Get a sharded variable concatenated into one tensor.

Linear map: sum_i(args[i] * W[i]), where W[i] is a variable. Args: args: a 2D Tensor or a list of 2D, batch x n, Tensors. output_si

Get a list of sharded variables with the given dtype.

Returns initializer op that only runs for uninitialized ops.

Returns initializer op that only runs for uninitialized ops.

Function to build multilayer PLSTM :param cells: :param inputs: :param lens: 2D tensor, length of the sequences in the batch (for syn

Function to build multilayer PLSTM :param cells: :param inputs: :param lens: 2D tensor, length of the sequences in the batch (for syn

Returns an initializer that generates tensors with an exponential distribution. Args: minval: A python scalar or a scalar tensor. Lower bou

Returns an initializer that generates tensors with an exponential distribution. Args: minval: A python scalar or a scalar tensor. Lower bou

Run one step of LSTM. Args: inputs: input Tensor, 2D, batch x num_units. state: if `state_is_tuple` is False, this mu

Run one step of LSTM. Args: inputs: input Tensor, 2D, batch x num_units. state: if `state_is_tuple` is False, this must b

Initialize the parameters for an PLSTM cell. Args: num_units: int, The number of units in the LSTM cell input_size: Depre

Initializes all uninitialized variables in correct order. Initializers are only run for uninitialized variables, so it's safe to run this multiple

Initializes all uninitialized variables in correct order. Initializers are only run for uninitialized variables, so it's safe to run this multiple