tensorflow - machine learning framework

	Commit message (Collapse)	Author	Age
*	Removed redundant std::string -> string conversions.	A. Unique TensorFlower	2018-08-28
\| \| \| \|	PiperOrigin-RevId: 210596417
*	Replaced calls to tensorflow::StringPiece::ToString with std::string ↵	Peter Hawkins	2018-05-02
\| \| \| \| \| \| \| \| \| \|	conversions. That is, instances of sp.ToString() are replaced with std::string(sp). This will allow tensorflow::StringPiece::ToString to be removed, which is necessary before it can be replaced with absl::string_view. PiperOrigin-RevId: 195162126
*	Expose data type information in checkpoint reader.	A. Unique TensorFlower	2017-10-05
\| \| \| \|	PiperOrigin-RevId: 171147196
*	Replace owning raw pointers with unique pointers	A. Unique TensorFlower	2017-10-05
\| \| \| \|	PiperOrigin-RevId: 171132628
*	Fix memory leakage of CheckpointReader	A. Unique TensorFlower	2017-02-08
\| \| \| \|	Change: 146955950
*	checkpoint_reader: fix VarToShapeMap V2 impl.	Zongheng Yang	2016-10-14
\| \| \| \| \| \|	This change makes it adhere to the original semantics: all slices of a partitioned tensor are grouped under one entry. Change: 136229541
*	tensor_bundle: tighten Lookup() semantics, add LookupDtypeShape().	Zongheng Yang	2016-10-07
\| \| \| \| \|	This fixes a subtle allocation issue in CheckpointReader. Change: 135503615
*	TF Checkpoint V2: make CheckpointReader work with the V2 format.	Zongheng Yang	2016-09-28
\| \| \| \| \| \| \| \| \| \| \| \| \|	If the same checkpoint prefix identifies both a V1 checkpoint and a V2 checkpoint on disk, the V2 version takes priority -- which matches the same behavior as the RestoreV2 op. Typical usage: $ bazel run tensorflow/python/tools:inspect_checkpoint -- --file_name=<V2 ckpt prefix> Other changes: add DebugString() and Contains() to BundleReader. Change: 134543092
*	Move C API files (and related files used by SWIG wrappers)	A. Unique TensorFlower	2016-07-29
	to new tensorflow/c directory. Change: 128855990