QuantBatchMatmulV3TransposeFusionPass
Description
Fuses the Transpose/TransposeD nodes connected to the inputs x1 and x2 of QuantBatchMatmulV3 and transfers their information into the transpose_x1 and transpose_x2 attributes of QuantBatchMatmulV3.
x1 and x2 are independent of each other. The fusion can be triggered as long as a Transpose node appears on either input path.
Restrictions
None
Availability
Parent topic: Graph Fusion Patterns