tensorflow - machine learning framework

	Commit message (Collapse)	Author	Age
*	Global de-std::unique_ptr cleanup for xla::Literal.	A. Unique TensorFlower	2018-09-10
\| \| \| \|	PiperOrigin-RevId: 212313258
*	[XLA] Rename all (Mutable)ArraySlice to absl::Span.	Tim Shen	2018-08-30
\| \| \| \|	PiperOrigin-RevId: 210998142
*	Convert a couple more test files to HloVerifiedTestBase, and add default ↵	Dimitris Vardoulakis	2018-08-29
\| \| \| \| \| \|	arguments to the constructor to remove some boilerplate. PiperOrigin-RevId: 210855509
*	[XLA] Use absl::make_unique instead of xla::MakeUnique.	Justin Lebar	2018-08-20
\| \| \| \| \| \|	Same for WrapUnique. PiperOrigin-RevId: 209531124
*	[XLA] Make HloTestBase::GetDebugOptionsForTest virtual.	Justin Lebar	2018-08-20
\| \| \| \| \| \| \| \| \| \| \| \|	It's useful for tests to be able to override this and specify their own debug options. Making this function non-static requires a few add-on changes so that everything that calls it is non-static. For lambdas, this necessitates capturing [&] rather than [&foo], so that we can implicitly capture `this`. PiperOrigin-RevId: 209493738
*	[TF:XLA] Split literal_util into {literal, literal_util}.	Kay Zhu	2018-07-03
\| \| \| \| \| \| \| \| \|	Currently Literal classes sits in literal_util.{h,cc} instead of literal.{h,cc}. It also contains helper functions that are better fit to be their own separate class/namespace. This change starts this process by moving most static factory methods to LiteralUtil namespace. PiperOrigin-RevId: 203217065
*	[TF:XLA]	Dimitris Vardoulakis	2018-04-28
\| \| \| \| \| \| \|	- Require a module config when creating an HloModule. - All tests using HloTestBase create a module using CreateNewModule. PiperOrigin-RevId: 194684585
*	Minor cleanups to the gather expander; NFC	Sanjoy Das	2018-04-17
	This change is NFC now, but it makes the code more general and this generality will be used later on. For instance ExpandFirstDimIntoNDims(transposed_gather_indices, {1, shape.dimensions(0)}) does not work if shape is a scalar shape (and this fine because today shape is never scalar) but PrependDegenerateDims(transposed_gather_indices, 1) works fine if transposed_gather_indices is scalar (and it will be, in a future change). PiperOrigin-RevId: 193283404