def crossvalidate():
''' Iris dataset '''
path = '/host/Users/vandana/Documents/Grad_Studies/Spring2012/Machine Learning/Projects/iris/iris.data'
mldata = loadiris(path)
#normaldata = mldata.normalizeinput()
''' Cars dataset '''
path1 = '/host/Users/vandana/Documents/Grad_Studies/Spring2012/Machine Learning/Projects/cars/car.data'
mldata1 = loadcars(path1)
''' Mushroom dataset '''
path2 = '/host/Users/vandana/Documents/Grad_Studies/Spring2012/Machine Learning/Projects/mushroom/agaricus-lepiota.data'
mldata2 = loadmushroom(path2)
''' Voting dataset '''
path3 = '/host/Users/vandana/Documents/Grad_Studies/Spring2012/Machine Learning/Projects/voting/house-votes-84.data'
mldata3 = loadvoting(path3)
''' Heart Disease dataset '''
path4 = '/host/Users/vandana/Documents/Grad_Studies/Spring2012/Machine Learning/Projects/heart_disease/processed.cleveland.data'
mldata4 = loadheartdisease(path4, True)
#normaldata4 = mldata4.normalizeinput()
''' Wine dataset '''
path5 = '/host/Users/vandana/Documents/Grad_Studies/Spring2012/Machine Learning/Projects/wine/wine.data'
mldata5 = loadwine(path5)
#normaldata5 = mldata5.normalizeinput()
cvacc = {}
cvacc['Iris'] = cv10foldresult(mldata)
cvacc['Cars'] = cv10foldresult(mldata1)
cvacc['Mushrm'] = cv10foldresult(mldata2)
cvacc['Voting'] = cv10foldresult(mldata3)
cvacc['Heart'] = cv10foldresult(mldata4)
cvacc['Wine'] = cv10foldresult(mldata5)
for i in cvacc:
print i, ":", cvacc[i]
def crossvalidate():
''' Iris dataset '''
path = '/host/Users/vandana/Documents/Grad_Studies/Spring2012/Machine Learning/Projects/iris/iris.data'
mldata = loadiris(path)
#normaldata = mldata.normalizeinput()
''' Cars dataset '''
path1 = '/host/Users/vandana/Documents/Grad_Studies/Spring2012/Machine Learning/Projects/cars/car.data'
mldata1 = loadcars(path1)
''' Mushroom dataset '''
path2 = '/host/Users/vandana/Documents/Grad_Studies/Spring2012/Machine Learning/Projects/mushroom/agaricus-lepiota.data'
mldata2 = loadmushroom(path2)
''' Voting dataset '''
path3 = '/host/Users/vandana/Documents/Grad_Studies/Spring2012/Machine Learning/Projects/voting/house-votes-84.data'
mldata3 = loadvoting(path3)
''' Heart Disease dataset '''
path4 = '/host/Users/vandana/Documents/Grad_Studies/Spring2012/Machine Learning/Projects/heart_disease/processed.cleveland.data'
mldata4 = loadheartdisease(path4, True)
#normaldata4 = mldata4.normalizeinput()
''' Wine dataset '''
path5 = '/host/Users/vandana/Documents/Grad_Studies/Spring2012/Machine Learning/Projects/wine/wine.data'
mldata5 = loadwine(path5)
#normaldata5 = mldata5.normalizeinput()
cvacc = {}
cvacc['Iris'] = cv10foldresult(mldata)
cvacc['Cars'] = cv10foldresult(mldata1)
cvacc['Mushrm'] = cv10foldresult(mldata2)
cvacc['Voting'] = cv10foldresult(mldata3)
cvacc['Heart'] = cv10foldresult(mldata4)
cvacc['Wine'] = cv10foldresult(mldata5)
for i in cvacc:
print i, ":", cvacc[i]
def crossvalidate():
''' Iris dataset '''
path = '/host/Users/vandana/Documents/Grad_Studies/Spring2012/Machine Learning/Projects/iris/iris.data'
