`neural_compressor.tensorflow.quantization.utils.quantize_graph.quantize_graph_bn`

Quantize FusedBatchNormV3 to int8 op.

Module Contents

Classes

FuseNodeStartWithFusedBatchNormV3

Quantize FusedBatchNormV3 to int8 op and apply the fusion.

class neural_compressor.tensorflow.quantization.utils.quantize_graph.quantize_graph_bn.FuseNodeStartWithFusedBatchNormV3(**kwargs)[source]: Quantize FusedBatchNormV3 to int8 op and apply the fusion.