From 4852a7f339e59f471a5915223ebcebe94c089ed1 Mon Sep 17 00:00:00 2001
From: Shreyas <seekshreyas@gmail.com>
Date: Fri, 25 Apr 2014 11:49:21 -0700
Subject: [PATCH] added the classifier changes

---
 README.md     |  19 ++++
 classifier.py | 285 ++++++++++++++++++++++++++++++++++++++++++++++++++
 data.csv      |   1 +
 3 files changed, 305 insertions(+)
 create mode 100644 classifier.py
 create mode 100644 data.csv

diff --git a/README.md b/README.md
index 965acb7..c9bf029 100644
--- a/README.md
+++ b/README.md
@@ -2,3 +2,22 @@ classifier-scaffold
 ===================
 
 My reusable code for using many different classifiers on a dataset.
+
+It takes some command line flags, like:
+
+```bash
+python appClassifierBenchmark.py --file="exports/appFeatures.csv" --classifier="all" --sample="split"
+```
+
+where,
+
+- `--classifier`: takes inputs of what classifier we want to run. By default it runs all and takes key values of other models to specifically only run that model.
+- `--sample`: takes inputs for whether we want to run the classifiers on all dataset, or equal splits. By default it takes all.
+- `--help`: provides some output for the above help, but frankly it still needs a little improvement.
+
+
+Save output from console to a file as:
+
+```bash
+$python appClassifierBenchmark.py --file="exports/appFeatures.csv" > <outputfile>
+```
diff --git a/classifier.py b/classifier.py
new file mode 100644
index 0000000..e58a529
--- /dev/null
+++ b/classifier.py
@@ -0,0 +1,285 @@
+#! /usr/bin/env python
+# -*- coding: UTF-8 -*-
+"""
+Benchmark different classifiers
+
+Classifiers tried:
+
+- Naive Bayes
+- Random Forest
+- SVM
+    - linear
+    - kernelized
+"""
+from __future__ import division
+import sys
+import pandas as pd
+import numpy as np
+from optparse import OptionParser
+from sklearn import metrics, preprocessing
+from sklearn import svm, naive_bayes, neighbors, tree
+from sklearn.ensemble import AdaBoostClassifier
+
+
+
+def getUserInput(models):
+    """
+    Get User Input
+    """
+    optionparser = OptionParser(add_help_option=False, epilog="multiline")
+
+    optionparser.add_option('-c', '--classifier', dest='classifier', default="all")
+    optionparser.add_option('-s', '--sample', dest='sample', default="all")
+    optionparser.add_option('-h', '--help', dest='help', action='store_true',
+                  help='show this help message and exit')
+    optionparser.add_option('-f', '--file', dest='file')
+
+
+    (option, args) = optionparser.parse_args()
+
+    if option.help:
+        print optionparser.print_help()
+        print __doc__
+        print "Supported Classifier Models:"
+
+
+        # print models
+        for index, key in enumerate(models):
+            print "%2s % 20s" % (index, key)
+
+        print "Default option: 'all'\n"
+
+        print "To run the program, provide app features file path"
+        print "Usage: --file='path.to.appData'"
+
+        sys.exit()
+
+
+    if not option.file:
+        return optionparser.error('Data File path not provided.\n Usage: --file="path.to.data"')
+
+
+    return {
+        'classifier' : option.classifier,
+        'file': option.file,
+        'sample' : option.sample
+    }
+
+
+
+
+
+def loadAppData(datafile):
+    """
