Template Class OptimizerCloneableOptions#
Defined in File optimizer.h
Nested Relationships#
Nested Types#
Template Struct OptimizerCloneableOptions::_has_history_size
Template Struct OptimizerCloneableOptions::_has_initial_accumulator_value
Template Struct OptimizerCloneableOptions::_has_line_search_fn
Template Struct OptimizerCloneableOptions::_has_tolerance_change
Template Struct OptimizerCloneableOptions::_has_tolerance_grad
Template Struct OptimizerCloneableOptions::_has_weight_decay
Inheritance Relationships#
Base Type#
public torch::optim::OptimizerOptions(Class OptimizerOptions)
Class Documentation#
-
template<typename Derived>
class OptimizerCloneableOptions : public torch::optim::OptimizerOptions# OptimizerCloneableOptions provides parameter group inheritance functionality for PyTorch C++ optimizer options.
When creating parameter groups with partial options (e.g., AdamOptions().weight_decay(0.1)), fields not explicitly set by the user inherit from the optimizer’s default values, while explicitly set fields are preserved.
This enables Python-like behavior in C++:
// Python equivalent: // optimizer = Adam([{'params': params1, 'weight_decay': 0.1}], lr=0.01) // Result: weight_decay=0.1 preserved, lr=0.01 inherited AdamOptions defaults; defaults.lr(0.01).weight_decay(0.05); std::vector<OptimizerParamGroup> groups; groups.emplace_back(params1, std::make_unique<AdamOptions>( AdamOptions().weight_decay(0.1))); // Only weight_decay specified Adam optimizer(groups, defaults); // Result: group inherits lr=0.01, preserves weight_decay=0.1
Implementation: Uses SFINAE-based field detection and constructor-default comparison to distinguish explicitly set fields from default values. Fields that match constructor defaults are inherited; others are preserved.
