aboutsummaryrefslogtreecommitdiffhomepage
path: root/tensorflow/contrib/lite/kernels/bidirectional_sequence_lstm.cc
Commit message (Collapse)AuthorAge
* Merge the different LSTM EvalFloat/EvalHybrid calls into a single file.Gravatar A. Unique TensorFlower2018-10-05
| | | | PiperOrigin-RevId: 215870962
* Add the option of merging bidirectional RNN and LSTM outputs into a single ↵Gravatar A. Unique TensorFlower2018-10-03
| | | | | | | | output tensor. This is useful if the output of both directions will be passed to the next layer as a single output, as it avoids adding a concatenation op, which can be expensive on mobile devices where memory movement is relatively expensive. PiperOrigin-RevId: 215616140
* Simplify the logic for running through a sequence forwards and backwards.Gravatar A. Unique TensorFlower2018-09-26
| | | | PiperOrigin-RevId: 214618170
* Modularize TF Lite interface definitions and reorganize file structureGravatar Pete Warden2018-09-07
| | | | PiperOrigin-RevId: 212064501
* Introduce auxiliary input and allow "cross-linking" in the bidirectional ↵Gravatar A. Unique TensorFlower2018-09-05
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | LSTM Op. This introduces a connection between forward and backward cells across subsequent layers when stacking bidirectional LSTM Ops on top of each other. In more detail: Previously, the Op had only one input that was fed into the layer in the following way: INPUT (INPUT_REVERSED) | | ----------------------- | FW_LSTM BW_LSTM | <----- bidi-LSTM cell (with one input / two outputs) ----------------------- | | FW_OUT BW_OUT Now, the Op can have an (optional) auxiliary input in the following way: AUX_INPUT (AUX_INPUT_REVERSED) | | INPUT | (INPUT_R'D.)| | | | | ------------------------- | \ / \ / | | FW_LSTM BW_LSTM | <----- bidi-LSMT cell (with 2 inputs / 2 outputs) ------------------------- | | FW_OUT BW_OUT When stacking these Ops, previously, only the following flow was allowed: Input / \ FW_LSTM1 BW_LSTM1 | | | | FW_LSTM2 BW_LSTM2 | | | | FW_LSTM3 BW_LSTM3 \ / Output With the introduction of an auxiliary input to the bidi-LSTM layer, the forward (FW_LSTMi) output of the ith layer is fed into as the input to the next layer (hence, inputs to both FW_LSTM{i+1} and BW_LSTM{i+1}) and the backward output is fed as the auxiliary inputs to both FW_LSTM{i+1} and BW_LSTM{i+1}). This way, the stacking can be changed to allow for the "cross-linking" between subsequent layer in the following way: Input / \ FW_LSTM1 BW_LSTM1 | \ / | | / \ | FW_LSTM2 BW_LSTM2 | \ / | | / \ | FW_LSTM3 BW_LSTM3 \ / Output PiperOrigin-RevId: 211659472
* Add support for quantized (hybrid) bidirectional sequential LSTM Op.Gravatar A. Unique TensorFlower2018-09-04
| | | | PiperOrigin-RevId: 211552101
* Update bidirectional sequential LSTM to support state API.Gravatar A. Unique TensorFlower2018-09-03
| | | | PiperOrigin-RevId: 211378028
* Remove unnecessary <unistd.h> includesGravatar Jared Duke2018-07-23
| | | | PiperOrigin-RevId: 205744600
* LSTM CHECK_OK on input tensor checks.Gravatar Alan Chiao2018-07-12
| | | | PiperOrigin-RevId: 204341675
* Making GetOptionalInput from kernel_util.h return a pointer to const data.Gravatar A. Unique TensorFlower2018-05-16
| | | | PiperOrigin-RevId: 196932028
* Making GetInput from kernel_util.h return a pointer to const data.Gravatar A. Unique TensorFlower2018-05-11
| | | | PiperOrigin-RevId: 196340200
* Internal change.Gravatar A. Unique TensorFlower2018-05-04
| | | | PiperOrigin-RevId: 195374319
* Internal change.Gravatar A. Unique TensorFlower2018-04-30
| | | | PiperOrigin-RevId: 194877173
* Factor out the LstmBatchStep for the various LSTM Ops.Gravatar A. Unique TensorFlower2018-02-28
| | | | PiperOrigin-RevId: 187370622
* Add bidirectional sequence LSTM to TFLite Ops.Gravatar A. Unique TensorFlower2018-02-21
PiperOrigin-RevId: 186497571
generated by cgit v1.2.3 (git 2.39.1) at 2024-06-21 03:38:49 +0000