`neural_compressor.tensorflow.quantization.utils.graph_rewriter.int8.meta_op_optimizer`

Meta OP Graph Rewriter.

Module Contents

Fuse the pattern like Dequantize + MetaOp + Quantize into MetaOp(set its type to int8).

class neural_compressor.tensorflow.quantization.utils.graph_rewriter.int8.meta_op_optimizer.MetaInfoChangingMemOpOptimizer(model)[source]

Fuse the pattern like Dequantize + MetaOp + Quantize into MetaOp(set its type to int8).

With such changes, the Quantize and Dequantize OP will removed for better performance.