Commit Graph

4 Commits

Author SHA1 Message Date
Jianlong Zhong
152d2ae3a8 Implement CUDA version of GRU operator
Summary: Add CUDA version of GRU operator

Reviewed By: jamesr66a

Differential Revision: D5571043

fbshipit-source-id: 332aa64fc8a9116cc33382f2b2907080e58c13b3
2017-08-08 10:57:40 -07:00
Robert Verkuil
97193478c7 Implemented GRUCell
Summary: Implemented python logic and tests to create an RNNCell for GRU.  Uses the preexisting GRU Unit Op code.

Reviewed By: salexspb

Differential Revision: D5364893

fbshipit-source-id: 2451d7ec8c2eacb8d8c9b7c893bfd21b65fb9d18
2017-07-10 17:52:25 -07:00
Robert Verkuil
2409c2e359 GRUUnit Op Backwards Pass
Summary:
Just an implementation of the forward pass of the GRU Unit Op, not the full RNNCell.
Functions were created to mimic LSTM implementation as closely as possible.
Backwards pass implementations are defined in GRU_unit_op.{h, cc}
assertGradientChecks call added to gru_cell_test.py

Reviewed By: salexspb

Differential Revision: D5364856

fbshipit-source-id: 09cff4478091827763b40cc331e4e0abf0ec258f
2017-07-10 17:52:24 -07:00
Robert Verkuil
279f3f095e Implemented Gated Recurrent Unit (GRU) c++ operator forward pass
Summary:
Just an implementation of the forward pass of the GRU Unit Op, not the full RNNCell.
Functions were created to mimic LSTM implementation as closely as possible.
Implementation defined in GRU_unit_op.{h, cc}
tests put in gru_cell_test.py, which import rnn_cell_test_util.py for sigmoid, tanh, and _prepare_rnn functions.

Reviewed By: jamesr66a

Differential Revision: D5363697

fbshipit-source-id: f9ba9fe0be01ffc868dd22027be8be4975b84998
2017-07-10 17:52:23 -07:00