KNN Classifier Job Plugin¶
Plugin Name: KNNClassifierJobPlugin
This plugin runs a K-Nearest Neighbors Classification using the SciKit Learn library.
More information on this Classifier is available at http://scikit-learn.org/stable/modules/generated/sklearn.neighbors.KNeighborsClassifier.html
Note
Tags in the Example and Source TagSets SHOULD be the EXACT same length. Currently, features will be padded to the length of the longest Tag.
Launching a Job¶
Jobs are launched via the Plugin Job interface, by selecting “Launch Plugin Jobs” on the Project Files page.
Job Settings¶
Setting Name | Description |
---|---|
name | Not used internally. This is meant as a book-keeping item for later reference by the user. |
n_neighbors | Number of neighbors to use |
weights | weight function used in prediction.
|
algorithm | Algorithm used to compute the nearest neighbors.
|
exampleTagSet | The TagSet from which the Class Examples are loaded. |
sourceTagSet | Tags which will be classified. |
destinationTagSet | TagSet into which new Tags are created - these are copied from the sourceTagSet, with a new tagClass corresponding to the exampleTagSet examples. |
Spectra Details | |
numFrequencyBands | The number of bands (sections) in the spectra creation between minFrequency and maxFrequency. |
numTimeFramesPerSecond | The number of times per second in an audio file that a line of spectra is computed. |
dampingRatio | |
minFrequency | The minimum frequency used for the spectra creation. |
maxFrequency | The maximum frequency used for the spectra creation. |
Priority¶
This will set the Job priority to ‘0’, the default priority.