neural_compressor.compression.pruner.regs

Regularizer.

Module Contents

Classes

BaseReg	Regularizer.
GroupLasso	Regularizer.

Functions

register_reg(name)	Register a regularizator to the registry.
get_reg_type(config)	Obtain the regularizer type.
get_reg(config, modules, pattern)	Get registered regularizator class.

neural_compressor.compression.pruner.regs.register_reg(name)

Register a regularizator to the registry.

Parameters:

name – A string that defines the scheduler type.

Returns:

The class of register.

Return type:

cls

neural_compressor.compression.pruner.regs.get_reg_type(config)

Obtain the regularizer type.

Parameters:

config – A config dict object that includes information of the regularizer.

neural_compressor.compression.pruner.regs.get_reg(config, modules, pattern)

Get registered regularizator class.

Parameters:

• config – A config dict object that includes information of the regularizer.

• modules – A dict {“module_name”: Tensor} that stores the pruning modules’ weights.

• pattern – A config dict object that includes information of the pattern.

class neural_compressor.compression.pruner.regs.BaseReg(config: dict, modules: dict, pattern: neural_compressor.compression.pruner.patterns.base.PytorchBasePattern)

Regularizer.

The class that performs regularization.

Parameters:

• modules – A dict {“module_name”: Tensor} that stores the pruning modules’ weights.

• config – A config dict object that includes information of the regularizer.

• pattern – A config dict object that includes information of the pattern.

class neural_compressor.compression.pruner.regs.GroupLasso(config: dict, modules: dict, pattern: neural_compressor.compression.pruner.patterns.base.PytorchBasePattern, coeff)

Regularizer.

A regularizer class derived from BaseReg. In this class, the Group-lasso regularization will be performed. Group-lasso is a variable-selection and regularization method.

Parameters:

• modules – A dict {“module_name”: Tensor} that stores the pruning modules’ weights.

• config – A config dict object that includes information of the regularizer.

• pattern – A config dict object that includes information of the pattern.

reg_terms

A dict {“module_name”: Tensor} of regularization terms.

alpha

A float representing the coefficient related to group lasso.