tensorflow - machine learning framework

	Commit message (Collapse)	Author	Age
*	Import package xla_test instead of class XLATestCase.	A. Unique TensorFlower	2018-06-28
\| \| \| \|	PiperOrigin-RevId: 202572322
*	[TF:XLA] Implement ↵	Peter Hawkins	2018-06-22
\| \| \| \| \| \|	ResourceScatter{Update,Add,Sub,Mul,Div,Min,Max,NdAdd,NdUpdate}. PiperOrigin-RevId: 201723512
*	[TF:XLA] Add direct implementation of AssignVariableOp for XLA devices.	Peter Hawkins	2018-05-25
\| \| \| \| \| \|	This allows us to avoid an XLA compilation and tensor copies when assigning to a variable placed on an XLA device. PiperOrigin-RevId: 198127062
*	Update tests to enable flipping on bfloat16 types.	Jacques Pienaar	2018-04-02
\| \| \| \| \| \|	Skip individual tests that are currently failing with bfloat16. PiperOrigin-RevId: 191296618
*	[TF:XLA] Support for DT_INT64 in the VariableShape operation.	Asim Shankar	2017-12-06
\| \| \| \|	PiperOrigin-RevId: 178084701
*	[XLA:CPU] [XLA:GPU] Adds compiler support for C64 primitive type, including ↵	A. Unique TensorFlower	2017-10-27
\| \| \| \| \| \| \| \| \| \|	relevant elementwise unary and binary op lowering for CPU and GPU. We use a named LLVM struct "complex64", laid out the same as std::complex<float>. This named struct is accessed via the llvm::Module, which required changes to accessors of PrimitiveTypeToIrType & friends. Ops that require atan2 (in particular, angle and log) are only supported on GPU at this point. LLVM lacks a CPU intrinsic for atan or atan2, whereas libdevice provides this for GPU. PiperOrigin-RevId: 173676849
*	Reduce number of session.run calls from variable_ops_test.	A. Unique TensorFlower	2017-09-07
\| \| \| \|	PiperOrigin-RevId: 167902092
*	[TF:XLA] Implementing ResourceGather in TF2XLA.	A. Unique TensorFlower	2017-07-12
\| \| \| \|	PiperOrigin-RevId: 161730154
*	[TF:XLA] Add XLA implementation of ResourceStridedSliceAssign.	Peter Hawkins	2017-05-17
\| \| \| \|	PiperOrigin-RevId: 156341053
*	[TF:XLA] Fix compilation bug for computations with one non-constant output ↵	Peter Hawkins	2017-05-16
\| \| \| \| \| \|	and one or more variable outputs. PiperOrigin-RevId: 156193188
*	[TF:XLA] Improvements to resource variables:	Peter Hawkins	2017-02-27
\| \| \| \| \| \| \| \| \| \| \| \| \| \|	* enable compilation of VarIsInitializedOp. * fix deprecated variable initializer in variable_ops_test.py * simplify variable logic in XlaContext, move intelligence into XlaOpKernelContext. * add resource variable support in the contrib layers library. Cleanups and refactorings: * merge XlaCompiler::CompileSubComputation with XlaCompiler::CompileFunction. * pass XlaCompiler arguments consistently via XlaCompiler::Options. * split the two roles of XlaCompiler::CompilationResult::input_shapes into input_mapping and xla_input_shapes. * initialize the numpy and Python seeds to a constant for XLA test cases. Change: 148683645
*	[TF:XLA] Add support for resource variables to the Tensorflow/XLA bridge.	Peter Hawkins	2017-02-21
	Change: 148176223