+    Data File added
+    {
+        'fair' : False,
+        'unfair': True
+    }
+    """
+    df = pd.read_csv(datafile)
+
+    ## Remove the unnamed column as not sure
+    # cols = set(df.columns)
+    # cols.remove('Unnamed: 7')
+    # df = df[list(cols)]
+
+    ## Convert appLabel to boolean: True for 'unfair'
+    df['label'] = df['label'].map(lambda x: x=='unfair')
+
+    return df
+
+
+def trimDf(df):
+    """
+    Trim the dataframe provided
+    Remove features that we don't think are helping
+    """
+    cols = set(df.columns)
+
+    cols.remove('feat3') # bug in our feature extraction code
+    cols.remove('feat8') # considered only free apps
+
+
+    return df[list(cols)]
+
+
+
+
+def prepareSplitClassifier(df, models, choice):
+    """
+    Classify the apps for equal splits
+    """
+
+
+    def classificationOutput(clf, X, Y):
+        """
+        Fit the model and print the classification results
+        - confusion_matrix
+        - avg scores etc
+        """
+        n_samples = 36
+
+        print "\n\nClassifier: \n %s" % (clf)
+        print "#" * 79
+        # classifier_gnb = naive_bayes.GaussianNB() # initiating the classifier
+
+        clf.fit(X[:n_samples], Y[:n_samples]) # train on first n_samples and test on last 10
+
+        expected = Y[n_samples:]
+        predicted = clf.predict(X[n_samples:])
+        print("Classification report:\n%s\n" % (metrics.classification_report(expected, predicted)))
+        print("\nConfusion matrix:\n%s" % metrics.confusion_matrix(expected, predicted))
+
+
+
+
+    def splitclassify(cDf):
+        """
+        Given the dataframe combined with equal fair and unfair apps,
+        classify them
+        """
+        cDf = cDf.reindex(np.random.permutation(cDf.index)) # shuffle the dataframe
+        featCols = set(cDf.columns)
+        featCols.remove('label')
+
+        features = cDf[list(featCols)].astype('float')
+
+        ## Scale the features to a common range
+        min_max_scaler = preprocessing.MinMaxScaler()
+        X = min_max_scaler.fit_transform(features.values)
+
+        Y = cDf['label'].values
+
+
+        if choice == 'all':
+            for key in models:
+                classifier = models[key]
+                classificationOutput(classifier, X, Y)
+        else:
+            if choice in models:
+                classifier = models[choice]
+                classificationOutput(classifier, X, Y)
+            else:
+                print "Incorrect Choice"
+
+
+
+    fairDf = df[df['label'] == False]
+    unfairDf = df[df['label'] == True]
+
+
+    # calculate total possible splits of fair data frame relatie to
+    # size of unfair dataframe
+    splits = len(fairDf) // len(unfairDf)
+
+    for i in range(splits):
+        clDf = fairDf[i : i+len(unfairDf)].append(unfairDf)
+
+        # print fairDf.values, unfairDf.values
+        print "Classifying %d th split of fair  with unfair " % (i)
+        print "-" * 79
+        splitclassify(clDf)
+        print "\n\n"
+
+
+
+
+def performClassification(clf, featVector, labelVector, fold=4):
+    """
+    Perform Classification
+    """
+
+    (numrow, numcol) = featVector.shape
+
+    foldsize = int(numrow//fold)
+
+    print "FoldSize: %s" % (foldsize)
