorize linalg matmuls

d0k
authored and
tensorflower-gardener
committed
Remove broken (and unused) kernel targets.

akuegel
authored and
tensorflower-gardener
committed
Integrate LLVM at https://github.com/llvm/llvm-project/commit/1ea289681acf

Stephan Herhut
authored and
tensorflower-gardener
committed
Add expiring_lru_cache
vnghia
committed
[TF:TRT] Fix the TF-TRT bridge to respect the device assignment information in

bixia1
authored and
tensorflower-gardener
committed
Merge pull request #41158 from vnvo2409:gcs-ram
tensorflower-gardener
committed
Qualify uses of std::string
tensorflower-gardener
committed
Disable MLIR bridge for NMS image ops test

smit-hinsu
authored and
tensorflower-gardener
committed
Make the "untracked resource" error less awful

allenlavoie
authored and
tensorflower-gardener
committed
Integrate LLVM at https://github.com/llvm/llvm-project/commit/24b62f28c5da
tensorflower-gardener
committed
Fork the checkpointing_test and move the keras related test to keras/distribute.

qlzh727
authored and
tensorflower-gardener
committed
Fix msan false positive for //third_party/tensorflow/go:tensorflow_test

gharibian
authored and
tensorflower-gardener
committed
Internal change

jpienaar
authored and
tensorflower-gardener
committed
[MLIR][NFC] Adopt variadic isa<> on MLIR Types and Attributes

jurahul
authored and
tensorflower-gardener
committed
Qualify uses of std::string
tensorflower-gardener
committed
Update syntax in pip_smoke_test.

Amit Patankar
authored and
tensorflower-gardener
committed
Add SparseTensor support to assert_type.
tensorflower-gardener
committed
Fix a race condition in attr_builder where the AttrTypeMap for the same op_name

Kuangyuan Chen
authored and
tensorflower-gardener
committed
Replace instances of "whitelist" with "allowlist" where possible. See Google Developer guidelines at https://developers.google.com/style/word-list#blacklist for more information.

karmel
authored and
tensorflower-gardener
committed
Rename xla_lhlo dialect into lmhlo

joker-eph
authored and
tensorflower-gardener
committed
Rename xla_chlo dialect into chlo

joker-eph
authored and
tensorflower-gardener
committed
Update v1 only saved_model_experimental_test test with proper reason.

pavithrasv
authored and
tensorflower-gardener
committed
Call TfTpu_Initialize() to initialize relevant bits of the TPU framework when loading the TPU library

Frank Chen
authored and
tensorflower-gardener
committed
Rename XlaHloDialect class into MhloDialect following the recent dialect namespace renaming

joker-eph
authored and
tensorflower-gardener
committed
Adjust error reporting and messages with remote targets.

haoyuz
authored and
tensorflower-gardener
committed
Fix typo in a message.
tensorflower-gardener
committed
Allow computing gradients of If/While in imported graphs.

saxenasaurabh
authored and
tensorflower-gardener
committed
Update whitelist to allowlist in comment.

lucarfox
authored and
tensorflower-gardener
committed
Add mlir_cpu_runner tests infra for CHLO->LHLO->LLVM lowering.

pifon2a
authored and
tensorflower-gardener
committed
Lower tf.IsNan op to an equivalent computation using tf.Equal op.

lucarfox
authored and
tensorflower-gardener
committed
Correctly set the experimental_io_device when restoring variable from a checkpoint.

kenfranko
authored and
tensorflower-gardener
committed
Update TF_AllocateOutput to just call ctx->allocate_output now that TF_Tensor just wraps a Tensor.

annarev
authored and
tensorflower-gardener
committed
Expose Interpreter from InterpreterWrapper.
tensorflower-gardener
committed
Make ParseOpData a stub function that returns an error for TFLM.

advaitjain
authored and
tensorflower-gardener
committed
Run gpu doctest with a fixed set of devices

crccw
authored and
tensorflower-gardener
committed
Add helper functions to tf_device.replicate op for determining and accessing replicated and packed inputs from block arguments.

