prediction_vector_sink

Module: missions.nodes.sink.prediction_vector_sink

Collect prediction vectors

Inheritance diagram for pySPACE.missions.nodes.sink.prediction_vector_sink:

PredictionVectorSinkNode

class pySPACE.missions.nodes.sink.prediction_vector_sink.PredictionVectorSinkNode(**kwargs)

Bases: pySPACE.missions.nodes.base_node.BaseNode

Collect all PredictionVectorDataset elements that are passed through it in a collection of type prediction_vector.

Note

The code is heavily based on its counterpart for FeatureVector elements that can be found in FeatureVectorSinkNode

Exemplary Call

-
    node: PredictionVectorSink

Input:

PredictionVector

Output:

PredictionVectorDataset

Author:

Andrei Ignat (andrei_cristian.ignat@dfki.de)

Created:

2014/10/15

POSSIBLE NODE NAMES:  

• PredictionVectorSinkNode
• PredictionVectorSink

POSSIBLE INPUT TYPES:  

• PredictionVector

Class Components Summary

_create_result_sets()	Instantiate the PredictionVectorDataset
_train(data, label)
get_result_dataset()	Return the result dataset
input_types
is_supervised()	Returns whether this node requires supervised training
is_trainable()	Returns whether this node is trainable.
process_current_split()	Compute the results of this sink node for the current split
reset()	Reset the state of the object to the clean state it had after its

input_types = ['PredictionVector']

__init__(**kwargs)

reset()

Reset the state of the object to the clean state it had after its initialization

is_trainable()

Returns whether this node is trainable.

Since we want to sink the training examples as well, this function wil return True

is_supervised()

Returns whether this node requires supervised training

_train(data, label)

_create_result_sets()

Instantiate the PredictionVectorDataset

process_current_split()

Compute the results of this sink node for the current split of the data into train and test data

get_result_dataset()

Return the result dataset