|
pulsatrix 1.1.0
|
Optimizer settings for the parameters selects picks.
More...
#include <param_groups.hpp>
Public Attributes | |
| std::string | name |
| For error messages and logs. | |
| ParamSelector | selects |
| float | learning_rate |
| float | weight_decay = 0.0f |
| L2 penalty: the step uses grad + weight_decay * value, as PyTorch's SGD and Adam do. | |
Optimizer settings for the parameters selects picks.
| float pulsatrix::ParamGroup::learning_rate |
| std::string pulsatrix::ParamGroup::name |
For error messages and logs.
| ParamSelector pulsatrix::ParamGroup::selects |
| float pulsatrix::ParamGroup::weight_decay = 0.0f |
L2 penalty: the step uses grad + weight_decay * value, as PyTorch's SGD and Adam do.