From: Yoav F. <yf...@us...> - 2008-09-15 15:24:48
|
Update of /cvsroot/jboost/jboost/scripts In directory sc8-pr-cvs17.sourceforge.net:/tmp/cvs-serv16557/scripts Modified Files: VisualizeScores.py Log Message: VisualizeScores.py can now be used to select a range of scores at a particular iteration and output the indices of the examples whose score is in that range into a file. Index: VisualizeScores.py =================================================================== RCS file: /cvsroot/jboost/jboost/scripts/VisualizeScores.py,v retrieving revision 1.6 retrieving revision 1.7 diff -C2 -d -r1.6 -r1.7 *** VisualizeScores.py 11 Sep 2008 21:25:45 -0000 1.6 --- VisualizeScores.py 15 Sep 2008 22:24:45 -0000 1.7 *************** *** 21,25 **** scoreline_p = re.compile('([-+\d.]+):\s+') #([+-1]+):') ! info_path = "/Users/yoavfreund/projects/jboost/demo/cvdata-09-10-13-54-55/ADD_ALL/" testfiles = glob.glob(info_path+"trial*.test.boosting.info") --- 21,25 ---- scoreline_p = re.compile('([-+\d.]+):\s+') #([+-1]+):') ! info_path = "/Users/yoavfreund/projects/jboost/demo/forVisualize/run1/" testfiles = glob.glob(info_path+"trial*.test.boosting.info") *************** *** 55,61 **** --- 55,63 ---- d=DataSet(iterList) + d.setOutputFilename(info_path+"/selectedExamples.txt") # read the data into the Dataset data structures. for filename in testfiles: + print filename infile = open(filename,'r') iterList=[]; |