`neural_compressor.adaptor.tf_utils.quantize_graph.qdq.fuse_qdq_matmul`

Quantize MatMul/BatchMatMul/BatchMatMulV2.

Module Contents

Classes

FuseNodeStartWithMatmul

Quantize MatMul/BatchMatMul/BatchMatMulV2 and apply the fusion.

class neural_compressor.adaptor.tf_utils.quantize_graph.qdq.fuse_qdq_matmul.FuseNodeStartWithMatmul(**kwargs)[source]: Quantize MatMul/BatchMatMul/BatchMatMulV2 and apply the fusion.