Support for quantized LstmCell, with initial reference runtime code.
authorA. Unique TensorFlower <gardener@tensorflow.org>
Wed, 7 Feb 2018 14:30:32 +0000 (06:30 -0800)
committerTensorFlower Gardener <gardener@tensorflow.org>
Wed, 7 Feb 2018 14:34:07 +0000 (06:34 -0800)
commit0255d9a9ca66d233ca4befcceff5d6e70b16d0e0
treeee6d7318e18b228d17e41d64b78c677ed238a8ae
parent68a7f292b324e08fefd2c47dde444720db361f4d
Support for quantized LstmCell, with initial reference runtime code.
The current 'optimized' code is just a copy of the reference code,
a true optimized implementation will follow separately.

PiperOrigin-RevId: 184830223
12 files changed:
tensorflow/contrib/lite/kernels/internal/common.h
tensorflow/contrib/lite/kernels/internal/optimized/optimized_ops.h
tensorflow/contrib/lite/kernels/internal/quantization_util.cc
tensorflow/contrib/lite/kernels/internal/quantization_util.h
tensorflow/contrib/lite/kernels/internal/quantization_util_test.cc
tensorflow/contrib/lite/kernels/internal/reference/reference_ops.h
tensorflow/contrib/lite/toco/graph_transformations/hardcode_min_max.cc
tensorflow/contrib/lite/toco/graph_transformations/quantize.cc
tensorflow/contrib/lite/toco/model.h
tensorflow/contrib/lite/toco/tooling_util.cc
tensorflow/contrib/lite/toco/tooling_util.h
tensorflow/workspace.bzl