mldata = loadiris(path)
normaldata = mldata.normalizeinput()
''' Cars dataset '''
path1 = '/host/Users/vandana/Documents/Grad_Studies/Spring2012/Machine Learning/Projects/cars/car.data'
mldata1 = loadcars(path1)
''' Mushroom dataset '''
path2 = '/host/Users/vandana/Documents/Grad_Studies/Spring2012/Machine Learning/Projects/mushroom/agaricus-lepiota.data'
mldata2 = loadmushroom(path2)
''' Voting dataset '''
path3 = '/host/Users/vandana/Documents/Grad_Studies/Spring2012/Machine Learning/Projects/voting/house-votes-84.data'
mldata3 = loadvoting(path3)
''' Heart Disease dataset '''
path4 = '/host/Users/vandana/Documents/Grad_Studies/Spring2012/Machine Learning/Projects/heart_disease/processed.cleveland.data'
#path4 = '/host/Users/vandana/Documents/Grad_Studies/Spring2012/Machine Learning/Projects/heart_disease/processed.switzerland.data'
#mldata4 = loadheartdisease(path4, True, True)
mldata4 = loadheartdisease(path4, True)
normaldata4 = mldata4.normalizeinput()
''' Wine dataset '''
path5 = '/host/Users/vandana/Documents/Grad_Studies/Spring2012/Machine Learning/Projects/wine/wine.data'
mldata5 = loadwine(path5)
normaldata5 = mldata5.normalizeinput()
cvacc = {}
"""
cvacc['Iris'] = cv10foldresult(normaldata, [3], learn_rate=0.05, epoch=700, stop_criterion=50)
cvacc['Cars'] = cv10foldresult(mldata1, [5], learn_rate=0.01, epoch=1000, stop_criterion=100)
cvacc['Mushrm'] = cv10foldresult(mldata2, [5], learn_rate=0.05, epoch=50, stop_criterion=4)
cvacc['Voting'] = cv10foldresult(mldata3, [5], learn_rate=0.01, epoch=500, stop_criterion=100)
cvacc['Heart'] = cv10foldresult(normaldata4, [10, 5], learn_rate=0.01, epoch=1000, stop_criterion=100)
cvacc['Wine'] = cv10foldresult(normaldata5, [5], learn_rate=0.005, epoch=500, stop_criterion=50)
"""
#cvacc['Iris'] = cv10foldresult(normaldata, None, learn_rate=0.05, epoch=700, stop_criterion=50)
cvacc['Cars'] = cv10foldresult(mldata1,
None,
learn_rate=0.01,
epoch=1000,
stop_criterion=50)
cvacc['Mushrm'] = cv10foldresult(mldata2,
None,
learn_rate=0.08,
epoch=50,
stop_criterion=2)
#cvacc['Voting'] = cv10foldresult(mldata3, None, learn_rate=0.05, epoch=500, stop_criterion=100)
#cvacc['Heart'] = cv10foldresult(normaldata4, None, learn_rate=0.01, epoch=1000, stop_criterion=100)
#cvacc['Wine'] = cv10foldresult(normaldata5, None, learn_rate=0.005, epoch=500, stop_criterion=50)
for i in cvacc:
print i, ":", cvacc[i]
f = open("nn_cv", 'w')
pickle.dump(cvacc, f)
f.close()
def crossvalidate():
''' Iris dataset '''
path = '/host/Users/vandana/Documents/Grad_Studies/Spring2012/Machine Learning/Projects/iris/iris.data'
mldata = loadiris(path)
normaldata = mldata.normalizeinput()
''' Cars dataset '''
path1 = '/host/Users/vandana/Documents/Grad_Studies/Spring2012/Machine Learning/Projects/cars/car.data'
mldata1 = loadcars(path1)
''' Mushroom dataset '''
path2 = '/host/Users/vandana/Documents/Grad_Studies/Spring2012/Machine Learning/Projects/mushroom/agaricus-lepiota.data'
mldata2 = loadmushroom(path2)
''' Voting dataset '''
