Operator Reference
serialize_class_knn (Operator)
serialize_class_knn
— Serialize a k-NN classifier.
Signature
serialize_class_knn( : : KNNHandle : SerializedItemHandle)
Description
serialize_class_knn
serializes a k-NN classifier (k-NN) and
its stored training samples (see fwrite_serialized_item
for an
introduction of the basic principle of serialization). The same data that
is written in a file by write_class_knn
is converted to a
serialized item. The k-NN classifier is defined by the handle
KNNHandle
. The serialized k-NN classifier is returned by the
handle SerializedItemHandle
and can be deserialized by
deserialize_class_knn
.
Execution Information
- Multithreading type: reentrant (runs in parallel with non-exclusive operators).
- Multithreading scope: global (may be called from any thread).
- Processed without parallelization.
Parameters
KNNHandle
(input_control) class_knn →
(handle)
Handle of the k-NN classifier.
SerializedItemHandle
(output_control) serialized_item →
(handle)
Handle of the serialized item.
Result
If the parameters are valid, the operator
serialize_class_knn
returns the value 2 (
H_MSG_TRUE)
. If
necessary, an exception is raised.
Possible Predecessors
train_class_knn
,
read_class_knn
Possible Successors
fwrite_serialized_item
,
send_serialized_item
,
deserialize_class_knn
See also
create_class_knn
,
read_class_knn
,
deserialize_class_knn
References
Marius Muja, David G. Lowe: “Fast Approximate Nearest Neighbors with Automatic Algorithm Configuration”; International Conference on Computer Vision Theory and Applications (VISAPP 09); 2009.
Module
Foundation