Read a training data set from a file.
The training examples are accessible with the key SampKey by calling procedures clear_sampset and learn_sampset_box. You may edit the file using an editor. Every row contains an array of attributes with corresponding class. An example for a format might be:
(1.0, 25.3, * , 17 | 3)
This row specifies an array of attributes which belong to class 3. In this
array the third attribute is unknown. Attributes upwards 5 are supposed to be
unknown, too.
You may insert comments like /* .. */ in any place.
|
FileName (input_control) |
filename -> string |
| Filename of the data set to train. | |
| Default value: 'sampset1' | |
|
SampKey (output_control) |
feature_set -> integer |
| Identification of the data set to train. | |
read_sampset returns 2 (H_MSG_TRUE). An exception handling is raised if it is not possible to open the file or it contains syntax errors or there is not enough memory.
read_sampset is local and processed completely exclusively without parallelization.
test_sampset_box, enquire_class_box, write_class_box, close_class_box, clear_sampset
test_sampset_box, clear_sampset, learn_sampset_box
Foundation