+
+    for i in range(fold):
+        X_test = featVector[i*foldsize:(i+1)*foldsize]
+        Y_test = labelVector[i*foldsize:(i+1)*foldsize]
+
+        X_train = np.concatenate((featVector[:i*foldsize], featVector[(i+1)*foldsize:]))
+        Y_train = np.concatenate((labelVector[:i*foldsize], labelVector[(i+1)*foldsize:]))
+
+        print " X_train: %s, Y_train: %s, X_test: %s, Y_test: %s" % (X_train.shape, Y_train.shape, X_test.shape, Y_test.shape)
+
+        print "#### Classifier: \n %s" % (clf)
+
+
+        clf.fit(X_train, Y_train) # train on first n_samples and test on last 10
+
+        expected = Y_test
+        predicted = clf.predict(X_test)
+        print "Classification report:\n%s\n" % metrics.classification_report(expected, predicted)
+        print "\nConfusion matrix:\n%s" % metrics.confusion_matrix(expected, predicted)
+
+
+
+
+def allClassifier(cDf, models, modelchoice):
+    """
+    Classifier for all apps
+    """
+
+    print "Data Size: %s, \t Model Choice: %s" % (cDf.shape, modelchoice)
+
+    cDf = cDf.reindex(np.random.permutation(cDf.index)) # shuffle the dataframe
+    featCols = set(cDf.columns)
+    featCols.remove('label')
+
+    features = cDf[list(featCols)].astype('float')
+
+    ## Scale the features to a common range
+    min_max_scaler = preprocessing.MinMaxScaler()
+    featVector = min_max_scaler.fit_transform(features.values) #scaled feature vector
+
+    labelVector = cDf['label'].values #label vector
+
+
+    if modelchoice == 'all':
+        for key in models:
+            if key != 'svm-nl':
+                classifier = models[key]
+                performClassification(classifier, featVector, labelVector)
+    else:
+        if modelchoice in models and modelchoice != 'svm-nl':
+            classifier = models[choice]
+            performClassification(classifier, featVector, labelVector)
+        else:
+            print "Incorrect Choice"
+
+
+
+
+
+
+def main():
+
+    # Supported classifier models
+    n_neighbors = 3
+    models = {
+        'nb' : naive_bayes.GaussianNB(),
+        'svm-l' : svm.SVC(),
+        'svm-nl' : svm.NuSVC(),
+        'tree' : tree.DecisionTreeClassifier(),
+        'forest': AdaBoostClassifier(tree.DecisionTreeClassifier(max_depth=1),algorithm="SAMME",n_estimators=200),
+        'knn-uniform' : neighbors.KNeighborsClassifier(n_neighbors, weights='uniform'),
+        'knn-distance' : neighbors.KNeighborsClassifier(n_neighbors, weights='distance')
+    }
+
+    userInput = getUserInput(models)
+    appDf = loadAppData(userInput['file'])
+    appDf = trimDf(appDf)
+
+    if userInput['sample'] == 'all':
+        allClassifier(appDf, models, userInput['classifier'])
+    else:
+        prepareSplitClassifier(appDf, models, userInput['classifier'])
+
+
+
+
+if __name__ == '__main__':
+    main()
diff --git a/data.csv b/data.csv
new file mode 100644
index 0000000..a5a9d2b
--- /dev/null
+++ b/data.csv
@@ -0,0 +1 @@
+feat0,feat1,feat2,feat3,feat4,feat5,feat6,feat7,feat8,feat9,feat10,label
4,4.051,1,0,TRUE,TRUE,TRUE,30000000,0,-3,601,fair
13,4.351,11,0,TRUE,TRUE,TRUE,30000000,0,2,1139,fair
23,4.555,20,0,FALSE,TRUE,TRUE,3000000,0,-4,2223,fair
10,4.623,5,0,TRUE,TRUE,FALSE,30000000,0,8,804,fair
