tensorflow - machine learning framework

	Commit message (Collapse)	Author	Age
*	Added gradient op for QR decomposition	A. Unique TensorFlower	2017-10-20
\| \| \| \|	PiperOrigin-RevId: 172895297
*	Replace tf.conj(tf.transpose(foo)) and tf.transpose(tf.conj(foo)) idioms ↵	A. Unique TensorFlower	2017-10-12
\| \| \| \| \| \|	with tf.linalg.adjoint(foo) or tf.transpose(foo, conjugate=True), and clean up a few places that can avoid explicit adjoints as inputs to matmul. PiperOrigin-RevId: 172027859
*	* Add mechanism to CudaSolver for capturing references to temporary tensors. ↵	A. Unique TensorFlower	2017-09-29
\| \| \| \| \| \| \| \| \| \| \| \|	This way users of the class don't have to remember to capture each one manually to avoid premature deallocation and memory races for asynchronous op kernels. * Add simple tests that run multiple ops concurrently for linalg ops that use CudaSolver. * Put a lock around the calls to cusolverDngetrs and cusolverDngesvd, which appear not to be threadsafe. * Misc. cleanup in linalg GPU kernels. I ran all the related tests 1000 times without failure. Before this change, tests for matrix_solve and svd would fail or hang occasionally. PiperOrigin-RevId: 170557380
*	Enable GPU support of QR decomposition for the cases left out in cl/169249336.	A. Unique TensorFlower	2017-09-19
\| \| \| \|	PiperOrigin-RevId: 169329120
*	Add GPU support for QR decomposition.	A. Unique TensorFlower	2017-09-19
\| \| \| \| \| \|	Remove support support for on-the-fly transpose in internal matrix_band_part functor recently added (in anticipation of using it for QR), since it turned out to not be useful. PiperOrigin-RevId: 169249336
*	Remove hourglass imports from kernel_tests	Justine Tunney	2016-12-14
\| \| \| \|	Change: 142080137
*	Adding a TF QR op.	A. Unique TensorFlower	2016-11-22
	Change: 139959769