From d55f2b98822faa7d71f5fce2bfa980f8265e0610 Mon Sep 17 00:00:00 2001
From: Josh Rosen <joshrosen@eecs.berkeley.edu>
Date: Wed, 9 Jan 2013 21:21:23 -0800
Subject: [PATCH] Use take() instead of takeSample() in PySpark kmeans example.
This is a temporary change until we port takeSample().
---
python/examples/kmeans.py | 4 +++-
1 file changed, 3 insertions(+), 1 deletion(-)
diff --git a/python/examples/kmeans.py b/python/examples/kmeans.py
index ad2be21178..72cf9f88c6 100644
--- a/python/examples/kmeans.py
+++ b/python/examples/kmeans.py
@@ -33,7 +33,9 @@ if __name__ == "__main__":
K = int(sys.argv[3])
convergeDist = float(sys.argv[4])
- kPoints = data.takeSample(False, K, 34)
+ # TODO: change this after we port takeSample()
+ #kPoints = data.takeSample(False, K, 34)
+ kPoints = data.take(K)
tempDist = 1.0
while tempDist > convergeDist:
--
GitLab