ksu 0.5.1

Implementation of the KSU compression algorithm https://www.cs.bgu.ac.il/~karyeh/compression-arxiv.pdf

KSU Compression Algorithm Implementation

Algortihm 1 from Nearest-Neighbor Sample Compression: Efficiency, Consistency, Infinite Dimensions

Installation

• With pip: pip install ksu
• From source:
  • git clone --depth=1 https://github.com/nimroha/ksu_classifier.git
  • cd ksu_classifier
  • python setup.py install

Usage

Command Line

This package provides two command line tools: e-net and ksu:

• e-net constructs an epsilon net for a given epsilon
• ksu runs the full algorithm

Both provide the -h flag to specify the arguments, and both can save the result to the disk in numpy's .npz format

Code

This package provides a class KSU(Xs, Ys, metric, [gram, prune, logLevel, n_jobs])

Xs and Ys are the data points and their respective labels as numpy arrays

metric is either a callable to compute the metric or a string that names one of our provided metrics (print KSU.METRICS.keys() for the full list)

gram (optional, default=None) a precomputed gramian matrix, will be calculated if not provided.

prune (optional, default=False) a boolean indicating whether to prune the compressed set or not (Algorithm 2 from Near-optimal sample compression for nearest neighbors)

logLevel (optional, default='CRITICAL') a string indicating the logging level (set to 'INFO' or 'DEBUG' to get more information)

n_jobs (optional, default=1) an integer defining how many cpus to use (scipy logic), pass -1 to use all. For n_jobs below -1, (n_cpus + 1 + n_jobs) are used. Thus for n_jobs = -2, all CPUs but one are used.

KSU provides a method compressData([delta, minCompress, maxCompress, greedy, stride, logLevel, numProcs])

Which selects the subset with the lowest estimated error with confidence 1 - delta. Can take arguments:

delta (optional, default=0.1) confidence for error upper bound

minCompress (optional, default=0.05) minimal compression ratio

maxCompress (optional, default=0.1) maximum compression ratio

greedy (optional, default=True) whether to use greedy or hierarichal strategy for net construction

stride (optional, default=200) how many gammas to skip between each iteration (since similar gammas will produce similar nets)

logLevel (optional, default='CRITICAL') a string indicating the logging level (set to 'INFO' or 'DEBUG' to get more information)

numProcs (optional, default=1) number of processes to use

You can then run getClassifier() which returns a 1-NN Classifer (based on sklearn's K-NN) fitted to the compressed data.

Or, run getCompressedSet() to get the compressed data as a tuple of numpy arrays (compressedXs, compressedYs).

See scripts/ for example usage

Built-in metrics

['chebyshev', 'yule', 'sokalmichener', 'canberra', 'EarthMover', 'rogerstanimoto', 'matching', 'dice', 'EditDistance', 'braycurtis', 'russellrao', 'cosine', 'cityblock', 'l1', 'manhattan', 'sqeuclidean', 'jaccard', 'seuclidean', 'sokalsneath', 'kulsinski', 'minkowski', 'mahalanobis', 'euclidean', 'l2', 'hamming', 'correlation', 'wminkowski']