22,4.046,16,0,FALSE,TRUE,TRUE,7500000,0,-11,1867,fair
18,4.595,6,0,TRUE,TRUE,FALSE,30000000,0,1,1162,fair
18,4.526,60,0,TRUE,TRUE,TRUE,30000000,0,-4,1522,fair
13,4.039,19,0,FALSE,TRUE,FALSE,30000000,0,-5,1895,fair
11,4.4,10,0,TRUE,TRUE,TRUE,3000000,0,-2,1195,fair
19,3.935,11,0,TRUE,TRUE,TRUE,300000,0,-4,1488,fair
18,4.075,35,0,TRUE,TRUE,FALSE,3000000,0,-5,1864,fair
19,3.983,14,0,FALSE,TRUE,FALSE,750000,0,-2,2049,fair
8,4.238,2,0,TRUE,TRUE,FALSE,30000000,0,1,417,fair
16,3.915,11,0,TRUE,TRUE,FALSE,3000000,0,-3,1276,fair
13,4.05,12,0,TRUE,TRUE,FALSE,750000,0,-3,1210,fair
20,3.795,24,0,TRUE,TRUE,TRUE,750000,0,-7,2038,fair
12,3.997,10,0,TRUE,TRUE,FALSE,7500000,0,1,1044,fair
15,3.212,13,0,TRUE,TRUE,TRUE,3000000,0,-5,1245,fair
2,2.611,2,0,TRUE,TRUE,FALSE,3000000,0,-1,225,fair
15,4.547,10,0,TRUE,TRUE,FALSE,30000000,0,-3,1120,fair
22,2.671,11,0,TRUE,TRUE,TRUE,3000000,0,4,1406,fair
13,4.045,10,0,FALSE,TRUE,TRUE,750000,0,-3,1063,fair
7,3.555,10,0,TRUE,TRUE,FALSE,750000,0,-9,855,fair
17,4.59,38,0,TRUE,TRUE,TRUE,3000000,0,-5,2147,fair
12,4.258,13,0,TRUE,TRUE,TRUE,7500000,0,-9,1189,fair
19,4.428,8,0,TRUE,TRUE,FALSE,750000,0,-10,1804,fair
26,4.401,5,0,TRUE,TRUE,FALSE,3000000,0,5,1514,fair
12,4.275,12,0,TRUE,TRUE,TRUE,3000000,0,-6,1272,fair
17,4.149,14,0,TRUE,TRUE,TRUE,750000,0,-8,1413,fair
7,4.396,2,0,TRUE,TRUE,TRUE,300000,0,-3,610,fair
14,4.113,5,0,TRUE,TRUE,TRUE,3000000,0,2,1145,fair
14,4.24,14,0,TRUE,TRUE,FALSE,30000,0,2,1413,fair
2,4.241,14,0,TRUE,TRUE,TRUE,300000,0,-4,573,fair
12,3.989,16,0,TRUE,TRUE,TRUE,3000000,0,-6,1387,fair
19,4.31,7,0,TRUE,TRUE,TRUE,75000,0,-7,1336,fair
7,4.451,5,0,TRUE,TRUE,TRUE,3000000,0,5,817,fair
29,3.916,16,0,TRUE,TRUE,TRUE,300000,0,-9,2205,fair
7,4.761,3,0,TRUE,TRUE,FALSE,300000,0,1,541,fair
6,4.158,1,0,TRUE,TRUE,FALSE,300000,0,0,310,fair
6,2.972,4,0,TRUE,TRUE,FALSE,300000,0,-4,880,fair
5,3.903,4,0,TRUE,TRUE,FALSE,300000,0,-6,583,fair
19,3.433,20,0,TRUE,TRUE,FALSE,3000000,0,-9,1888,fair
13,4.412,4,0,TRUE,TRUE,FALSE,3000000,0,-3,1122,fair
15,4.461,21,0,FALSE,TRUE,TRUE,3000000,0,15,1613,fair
5,3.564,11,0,TRUE,FALSE,FALSE,300000,0,-2,1014,fair
8,4.131,6,0,TRUE,FALSE,FALSE,3000000,0,5,1115,fair
6,3.55,8,0,TRUE,TRUE,TRUE,30000,0,-1,437,fair
2,4.435,0,0,TRUE,TRUE,TRUE,3000000,0,1,307,fair
13,4.233,4,0,TRUE,TRUE,TRUE,300000,0,-3,991,fair
7,3.975,6,0,TRUE,TRUE,FALSE,3000000,0,1,578,fair
10,3.926,7,0,TRUE,TRUE,TRUE,300000,0,4,962,fair
13,4.59,6,0,TRUE,TRUE,FALSE,3000000,0,2,1267,fair
9,3.601,8,0,TRUE,TRUE,FALSE,300000,0,-9,1300,fair
16,3.701,12,0,TRUE,TRUE,FALSE,300000,0,3,1051,fair
16,2.931,10,0,TRUE,TRUE,TRUE,300000,0,-13,1822,fair
7,4.564,0,0,TRUE,TRUE,FALSE,3000000,0,0,535,fair
11,4.179,8,0,TRUE,TRUE,FALSE,7500000,0,-7,1075,fair
6,4.466,4,0,TRUE,TRUE,TRUE,3000000,0,7,691,fair
7,4.34,5,0,FALSE,TRUE,TRUE,750000,0,-3,991,fair
7,4.539,14,0,TRUE,TRUE,FALSE,300000,0,3,805,fair
23,4.135,18,0,TRUE,TRUE,TRUE,3000000,0,0,1509,fair
22,4.48,13,0,TRUE,FALSE,FALSE,300000,0,0,1661,fair
22,4.141,20,0,TRUE,TRUE,TRUE,7500000,0,3,1484,fair
15,4.557,13,0,TRUE,FALSE,FALSE,3000000,0,-6,1431,fair
12,4.452,16,0,TRUE,FALSE,FALSE,300000,0,2,929,fair
10,4.482,4,0,TRUE,FALSE,FALSE,3000000,0,4,577,fair
9,4.572,6,0,TRUE,TRUE,FALSE,3000000,0,-4,833,fair
13,4.789,5,0,TRUE,TRUE,FALSE,300000,0,8,869,fair
14,4.212,21,0,TRUE,FALSE,FALSE,750000,0,-3,1140,fair
