Return the parameters of a multilayer perceptron.
get_params_class_mlp returns the parameters of a multilayer perceptron (MLP) that were specified when the MLP was created with create_class_mlp. This is particularly useful if the MLP was read with read_class_mlp. The output of get_params_class_mlp can, for example, be used to check whether the feature vectors and, if necessary, the target data to be used with the MLP have the correct lengths. For a description of the parameters, see create_class_mlp.
|
MLPHandle (input_control) |
class_mlp -> integer |
| MLP handle. | |
|
NumInput (output_control) |
integer -> integer |
| Number of input variables (features) of the MLP. | |
|
NumHidden (output_control) |
integer -> integer |
| Number of hidden units of the MLP. | |
|
NumOutput (output_control) |
integer -> integer |
| Number of output variables (classes) of the MLP. | |
|
OutputFunction (output_control) |
string -> string |
| Type of the activation function in the output layer of the MLP. | |
|
Preprocessing (output_control) |
string -> string |
| Type of preprocessing used to transform the feature vectors. | |
|
NumComponents (output_control) |
integer -> integer |
| Preprocessing parameter: Number of transformed features. | |
If the parameters are valid, the operator get_params_class_mlp returns the value 2 (H_MSG_TRUE). If necessary an exception handling is raised.
get_params_class_mlp is reentrant and processed without parallelization.
create_class_mlp, read_class_mlp
add_sample_class_mlp, train_class_mlp
evaluate_class_mlp, classify_class_mlp
Foundation
