diff tvii/dataset/gauss.py @ 87:9d5a5e9f5c3b

add kmeans + dataset
author Jeff Hammel <k0scist@gmail.com>
date Sun, 17 Dec 2017 14:05:57 -0800
parents
children
line wrap: on
line diff
--- /dev/null	Thu Jan 01 00:00:00 1970 +0000
+++ b/tvii/dataset/gauss.py	Sun Dec 17 14:05:57 2017 -0800
@@ -0,0 +1,46 @@
+#!/usr/bin/env python
+
+"""
+generate a Gaussian dataset
+"""
+
+import random
+import sys
+from .cli import DatasetGenerationParser
+
+
+class GaussianSampler(object):
+
+    def __init__(self, D, sigma=1., center=None):
+        assert D > -1
+        self.D = D
+        if center is None:
+            self.center = [0]*self.D
+        else:
+            assert len(center) == self.D
+            self.center = center
+        self.sigma = sigma
+
+    def __call__(self):
+        """yield one point of a Gaussian distribution"""
+
+        return [random.gauss(x, self.sigma)
+                for x in self.center]
+
+
+def main(args=sys.argv[1:]):
+    """CLI"""
+
+    # parse command line
+    parser = DatasetGenerationParser(description=__doc__)
+    options = parser.parse_args(args)
+
+    # point generator
+    generator = GaussianSampler(D=options.dimensions)
+    points = [generator() for _ in xrange(options.number)]
+
+    # output
+    parser.writer().writerows(points)
+
+if __name__ == '__main__':
+    main()