15,4.226,7,0,TRUE,TRUE,TRUE,3000000,0,3,651,fair
7,4.182,3,0,TRUE,FALSE,TRUE,3000000,0,-3,516,fair
7,4.528,3,0,TRUE,TRUE,FALSE,300000,0,0,326,fair
8,4.384,4,0,TRUE,TRUE,FALSE,3000000,0,0,790,fair
16,4.123,3,0,TRUE,TRUE,FALSE,750000,0,-7,1138,fair
19,4.399,10,0,TRUE,TRUE,FALSE,30000,0,0,1284,fair
26,4.239,18,0,TRUE,TRUE,FALSE,3000000,0,-8,2334,fair
5,4.249,1,0,TRUE,FALSE,FALSE,750000,0,-1,585,fair
30,3.228,24,0,TRUE,TRUE,TRUE,30000,0,2,2314,fair
14,4.301,10,0,TRUE,TRUE,FALSE,3000000,0,-6,1573,fair
13,4.095,11,0,TRUE,FALSE,FALSE,750000,0,5,952,fair
1,4.223,0,0,TRUE,FALSE,FALSE,75000,0,1,156,fair
9,4.413,4,0,FALSE,TRUE,FALSE,300000,0,2,607,fair
8,4.018,7,0,TRUE,FALSE,FALSE,75000,0,-1,441,fair
7,4.529,3,0,TRUE,TRUE,TRUE,3000000,0,-1,674,fair
31,4.341,16,0,TRUE,FALSE,FALSE,300000,0,-2,1784,fair
2,4.613,1,0,TRUE,TRUE,FALSE,750000,0,5,153,fair
26,4.664,7,0,TRUE,TRUE,TRUE,750000,0,-5,1339,fair
7,4.402,2,0,TRUE,TRUE,FALSE,3000000,0,4,519,fair
12,4.573,9,0,TRUE,FALSE,FALSE,300000,0,-1,928,fair
8,4.302,0,0,TRUE,TRUE,TRUE,750000,0,0,386,fair
11,4.035,0,0,TRUE,TRUE,FALSE,300000,0,-3,664,fair
12,4.256,9,0,TRUE,TRUE,TRUE,3000000,0,0,909,fair
2,3.805,2,0,TRUE,TRUE,TRUE,750000,0,3,144,fair
7,4.31,1,0,TRUE,TRUE,FALSE,300000,0,0,376,fair
14,4.528,7,0,TRUE,TRUE,FALSE,3000000,0,3,740,fair
19,2.117,17,0,TRUE,TRUE,TRUE,75000,0,-9,1787,fair
22,4.421,15,0,TRUE,TRUE,FALSE,30000,0,-2,1289,fair
3,4.261,1,0,TRUE,TRUE,TRUE,30000000,0,3,547,fair
8,4.629,8,0,TRUE,TRUE,FALSE,30000,0,0,757,fair
13,4.05,9,0,TRUE,TRUE,TRUE,75000,0,-7,1484,fair
10,4.685,4,0,TRUE,TRUE,FALSE,300000,0,0,687,fair
8,4.339,12,0,TRUE,FALSE,FALSE,30000,0,-1,636,fair
10,4.646,12,0,TRUE,TRUE,FALSE,30000,0,-4,781,fair
3,4.182,1,0,TRUE,FALSE,FALSE,75000,0,2,161,fair
5,4.656,7,0,TRUE,TRUE,FALSE,30000,0,3,614,fair
2,4.465,2,0,TRUE,FALSE,FALSE,75000,0,1,34,fair
3,4.279,2,0,TRUE,FALSE,FALSE,750000,0,1,166,fair
3,4.677,0,0,TRUE,TRUE,FALSE,7500,0,1,77,fair
2,3.901,1,0,TRUE,FALSE,FALSE,300000,0,-3,420,fair
5,4.115,3,0,TRUE,TRUE,FALSE,750000,0,4,429,fair
1,4.429,1,0,TRUE,FALSE,FALSE,3000,0,0,69,fair
9,4.787,11,0,TRUE,TRUE,FALSE,7500,0,1,969,fair
3,4.815,2,0,TRUE,TRUE,FALSE,7500,0,4,186,fair
11,4.147,15,0,TRUE,TRUE,TRUE,75000,0,-1,1138,fair
1,4.696,3,0,TRUE,TRUE,FALSE,7500,0,0,187,fair
7,4.408,10,0,TRUE,FALSE,FALSE,300000,0,7,346,fair
11,4.52,5,0,TRUE,TRUE,FALSE,300000,0,2,534,fair
0,4.222,0,0,TRUE,TRUE,FALSE,7500,0,0,17,fair
10,4.197,10,0,TRUE,TRUE,FALSE,30000,0,1,1044,fair
2,3.952,2,0,TRUE,TRUE,FALSE,7500,0,2,176,fair
17,4.271,14,0,TRUE,TRUE,TRUE,300000000,0,-7,1493,fair
13,3.956,11,0,TRUE,TRUE,TRUE,300000000,0,-10,1519,fair
16,4.424,14,0,TRUE,TRUE,TRUE,30000000,0,-4,1187,fair
14,4.539,17,0,TRUE,TRUE,TRUE,300000000,0,-11,1365,fair
15,3.994,11,0,TRUE,TRUE,TRUE,75000000,0,-5,1743,fair
4,4.436,5,0,TRUE,TRUE,TRUE,75000000,0,-8,1106,fair
15,4.559,26,0,TRUE,TRUE,TRUE,300000000,0,1,1639,fair
15,4.425,11,0,TRUE,TRUE,TRUE,300000000,0,-3,1070,fair
19,4.18,18,0,FALSE,TRUE,TRUE,300000000,0,-7,1221,fair
14,4.179,6,0,FALSE,TRUE,FALSE,300000000,0,-9,1122,fair
20,4.395,12,0,FALSE,TRUE,TRUE,75000000,0,-2,1457,fair
16,4.201,21,0,FALSE,TRUE,TRUE,30000000,0,-5,1638,fair
14,4.164,18,0,TRUE,TRUE,TRUE,3000000,0,-7,1255,fair
20,4.394,24,0,TRUE,TRUE,TRUE,30000000,0,-8,1469,fair
