Download Latest Version t10k-images-idx3-ubyte.gz (1.6 MB)
Email in envelope

Get an email when there's a new version of Sampling-k-means

Name Modified Size InfoDownloads / Week
Parent folder
train-labels-idx1-ubyte.gz 2013-07-29 28.9 kB
yeast.txt 2013-07-29 52.6 kB
train-images-idx3-ubyte.gz 2013-07-29 9.9 MB
test.txt 2013-07-29 145 Bytes
t10k-labels-idx1-ubyte.gz 2013-07-29 4.5 kB
t10k-images-idx3-ubyte.gz 2013-07-29 1.6 MB
t10k-images-idx3-ubyte 2013-07-29 7.8 MB
s1.txt 2013-07-29 105.0 kB
kddcup.data_10_percent.gz 2013-07-29 2.1 MB
kddcup.data_10_percent 2013-07-29 74.9 MB
kdd.txt 2013-07-29 64.2 MB
data8 2013-07-29 784.0 kB
Totals: 12 Items   161.6 MB 0
Here is a brief description of the files:

PTAS_Heuristic.m: This is the implementation of our Heuristic Algorithm.
		  This program uses the following subroutines:
	1. evalCost.m: Computes the cost of a given solution.
	2. pickNewCenter.m: Picks a new center with by computing the centroid.
	3. pickSubsets.m: This picks the subsets that are supposed to be considered
			  in the algorithm.
	4. samplePintsUsingD2Sampling.m: This samples the points.
	5. updateNearestDistances.m: This function updates the closest center for each 				point after a new center has been chosen.

readKDD.m: Reads the KDD data given in the folder "Data"
readMNIST.m Reads the MNIST data given in the folder "Data"

Other matlab files: 

There are some other matlab files that were used to either plot, or experiment, or generate data. If not needed you may delete this.



Source: README.txt, updated 2013-07-29