ScopeDynamicRNNPass

Description

Fuses a scope containing the following operators into a DynamicRNN operator: The scope contains a while subscope and does not contain Transpose operators. The while subscope contains a multiple of 4 BiasAdd operators, a multiple of 2 Tanh operators, eight MatMul operators, and a Split operator.

Scope Details

See the following scope example.

The while subscope contains a multiple of 4 BiasAdd operators, a multiple of 2 Tanh operators, eight MatMul operators, and a Split operator.

Result Operator Prototype

DynamicRNN. For details, see Operator Acceleration Library API Reference.

Fusion Mapping

The input of TensorArrayUnstack/TensorArrayScatter/TensorArrayScatterV3 is used as the first input of the fused operator.

The input of while/split/ReadVariableOp/Enter is used as the second input of the fused operator.

The input of while/split_1/ReadVariableOp/Enter is used as the third input of the fused operator.

The output of TensorArrayStack/TensorArrayGatherV3 is used as the first output of the fused operator.

Applicable Network

TacoTron

Fusion Pattern Type

Non-general fusion pattern