Artem Volkhin 000db87bc7 Half-floats support for the rest of segment ops ... Summary: previously fp16 type was supported in SparseLengthsSum operator, now it works in all other segment operator as well. Reviewed By: dzhulgakov Differential Revision: D4624312 fbshipit-source-id: c9d72110e3762167270bb088405eaf9c56e88493		2017-02-28 11:19:15 -08:00
..
__init__.py	fbsync. TODO: check if build files need update.	2016-11-15 00:00:46 -08:00
batch_lr_loss.py	NextScopedBlob with well-defined behavior and respect namescope	2017-02-16 17:16:36 -08:00
concat.py	NextScopedBlob with well-defined behavior and respect namescope	2017-02-16 17:16:36 -08:00
dot_product.py	NextScopedBlob with well-defined behavior and respect namescope	2017-02-16 17:16:36 -08:00
expand_dims.py	NextScopedBlob with well-defined behavior and respect namescope	2017-02-16 17:16:36 -08:00
fc.py	model and preprocessor can handle empty dense inputs	2017-02-22 11:19:15 -08:00
functional.py	Add a way do describe layers in a more AdHoc manner.	2017-02-27 23:30:39 -08:00
layers.py	Add a way do describe layers in a more AdHoc manner.	2017-02-27 23:30:39 -08:00
simple_operator_layers.py	Fix random issues with some of the layers getting missing from registry.	2017-01-10 15:14:31 -08:00
sparse_lookup.py	Half-floats support for the rest of segment ops	2017-02-28 11:19:15 -08:00
sparse_to_dense.py	NextScopedBlob with well-defined behavior and respect namescope	2017-02-16 17:16:36 -08:00
split.py	NextScopedBlob with well-defined behavior and respect namescope	2017-02-16 17:16:36 -08:00
tags.py	fbsync. TODO: check if build files need update.	2016-11-15 00:00:46 -08:00