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

Convert fp32 op to int8 and fuse the pattern.

Module Contents

Classes

QuantizeGraphForIntel

Quantize the graph.

class neural_compressor.tensorflow.quantization.utils.quantize_graph.quantize_graph_for_intel_cpu.QuantizeGraphForIntel(input_graph, input_node_names, output_node_names, op_wise_config, op_wise_sequences, device, fake_quant=False, new_api=False, performance_only=False, itex_mode=False)[source]: Quantize the graph.