--feature

Description

This option applies only to TensorFlow models.

Enables model compression features.

See Also

None

Arguments

Arguments:

• ptq (default): enables only PTQ.
• approximate: enables only operator approximation calibration.
• approximate|ptq: enables both operator approximation calibration and PTQ.

Format: If both operator approximation calibration and PTQ are enabled, the argument must be enclosed in double quotation marks, for example, "approximate|ptq".

Suggestions and Benefits

None

Example

--feature="approximate|ptq"