24,4.273,19,0,TRUE,TRUE,FALSE,7500000,0,-3,1717,fair
6,4.659,2,0,TRUE,TRUE,TRUE,30000000,0,2,404,fair
15,3.958,16,0,TRUE,TRUE,FALSE,7500000,0,-8,1561,fair
5,4.333,14,0,TRUE,TRUE,FALSE,75000000,0,-4,1100,fair
15,4.135,15,0,TRUE,TRUE,FALSE,30000000,0,-10,1389,fair
15,4.42,3,0,TRUE,TRUE,FALSE,30000000,0,-9,1508,fair
14,4.398,11,0,FALSE,TRUE,FALSE,75000000,0,-9,1243,fair
20,4.242,29,0,TRUE,TRUE,TRUE,3000000,0,-8,1591,fair
10,2.843,2,0,FALSE,TRUE,FALSE,30000000,0,-5,942,fair
14,4.229,11,0,TRUE,TRUE,TRUE,30000000,0,-9,1413,fair
18,3.991,16,0,TRUE,TRUE,TRUE,3000000,0,-4,1969,fair
14,4.12,6,0,TRUE,TRUE,FALSE,7500000,0,-9,1392,fair
12,4.33,26,0,TRUE,TRUE,TRUE,7500000,0,-3,1528,fair
16,4.263,13,0,FALSE,TRUE,TRUE,750000000,0,-9,1295,fair
9,3.97,26,0,TRUE,TRUE,TRUE,3000000,0,-13,1676,fair
8,4.545,14,0,TRUE,TRUE,TRUE,75000000,0,-6,1294,fair
16,3.77,20,0,TRUE,TRUE,TRUE,3000000,0,-9,1919,fair
7,4.386,37,0,TRUE,TRUE,TRUE,3000000,0,-4,908,fair
15,3.46,15,0,TRUE,TRUE,TRUE,300000000,0,2,1345,fair
18,4.289,11,0,TRUE,TRUE,FALSE,300000000,0,-8,1970,fair
14,4.364,24,0,TRUE,TRUE,FALSE,7500000,0,5,1337,fair
14,3.131,15,0,TRUE,TRUE,TRUE,3000000,0,-2,1123,fair
3,4.327,1,0,TRUE,TRUE,TRUE,3000000,0,-1,211,fair
9,3.616,17,0,TRUE,TRUE,FALSE,3000000,0,-1,1508,fair
13,3.683,18,0,TRUE,TRUE,FALSE,3000000,0,2,1252,fair
18,4.517,6,0,TRUE,TRUE,TRUE,3000000,0,-5,1590,fair
21,4.317,10,0,TRUE,TRUE,FALSE,30000000,0,-8,1358,fair
10,3.545,15,0,TRUE,TRUE,FALSE,3000000,0,-4,1350,fair
13,4.337,17,0,TRUE,TRUE,FALSE,3000000,0,-12,1577,fair
9,4.009,8,0,TRUE,FALSE,FALSE,300000,0,3,728,fair
19,3.957,24,0,TRUE,TRUE,FALSE,300000000,0,-1,1707,fair
16,3.298,13,0,TRUE,TRUE,FALSE,300000,0,-3,1817,fair
9,4.329,7,0,TRUE,TRUE,FALSE,30000000,0,-2,1058,fair
11,4.185,5,0,TRUE,TRUE,FALSE,3000000,0,-4,1217,fair
17,4.303,26,0,TRUE,TRUE,TRUE,3000000,0,-2,1447,fair
8,4.14,19,0,TRUE,TRUE,FALSE,30000000,0,-9,1574,fair
24,4.551,4,0,TRUE,TRUE,FALSE,300000,0,0,1354,fair
12,4.388,15,0,TRUE,TRUE,FALSE,30000000,0,-9,1427,fair
17,4.241,12,0,TRUE,TRUE,TRUE,3000000,0,-5,1685,fair
9,4.168,6,0,TRUE,TRUE,TRUE,750000,0,-3,946,fair
13,3.932,15,0,TRUE,TRUE,FALSE,300000,0,-1,1139,fair
9,4.541,15,0,TRUE,TRUE,TRUE,300000,0,-7,1251,fair
32,4.361,10,0,TRUE,TRUE,TRUE,7500000,0,-14,2454,fair
13,4.433,7,0,TRUE,TRUE,TRUE,3000000,0,3,677,fair
12,4.491,7,0,TRUE,TRUE,TRUE,75000,0,1,999,fair
15,4.198,10,0,TRUE,TRUE,TRUE,3000000,0,-3,1053,fair
17,3.822,12,0,TRUE,TRUE,TRUE,3000000,0,-9,1737,fair
16,4.436,2,0,TRUE,TRUE,TRUE,7500000,0,3,1257,fair
20,4.763,10,0,TRUE,TRUE,TRUE,3000000,0,-7,1347,fair
20,3.622,12,0,TRUE,TRUE,FALSE,3000000,0,-1,1669,fair
9,4.684,10,0,TRUE,TRUE,FALSE,30000000,0,4,877,fair
15,4.448,7,0,TRUE,TRUE,FALSE,3000000,0,-4,1015,fair
5,4.199,6,0,TRUE,TRUE,FALSE,300000,0,-2,601,fair
6,4.42,10,0,TRUE,TRUE,FALSE,3000000,0,2,1059,fair
24,4.188,10,0,TRUE,TRUE,TRUE,3000000,0,-7,1868,fair
2,3.835,1,0,TRUE,TRUE,FALSE,3000000,0,3,274,fair
18,4.468,12,0,TRUE,TRUE,TRUE,750000,0,1,1271,fair
16,4.603,9,0,TRUE,TRUE,TRUE,7500000,0,5,1251,fair
11,4.253,12,0,TRUE,TRUE,TRUE,3000000,0,-3,1568,fair
17,4.578,12,0,TRUE,TRUE,TRUE,3000000,0,2,1352,fair
16,4.282,33,0,TRUE,TRUE,FALSE,3000000,0,-4,1579,fair
13,4.287,2,0,TRUE,FALSE,FALSE,750000,0,9,719,fair
20,4.494,15,0,TRUE,TRUE,TRUE,7500000,0,-2,1460,fair
