convert

Applicability

Product	Supported
Atlas 350 Accelerator Card	√
Atlas A3 training product/Atlas A3 inference product	√
Atlas A2 training product/Atlas A2 inference product	√
Atlas 200I/500 A2 inference product	x
Atlas inference product	x
Atlas training product	x

Description

Converts a quantization calibration model into a quantized deployable model.

Prototype

convert(model)

Parameters

Parameter	Input/Output	Description
model	Input/Output	Input: quantization calibration model Output: quantized deployable model Data type: torch.nn.Module

Parameter

Input/Output

Description

model

Input/Output

Input: quantization calibration model

Output: quantized deployable model

Data type: torch.nn.Module

Returns

Quantized deployable model

Restrictions

None

Example

# Call the preceding API to generate a quantization calibration model.
quantize(model, cfg)
# Call this API to convert the quantization calibration model into a quantized deployable model.
convert(model)

Parent topic: Torch Module-based Quantization APIs