experiments/STT-tensorflow
1
0
Fork 0
Code Issues Pull Requests Projects Releases Wiki Activity

Improve recurrent layers Call API mask documentation

Browse Source
This commit is contained in:
Roger Cheng 2020-11-25 14:08:28 -08:00 committed by Roger Cheng
parent a7eb0f4531
commit 9a74ca8405
2 changed files with 33 additions and 5 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

16
tensorflow/python/keras/layers/recurrent.py
View File

@ -274,7 +274,9 @@ class RNN(Layer):
Call arguments:
inputs: Input tensor.
mask: Binary tensor of shape `[batch_size, timesteps]` indicating whether
a given timestep should be masked.
a given timestep should be masked. An individual `True` entry indicates
that the corresponding timestep should be utilized, while a `False`
entry indicates that the corresponding timestep should be ignored.
training: Python boolean indicating whether the layer should behave in
training mode or in inference mode. This argument is passed to the cell
when calling it. This is for use with cells that use dropout.
@ -1491,7 +1493,9 @@ class SimpleRNN(RNN):
Call arguments:
inputs: A 3D tensor, with shape `[batch, timesteps, feature]`.
mask: Binary tensor of shape `[batch, timesteps]` indicating whether
a given timestep should be masked.
a given timestep should be masked. An individual `True` entry indicates
that the corresponding timestep should be utilized, while a `False` entry
indicates that the corresponding timestep should be ignored.
training: Python boolean indicating whether the layer should behave in
training mode or in inference mode. This argument is passed to the cell
when calling it. This is only relevant if `dropout` or
@ -2036,7 +2040,9 @@ class GRU(RNN):
Call arguments:
inputs: A 3D tensor.
mask: Binary tensor of shape `(samples, timesteps)` indicating whether
a given timestep should be masked.
a given timestep should be masked. An individual `True` entry indicates
that the corresponding timestep should be utilized, while a `False`
entry indicates that the corresponding timestep should be ignored.
training: Python boolean indicating whether the layer should behave in
training mode or in inference mode. This argument is passed to the cell
when calling it. This is only relevant if `dropout` or
@ -2710,7 +2716,9 @@ class LSTM(RNN):
Call arguments:
inputs: A 3D tensor.
mask: Binary tensor of shape `(samples, timesteps)` indicating whether
a given timestep should be masked.
a given timestep should be masked. An individual `True` entry indicates
that the corresponding timestep should be utilized, while a `False`
entry indicates that the corresponding timestep should be ignored.
training: Python boolean indicating whether the layer should behave in
training mode or in inference mode. This argument is passed to the cell
when calling it. This is only relevant if `dropout` or

22
tensorflow/python/keras/layers/recurrent_v2.py
View File

@ -335,6 +335,9 @@ class GRU(recurrent.DropoutRNNCellMixin, recurrent.GRU):
inputs: A 3D tensor, with shape `[batch, timesteps, feature]`.
mask: Binary tensor of shape `[samples, timesteps]` indicating whether
a given timestep should be masked (optional, defaults to `None`).
An individual `True` entry indicates that the corresponding timestep
should be utilized, while a `False` entry indicates that the
corresponding timestep should be ignored.
training: Python boolean indicating whether the layer should behave in
training mode or in inference mode. This argument is passed to the cell
when calling it. This is only relevant if `dropout` or
@ -571,7 +574,9 @@ def standard_gru(inputs, init_h, kernel, recurrent_kernel, bias, mask,
bias: Weights for cell kernel bias and recurrent bias. The bias contains the
combined input_bias and recurrent_bias.
mask: Binary tensor of shape `(samples, timesteps)` indicating whether
a given timestep should be masked.
a given timestep should be masked. An individual `True` entry indicates
that the corresponding timestep should be utilized, while a `False` entry
indicates that the corresponding timestep should be ignored.
time_major: Boolean, whether the inputs are in the format of
[time, batch, feature] or [batch, time, feature].
go_backwards: Boolean (default False). If True, process the input sequence
@ -741,6 +746,9 @@ def gru_with_backend_selection(inputs, init_h, kernel, recurrent_kernel, bias,
bias: Weights for cell kernel bias and recurrent bias. Only recurrent bias
is used in this case.
mask: Boolean tensor for mask out the steps within sequence.
An individual `True` entry indicates that the corresponding timestep
should be utilized, while a `False` entry indicates that the corresponding
timestep should be ignored.
time_major: Boolean, whether the inputs are in the format of
[time, batch, feature] or [batch, time, feature].
go_backwards: Boolean (default False). If True, process the input sequence
@ -1063,6 +1071,9 @@ class LSTM(recurrent.DropoutRNNCellMixin, recurrent.LSTM):
inputs: A 3D tensor with shape `[batch, timesteps, feature]`.
mask: Binary tensor of shape `[batch, timesteps]` indicating whether
a given timestep should be masked (optional, defaults to `None`).
An individual `True` entry indicates that the corresponding timestep
should be utilized, while a `False` entry indicates that the corresponding
timestep should be ignored.
training: Python boolean indicating whether the layer should behave in
training mode or in inference mode. This argument is passed to the cell
when calling it. This is only relevant if `dropout` or
@ -1349,6 +1360,9 @@ def standard_lstm(inputs, init_h, init_c, kernel, recurrent_kernel, bias,
bias: weights for cell kernel bias and recurrent bias. Only recurrent bias
is used in this case.
mask: Boolean tensor for mask out the steps within sequence.
An individual `True` entry indicates that the corresponding timestep
should be utilized, while a `False` entry indicates that the corresponding
timestep should be ignored.
time_major: boolean, whether the inputs are in the format of
[time, batch, feature] or [batch, time, feature].
go_backwards: Boolean (default False). If True, process the input sequence
@ -1421,6 +1435,9 @@ def gpu_lstm(inputs, init_h, init_c, kernel, recurrent_kernel, bias, mask,
bias: Weights for cell kernel bias and recurrent bias. Only recurrent bias
is used in this case.
mask: Boolean tensor for mask out the steps within sequence.
An individual `True` entry indicates that the corresponding timestep
should be utilized, while a `False` entry indicates that the corresponding
timestep should be ignored.
time_major: Boolean, whether the inputs are in the format of [time, batch,
feature] or [batch, time, feature].
go_backwards: Boolean (default False). If True, process the input sequence
@ -1551,6 +1568,9 @@ def lstm_with_backend_selection(inputs, init_h, init_c, kernel,
bias: Weights for cell kernel bias and recurrent bias. Only recurrent bias
is used in this case.
mask: Boolean tensor for mask out the steps within sequence.
An individual `True` entry indicates that the corresponding timestep
should be utilized, while a `False` entry indicates that the corresponding
timestep should be ignored.
time_major: Boolean, whether the inputs are in the format of
[time, batch, feature] or [batch, time, feature].
go_backwards: Boolean (default False). If True, process the input sequence