SSH-Master-Thesis/ethan_data_processing_scripts

Using classifier.py

The train(classifications: int, data: list, results: list, testdata: list, testresults: list) function should be used.

Data and Testdata should be arrays of arrays of features, ex:

Suppose there are 3 features, each a float from 0 to 1. Data could be: [[.3, .2, .3], [.3, .3, .3], [.3, .4, .5]...]

Results should be ints in an array, each result accoring to the list of features it should represent the classification of.

Current state of nearestneighbors.py

Usage: nearestneighbors.py datafile.bin classificationsfile.bin testdatafile.bin -p/e

if -p, a tuple of the pickle dump of the test data array and their classifications are written if -e, an english copy of the printout is written

A command line utility that reads in FeatureVectors and runs a KNN classification on them.

Plan for classifier.py: Discuss data formatting at meeting, expand utility to include choice of classification and make more robust in general