get_regularization_params_class_mlpT_get_regularization_params_class_mlpGetRegularizationParamsClassMlpGetRegularizationParamsClassMlpget_regularization_params_class_mlp (算子)
名称
get_regularization_params_class_mlpT_get_regularization_params_class_mlpGetRegularizationParamsClassMlpGetRegularizationParamsClassMlpget_regularization_params_class_mlp — 返回多层感知机的正则化参数。
签名
描述
get_regularization_params_class_mlpget_regularization_params_class_mlpGetRegularizationParamsClassMlpGetRegularizationParamsClassMlpGetRegularizationParamsClassMlpget_regularization_params_class_mlp returns the
regularization parameters of a multilayer perceptron (MLP) that were
specified with set_regularization_params_class_mlpset_regularization_params_class_mlpSetRegularizationParamsClassMlpSetRegularizationParamsClassMlpSetRegularizationParamsClassMlpset_regularization_params_class_mlp。Furthermore, get_regularization_params_class_mlpget_regularization_params_class_mlpGetRegularizationParamsClassMlpGetRegularizationParamsClassMlpGetRegularizationParamsClassMlpget_regularization_params_class_mlp returns the
parameters that were determined by an automatic determination of the
regularization parameters. For a description of the parameters, see
set_regularization_params_class_mlpset_regularization_params_class_mlpSetRegularizationParamsClassMlpSetRegularizationParamsClassMlpSetRegularizationParamsClassMlpset_regularization_params_class_mlp。
执行信息
- 多线程类型：可重入（与非独占算子并行运行）。
- 多线程作用域：全局（可从任何线程调用）。
- 未采用并行化处理。
参数
MLPHandleMLPHandleMLPHandleMLPHandleMLPHandlemlphandle (输入控制) class_mlp → HClassMlp, HTupleHHandleHTupleHtuple (handle) (IntPtr) (HHandle) (handle)
MLP 句柄。
GenParamNameGenParamNameGenParamNameGenParamNamegenParamNamegen_param_name (输入控制) string → HTuplestrHTupleHtuple (string) (string) (HString) (char*)
Name of the regularization parameter to return.
默认值：
'weight_prior'
"weight_prior"
"weight_prior"
"weight_prior"
"weight_prior"
"weight_prior"
值列表：
'fraction_well_determined_params'"fraction_well_determined_params""fraction_well_determined_params""fraction_well_determined_params""fraction_well_determined_params""fraction_well_determined_params", 'noise_prior'"noise_prior""noise_prior""noise_prior""noise_prior""noise_prior", 'num_inner_iterations'"num_inner_iterations""num_inner_iterations""num_inner_iterations""num_inner_iterations""num_inner_iterations", 'num_outer_iterations'"num_outer_iterations""num_outer_iterations""num_outer_iterations""num_outer_iterations""num_outer_iterations", 'num_well_determined_params'"num_well_determined_params""num_well_determined_params""num_well_determined_params""num_well_determined_params""num_well_determined_params", 'weight_prior'"weight_prior""weight_prior""weight_prior""weight_prior""weight_prior"
GenParamValueGenParamValueGenParamValueGenParamValuegenParamValuegen_param_value (输出控制) number(-array) → HTupleSequence[Union[float, int]]HTupleHtuple (real / integer) (double / int / long) (double / Hlong) (double / Hlong)
Value of the regularization parameter.
结果
如果参数有效，算子
get_regularization_params_class_mlpget_regularization_params_class_mlpGetRegularizationParamsClassMlpGetRegularizationParamsClassMlpGetRegularizationParamsClassMlpget_regularization_params_class_mlp 返回值 2 ( H_MSG_TRUE )。如有必要，则抛出异常。
可能的前趋
set_regularization_params_class_mlpset_regularization_params_class_mlpSetRegularizationParamsClassMlpSetRegularizationParamsClassMlpSetRegularizationParamsClassMlpset_regularization_params_class_mlp,
read_class_mlpread_class_mlpReadClassMlpReadClassMlpReadClassMlpread_class_mlp
可能的后继
train_class_mlptrain_class_mlpTrainClassMlpTrainClassMlpTrainClassMlptrain_class_mlp
模块
基础