save_compressed_retrain_model

Applicability

Product	Supported
Atlas 350 Accelerator Card	QAT INT8 quantization: √ Filter-level sparsity: √ 2:4 structured sparsity: x
Atlas A3 training product/Atlas A3 inference product	QAT INT8 quantization: √ Filter-level sparsity: √ 2:4 structured sparsity: √
Atlas A2 training product/Atlas A2 inference product	QAT INT8 quantization: √ Filter-level sparsity: √ 2:4 structured sparsity: √
Atlas 200I/500 A2 inference product	QAT INT8 quantization: √ Filter-level sparsity: √ 2:4 structured sparsity: √
Atlas inference product	QAT INT8 quantization: √ Filter-level sparsity: √ 2:4 structured sparsity: x
Atlas training product	QAT INT8 quantization: √ Filter-level sparsity: √ 2:4 structured sparsity: x

Note: For the products marked with x, no error is reported when the API is called, but no performance gains are obtained.

Description

Applies to static compression combination. Generates the resultant model of static compression combination from the retrained model.

Prototype

save_compressed_retrain_model(pb_model, outputs, record_file, save_path)

Parameters

Parameter	Input/Output	Description
pb_model	Input	.pb model of the inference mode with sparsity and quantization operators. A string.
outputs	Input	List of output operators in a graph. A list of strings.
record_file	Input	Path (including the file name) of the sparsity and quantization factor record file. A string.
save_path	Input	Path for storing the compressed model. Must include the prefix of the model name, for example, **./compressed_model/model***. A string.

Returns

None

Example

import amct_tensorflow as amct
amct.save_compressed_retrain_model(masked_pb_path, [operation_name_1, operation_name_2], './tmp/record.txt', './pb_model/final_model')

Flush file: compressed model, which can be fake-quantized in the TensorFlow environment and inferred on the AI processor.

Parent topic: Compression Combination APIs