andyly
authored and
tensorflower-gardener
committed
Go: Update generated wrapper functions for TensorFlow ops.
tensorflower-gardener
committed
Merge pull request #41144 from xingyu-long:xingyu_simplify_cpu_example
tensorflower-gardener
committed
Canonicalize identity arithmetic operations with dynamic shapes

ezhulenev
authored and
tensorflower-gardener
committed
Clean up the LSTM/GRU layer with the new grappler selector.

yhliang2018
authored and
tensorflower-gardener
committed
Add tf.TPUPartitionedCall to TensorFlow MLIR ODS.

andyly
authored and
tensorflower-gardener
committed
[tf.data] Change buckets for tf.data's iterator.getNext() duration.

Michael Kuchnik
authored and
tensorflower-gardener
committed
Fix cleanup of TFLite Java NNAPI delegate

jdduke
authored and
tensorflower-gardener
committed
Add shape inference pass to the beginning of the TPUBridgeV1Compat pipeline.

andyly
authored and
tensorflower-gardener
committed
Add pass to update tpu embedding ops pass.
tensorflower-gardener
committed
Fix critical bug with `add_loss` TFOpLayer graph construction

fchollet
authored and
tensorflower-gardener
committed
[MLIR:TF] Canonicalize redundant reshape

ezhulenev
authored and
tensorflower-gardener
committed
HloModule::Clone function copies input_output_alias_config. Without this copy, the cloned module is not identical to the original module because it does not have the input_output_alias_config.

mangpo
authored and
tensorflower-gardener
committed
Update the docstring of reduce_to and batch_reduce_to

crccw
authored and
tensorflower-gardener
committed
Update Grappler Remapper documentation to include DepthwiseConv2dNative fusion.

lucarfox
authored and
tensorflower-gardener
committed
Reland "Fix issues with 32-bit ARM builds"

jdduke
authored and
tensorflower-gardener
committed
Update TPUUpdateEmbeddingEnqueueOpInputs pass to use llvm::StringMap instead of std::map and fix error message style. (NFC)

andyly
authored and
tensorflower-gardener
committed
[TF-numpy] Re-enables signature check against numpy, and fixes problems found by the check.

wangpengmit
authored and
tensorflower-gardener
committed
Convert path to string right before passing it to the filesystem layer
lgeiger
committed
Update v1 only training/adagrad_test with proper reason.

pavithrasv
authored and
tensorflower-gardener
committed
Fix typo, AddIntemediateTensorsToFusedOp to AddIntermediateTensorsToFusedOp.
tensorflower-gardener
committed
Update TF's six dependency.

gunan
authored and
tensorflower-gardener
committed
Expose XlaOptions as an argument to tpu.rewrite

Yuanzhong Xu
authored and
tensorflower-gardener
committed
Improve tf.train.list_variables API doc.

pavithrasv
authored and
tensorflower-gardener
committed
Create consume_leading_digits_fuzz.cc
grasskin
authored
Create arg_def_case_fuzz.cc
grasskin
authored
Create consume_non_whitespace_fuzz.cc
grasskin
authored
Update BUILD
grasskin
authored
Expose x*sigmoid(x) as silu in addition to swish, and credit silu paper.

martinwicke
authored and
tensorflower-gardener
committed
[ROCm] Fix for ROCm Breakage - 200708
deven-amd
committed
Add dependencies to TPU library for graph rewrite registration

Frank Chen
authored and
tensorflower-gardener
committed
Internal Build Change.

tf-marissaw
authored and
tensorflower-gardener
committed
Implement policy for constant folding TF ops to avoid creating large constants.

Prakalp Srivastava
authored and
tensorflower-gardener
committed
Disable test which has flaky timeouts and breaks build

mihaimaruseac
authored and
tensorflower-gardener
committed

Commits on Jul 9, 2020

Commits on Jul 10, 2020