[mlpack-svn] r15299 - mlpack/conf/jenkins-conf/benchmark/methods/mlpack

fastlab-svn at coffeetalk-1.cc.gatech.edu fastlab-svn at coffeetalk-1.cc.gatech.edu
Mon Jun 24 16:34:55 EDT 2013


Author: marcus
Date: 2013-06-24 16:34:55 -0400 (Mon, 24 Jun 2013)
New Revision: 15299

Added:
   mlpack/conf/jenkins-conf/benchmark/methods/mlpack/allkfn.py
Log:
Add script to benchmark the All K-Furthest-Neighbors method.

Added: mlpack/conf/jenkins-conf/benchmark/methods/mlpack/allkfn.py
===================================================================
--- mlpack/conf/jenkins-conf/benchmark/methods/mlpack/allkfn.py	                        (rev 0)
+++ mlpack/conf/jenkins-conf/benchmark/methods/mlpack/allkfn.py	2013-06-24 20:34:55 UTC (rev 15299)
@@ -0,0 +1,125 @@
+'''
+  @file allkfn.py
+  @author Marcus Edel
+
+  Class to benchmark the mlpack All K-Furthest-Neighbors method.
+'''
+
+import os
+import sys
+import inspect
+
+# Import the util path, this method even works if the path contains
+# symlinks to modules.
+cmd_subfolder = os.path.realpath(os.path.abspath(os.path.join(
+	os.path.split(inspect.getfile(inspect.currentframe()))[0], '../../util')))
+if cmd_subfolder not in sys.path:
+	sys.path.insert(0, cmd_subfolder)
+
+from log import *
+
+import shlex
+import subprocess
+import re
+import collections
+
+class ALLKFN(object):
+
+	# Create the All K-Furthest-Neighbors instance, 
+	# show some informations and return the instance.
+	def __init__(self, dataset, path='/usr/local/bin/', verbose=True): 
+		self.verbose = verbose
+		self.dataset = dataset
+		self.path = path
+
+		# Get description from executable.
+		cmd = shlex.split(self.path + "allkfn -h")
+		s = subprocess.check_output(cmd, stderr=subprocess.STDOUT, shell=False)	
+
+		# Use regular expression pattern to get the description.
+		pattern = re.compile(r"""(.*?)Required.*?options:""", 
+				re.VERBOSE|re.MULTILINE|re.DOTALL)
+		
+		match = pattern.match(s)
+		if not match:
+			Log.Warn("Can't parse description", self.verbose)
+			description = ''
+		else:
+			description = match.group(1)
+		
+		# Show method informations.
+		# Log.Notice(description)
+		# Log.Notice('\n')
+
+	# Remove created files.
+	def __del__(self):		
+		Log.Info('Clean up.', self.verbose)
+		filelist = ['gmon.out', 'distances.csv', 'neighbors.csv']
+		for f in filelist:
+			if os.path.isfile(f):
+				os.remove(f)				
+
+	# Perform All K-Furthest-Neighbors and return
+	# the elapsed time.
+	def RunMethod(self, options):
+		Log.Info('Perform ALLKFN.', self.verbose)
+
+		# If the dataset contains two files then the second file is the query
+		# file. In this case we add this to the command line.
+		if len(self.dataset) == 2:
+			cmd = shlex.split(self.path + "allkfn -r " + self.dataset[0] + 
+				" -q " + self.dataset[1] + 
+				" -v -n neighbors.csv -d distances.csv" + options)
+		else:
+			cmd = shlex.split(self.path + "allkfn -r " + self.dataset + 
+				" -v -n neighbors.csv -d distances.csv " + options)		
+
+
+		# Run command with the nessecary arguments and return its output as
+		# a byte string. We have untrusted input so we disables all shell 
+		# based features.
+		s = subprocess.check_output(cmd, stderr=subprocess.STDOUT, shell=False)	
+
+		# Return the elapsed time.
+		timer = self.parseTimer(s)
+		if not timer:
+			Log.Fatal("Can't parse the timer", self.verbose)
+			return 0
+		else:
+			time = self.GetTime(timer)
+			Log.Info(('total time: %fs' % (time)), self.verbose)
+
+			return time
+
+	# Parse the timer data.
+	def parseTimer(self, data):
+		# Compile the regular expression pattern into a regular expression object
+		# to parse the timer data.
+		pattern = re.compile(r"""
+							.*?loading_data: (?P<loading_data>.*?)s.*?
+							.*?saving_data: (?P<saving_data>.*?)s.*?
+							.*?total_time: (?P<total_time>.*?)s.*?
+							""", re.VERBOSE|re.MULTILINE|re.DOTALL)
+		
+		match = pattern.match(data)
+
+		if not match:
+			print "Can't parse the data: wrong format"
+			return False
+		else:
+			# Create a namedtuple and return the timer data.
+			timer = collections.namedtuple('timer', ['loading_data', 
+				'saving_data', 'total_time'])
+			if match.group("loading_data").count(".") == 1:
+				return timer(float(match.group("loading_data")),
+						 	float(match.group("saving_data")),
+						 	float(match.group("total_time")))
+			else:
+				return timer(float(match.group("loading_data").replace(",", ".")),
+						 	float(match.group("saving_data").replace(",", ".")),
+						 	float(match.group("total_time").replace(",", ".")))
+
+	# Return the elapsed time.
+	def GetTime(self, timer):
+		time = timer.total_time - timer.loading_data - timer.saving_data
+		return time
\ No newline at end of file




More information about the mlpack-svn mailing list