16,2.845,24,0,TRUE,TRUE,FALSE,3000000,0,-12,1792,fair
16,4.458,6,0,TRUE,TRUE,TRUE,3000000,0,-3,1299,fair
12,4.264,10,0,TRUE,TRUE,FALSE,3000000,0,9,791,fair
18,4.309,15,0,TRUE,TRUE,TRUE,3000000,0,-4,1413,fair
10,4.38,16,0,FALSE,TRUE,TRUE,300000,0,3,1126,fair
8,3.035,13,0,TRUE,TRUE,FALSE,75000,0,0,1058,fair
11,4.732,9,0,TRUE,TRUE,FALSE,300000,0,-4,860,fair
5,4.325,1,0,TRUE,TRUE,TRUE,75000,0,-2,483,fair
12,4.291,1,0,TRUE,TRUE,TRUE,3000000,0,-1,572,fair
13,4.367,7,0,TRUE,TRUE,FALSE,30000000,0,-3,931,fair
13,4.612,8,0,TRUE,TRUE,TRUE,7500000,0,-5,824,fair
23,4.2,10,0,TRUE,TRUE,FALSE,3000000,0,-1,1731,fair
14,4.594,14,0,TRUE,TRUE,FALSE,750000,0,2,960,fair
23,4.845,11,0,TRUE,TRUE,TRUE,3000000,0,10,1540,fair
6,4.208,14,0,TRUE,TRUE,TRUE,750000,0,2,570,fair
13,4.391,17,0,TRUE,TRUE,FALSE,3000000,0,-4,1617,fair
10,3.965,10,0,TRUE,TRUE,TRUE,300000,0,4,1189,fair
7,4.522,11,0,TRUE,TRUE,FALSE,7500000,0,-3,914,fair
10,3.858,4,0,TRUE,TRUE,TRUE,3000000,0,4,635,fair
5,4.405,21,0,TRUE,TRUE,FALSE,3000000,0,8,584,fair
18,4.211,11,0,TRUE,TRUE,TRUE,3000000,0,-3,1693,fair
5,3.734,28,0,TRUE,TRUE,TRUE,3000000,0,0,814,fair
13,4.021,14,0,TRUE,TRUE,TRUE,3000000,0,0,694,fair
13,3.987,17,0,TRUE,TRUE,TRUE,300000,0,1,1274,fair
14,4.126,3,0,TRUE,TRUE,FALSE,750000,0,-2,1194,fair
11,3.771,7,0,TRUE,TRUE,TRUE,3000000,0,-2,1355,fair
11,3.505,10,0,TRUE,TRUE,TRUE,300000,0,0,931,fair
18,4.528,7,0,TRUE,TRUE,FALSE,7500000,0,-1,1188,fair
16,4.527,7,0,TRUE,TRUE,TRUE,3000000,0,-9,1776,fair
18,3.603,14,0,TRUE,TRUE,FALSE,750000,0,-8,1640,fair
33,4.227,17,0,TRUE,TRUE,TRUE,300000,0,3,1782,fair
12,4.038,9,0,TRUE,TRUE,FALSE,3000000,0,3,939,fair
10,4.075,3,0,TRUE,TRUE,TRUE,3000000,0,-6,1064,fair
7,4.808,3,0,TRUE,TRUE,FALSE,3000000,0,-4,859,fair
10,4.289,6,0,TRUE,TRUE,FALSE,7500000,0,6,785,fair
24,4.807,37,0,TRUE,TRUE,FALSE,3000000,0,-1,1484,fair
21,4.45,6,0,TRUE,TRUE,TRUE,3000000,0,10,1501,fair
8,3.976,7,0,FALSE,TRUE,TRUE,300000,0,-3,770,fair
20,3.844,13,0,TRUE,TRUE,FALSE,750000,0,-15,1642,fair
28,4.592,15,0,TRUE,TRUE,TRUE,750000,0,4,1941,fair
7,3.167,12,0,TRUE,TRUE,FALSE,300000,0,-6,989,fair
8,3.998,11,0,TRUE,TRUE,TRUE,3000000,0,0,721,fair
2,3.359,23,0,TRUE,FALSE,FALSE,750000,0,0,691,fair
4,3.727,20,0,TRUE,TRUE,TRUE,750000000,0,-6,673,fair
11,4.601,30,0,TRUE,TRUE,TRUE,300000000,0,-2,1294,fair
20,3.726,14,0,TRUE,TRUE,TRUE,30000000,0,-10,1219,fair
10,3.968,13,0,TRUE,TRUE,TRUE,300000000,0,-5,1177,fair
13,4.389,13,0,TRUE,TRUE,TRUE,300000000,0,-4,1222,fair
18,4.73,15,0,FALSE,TRUE,TRUE,30000000,0,-5,1431,fair
16,3.435,9,0,TRUE,TRUE,TRUE,30000000,0,-1,1567,fair
6,4.286,8,0,TRUE,TRUE,FALSE,30000000,0,-3,873,fair
16,4.225,28,0,TRUE,TRUE,TRUE,30000000,0,-5,1353,fair
17,4.347,10,0,TRUE,TRUE,TRUE,30000000,0,-4,1390,fair
9,4.152,12,0,FALSE,TRUE,TRUE,30000000,0,-1,1112,fair
11,4.237,15,0,TRUE,TRUE,TRUE,7500000,0,-4,1553,fair
22,4.33,7,0,TRUE,TRUE,TRUE,7500000,0,-1,1380,fair
17,4.197,15,0,TRUE,TRUE,TRUE,7500000,0,-4,1156,fair
19,4.337,8,0,TRUE,TRUE,TRUE,3000000,0,5,942,fair
13,4.238,8,0,TRUE,TRUE,TRUE,7500000,0,-2,1066,fair
10,4.135,14,0,TRUE,TRUE,TRUE,7500000,0,-8,912,fair
12,4.214,20,0,TRUE,TRUE,TRUE,7500000,0,-3,1099,fair
4,4.061,6,0,TRUE,TRUE,TRUE,30000000,0,-2,672,fair
11,4.289,17,0,TRUE,TRUE,TRUE,3000000,0,-11,1376,fair
