View source on GitHub |
Helper class that converts between params of Cudnn and TF GRU.
tf.contrib.cudnn_rnn.CudnnParamsFormatConverterGRU(
num_layers, num_units, input_size, num_proj=None,
input_mode=CUDNN_INPUT_LINEAR_MODE, direction=CUDNN_RNN_UNIDIRECTION
)
Args | |
---|---|
num_layers
|
the number of layers for the RNN model. |
num_units
|
the number of units within the RNN model. |
input_size
|
the size of the input, it could be different from the num_units. |
num_proj
|
The output dimensionality for the projection matrices. If None or 0, no projection is performed. |
input_mode
|
indicate whether there is a linear projection between the input and the actual computation before the first layer. It could be one of 'linear_input', 'skip_input' or 'auto_select'. * 'linear_input' (default) always applies a linear projection of input onto RNN hidden state. (standard RNN behavior). * 'skip_input' is only allowed when input_size == num_units; * 'auto_select' implies 'skip_input' when input_size == num_units; otherwise, it implies 'linear_input'. |
direction
|
the direction model that the model operates. Could be either 'unidirectional' or 'bidirectional' |
Methods
opaque_to_tf_canonical
opaque_to_tf_canonical(
opaque_param
)
Converts cudnn opaque param to tf canonical weights.
tf_canonical_to_opaque
tf_canonical_to_opaque(
tf_canonicals, weights_proj=None
)
Converts tf canonical weights to cudnn opaque param.