2,4.356,2,0,TRUE,TRUE,TRUE,3000000,0,-1,220,fair
18,4.527,18,0,TRUE,TRUE,TRUE,7500000,0,3,795,fair
7,4.533,7,0,TRUE,TRUE,TRUE,30000000,0,2,602,fair
3,3.724,13,0,TRUE,FALSE,FALSE,3000000,0,-5,708,fair
12,4.213,17,0,TRUE,TRUE,TRUE,30000000,0,-6,1552,fair
17,4.138,14,0,FALSE,TRUE,TRUE,300000000,0,-8,1500,fair
12,4.098,8,0,TRUE,TRUE,TRUE,30000000,0,-1,1204,fair
20,4.386,14,0,TRUE,TRUE,TRUE,7500000,0,-3,1220,fair
13,3.806,6,0,TRUE,TRUE,FALSE,3000000,0,-5,972,fair
14,4.196,10,0,TRUE,TRUE,TRUE,30000000,0,-2,968,fair
15,4.247,12,0,TRUE,TRUE,TRUE,3000000,0,-2,1534,fair
13,4.413,10,0,TRUE,TRUE,TRUE,3000000,0,0,714,fair
12,4.193,5,0,TRUE,TRUE,FALSE,7500000,0,-5,732,fair
15,4.209,8,0,TRUE,FALSE,FALSE,750000,0,-5,1102,fair
28,4.104,20,0,TRUE,TRUE,TRUE,3000000,0,-6,2239,fair
24,4.147,19,0,TRUE,TRUE,TRUE,300000,0,-13,1939,fair
5,4.405,5,0,TRUE,TRUE,TRUE,3000000,0,2,389,fair
9,4.136,5,0,TRUE,TRUE,TRUE,3000000,0,-5,598,fair
26,2.946,11,0,TRUE,TRUE,TRUE,3000000,0,-17,1904,fair
9,3.978,10,0,TRUE,TRUE,TRUE,3000000,0,-1,1029,fair
18,4.576,25,0,TRUE,TRUE,TRUE,30000000,0,1,1342,fair
13,4.548,15,0,TRUE,TRUE,TRUE,3000000,0,-1,1354,fair
17,3.099,10,0,TRUE,TRUE,TRUE,3000000,0,-2,1177,fair
12,4.025,13,0,TRUE,FALSE,FALSE,300000,0,-7,952,fair
5,4.324,10,0,TRUE,TRUE,FALSE,3000000,0,0,669,fair
15,4.335,14,0,TRUE,TRUE,TRUE,7500000,0,-9,1348,fair
3,4.381,46,0,TRUE,TRUE,TRUE,7500000,0,5,525,fair
12,3.594,12,0,TRUE,TRUE,FALSE,300000,0,-1,970,fair
2,4.403,18,0,TRUE,TRUE,TRUE,3000000,0,3,384,fair
19,4.125,46,0,TRUE,TRUE,TRUE,3000000,0,-5,1179,fair
21,3.088,9,0,TRUE,TRUE,TRUE,3000000,0,-9,1593,fair
17,4.534,11,0,TRUE,TRUE,TRUE,3000000,0,2,1085,fair
12,4.602,5,0,TRUE,TRUE,FALSE,3000000,0,-4,992,fair
9,3.977,6,0,TRUE,TRUE,TRUE,7500000,0,-2,714,fair
6,4.464,2,0,TRUE,TRUE,TRUE,3000000,0,1,433,fair
25,4.443,14,0,FALSE,TRUE,FALSE,7500000,0,-5,2242,fair
10,4.391,5,0,FALSE,TRUE,FALSE,3000000,0,-2,1176,fair
14,4.163,8,0,TRUE,TRUE,TRUE,3000000,0,-7,1554,fair
20,4.676,19,0,TRUE,TRUE,FALSE,300000,0,7,990,fair
16,4.397,109,0,TRUE,TRUE,TRUE,7500000,0,-3,1713,fair
18,3.225,9,0,TRUE,TRUE,FALSE,30000,0,-17,2087,unfair
2,1,19,0,TRUE,TRUE,FALSE,3000,0,-1,306,unfair
7,3.149,20,0,TRUE,TRUE,FALSE,30000,0,-6,588,unfair
7,1.951,4,0,TRUE,TRUE,FALSE,30000,0,-9,843,unfair
19,2.062,9,0,TRUE,TRUE,TRUE,750000,0,-7,1281,unfair
15,2.091,6,0,TRUE,TRUE,FALSE,300,0.9,-4,1002,unfair
13,4.554,12,0,FALSE,TRUE,TRUE,3000000,0,1,1492,unfair
8,4.522,98,0,TRUE,TRUE,TRUE,30000000,0,-4,1253,unfair
13,4.054,12,0,TRUE,TRUE,FALSE,750000,0,-3,1210,unfair
7,3.551,10,0,TRUE,TRUE,FALSE,750000,0,-9,855,unfair
9,4.256,9,0,TRUE,TRUE,TRUE,7500000,0,-4,955,unfair
2,4.117,2,0,TRUE,TRUE,TRUE,3000000,0,1,219,unfair
41,4.314,8,0,TRUE,TRUE,FALSE,30000,0,13,2388,unfair
17,3.987,17,0,TRUE,TRUE,TRUE,3000000,0,-10,1997,unfair
2,4.111,2,0,TRUE,TRUE,FALSE,300000,0,1,212,unfair
8,2.949,12,0,TRUE,TRUE,FALSE,300000,0,-6,969,unfair
7,3.586,14,0,TRUE,FALSE,FALSE,300000,0,-1,1054,unfair
10,4.047,1,0,TRUE,TRUE,TRUE,30000,0,1,775,unfair
10,3.586,7,0,TRUE,FALSE,FALSE,300000,0,-9,1414,unfair
7,4.175,3,0,TRUE,FALSE,TRUE,3000000,0,-3,516,unfair
11,4.424,10,0,TRUE,TRUE,TRUE,30000000,0,4,934,unfair
12,4.535,18,0,TRUE,TRUE,TRUE,300000000,0,-9,979,unfair
3,4.554,7,0,TRUE,TRUE,TRUE,300000000,0,1,559,unfair
\ No newline at end of file