path3 = '/host/Users/vandana/Documents/Grad_Studies/Spring2012/Machine Learning/Projects/voting/house-votes-84.data'
mldata3 = loadvoting(path3)
''' Heart Disease dataset '''
path4 = '/host/Users/vandana/Documents/Grad_Studies/Spring2012/Machine Learning/Projects/heart_disease/processed.cleveland.data'
#path4 = '/host/Users/vandana/Documents/Grad_Studies/Spring2012/Machine Learning/Projects/heart_disease/processed.switzerland.data'
#mldata4 = loadheartdisease(path4, True, True)
mldata4 = loadheartdisease(path4, True)
normaldata4 = mldata4.normalizeinput()
''' Wine dataset '''
path5 = '/host/Users/vandana/Documents/Grad_Studies/Spring2012/Machine Learning/Projects/wine/wine.data'
mldata5 = loadwine(path5)
normaldata5 = mldata5.normalizeinput()
cvacc = {}
"""
cvacc['Iris'] = cv10foldresult(normaldata, [3], learn_rate=0.05, epoch=700, stop_criterion=50)
cvacc['Cars'] = cv10foldresult(mldata1, [5], learn_rate=0.01, epoch=1000, stop_criterion=100)
cvacc['Mushrm'] = cv10foldresult(mldata2, [5], learn_rate=0.05, epoch=50, stop_criterion=4)
cvacc['Voting'] = cv10foldresult(mldata3, [5], learn_rate=0.01, epoch=500, stop_criterion=100)
cvacc['Heart'] = cv10foldresult(normaldata4, [10, 5], learn_rate=0.01, epoch=1000, stop_criterion=100)
cvacc['Wine'] = cv10foldresult(normaldata5, [5], learn_rate=0.005, epoch=500, stop_criterion=50)
"""
#cvacc['Iris'] = cv10foldresult(normaldata, None, learn_rate=0.05, epoch=700, stop_criterion=50)
cvacc['Cars'] = cv10foldresult(mldata1, None, learn_rate=0.01, epoch=1000, stop_criterion=50)
cvacc['Mushrm'] = cv10foldresult(mldata2, None, learn_rate=0.08, epoch=50, stop_criterion=2)
#cvacc['Voting'] = cv10foldresult(mldata3, None, learn_rate=0.05, epoch=500, stop_criterion=100)
#cvacc['Heart'] = cv10foldresult(normaldata4, None, learn_rate=0.01, epoch=1000, stop_criterion=100)
#cvacc['Wine'] = cv10foldresult(normaldata5, None, learn_rate=0.005, epoch=500, stop_criterion=50)
for i in cvacc:
print i, ":", cvacc[i]
f = open("nn_cv", 'w')
pickle.dump(cvacc, f)
f.close()
def main(argv=None):
''' Iris dataset '''
path = '/host/Users/vandana/Documents/Grad_Studies/Spring2012/Machine Learning/Projects/iris/iris.data'
mldata = loadiris(path)
normaldata = mldata.normalizeinput()
sampledmldata = normaldata.sampledata({'train':70, 'test':15, 'cv':15})
#knn = NearestNeighbor(3, sampledmldata['train'])
#knn = NearestNeighbor(3, sampledmldata['train'], False, True, sampledmldata['cv'])
knn = NearestNeighbor(3, sampledmldata['train'], True)
knn.predict(sampledmldata['test'])
#knn.predict(sampledmldata['test'])
#knn.predictdistweighted(sampledmldata['test'])
#bestkexperiment('iris', normaldata)
''' Cars dataset '''
path1 = '/host/Users/vandana/Documents/Grad_Studies/Spring2012/Machine Learning/Projects/cars/car.data'
mldata1 = loadcars(path1)
sampledmldata1 = mldata1.sampledata({'train':70, 'test':15, 'cv':15})
#knn1 = NearestNeighbor(5, sampledmldata1['train'])
#knn1 = NearestNeighbor(5, sampledmldata1['train'], False, True, sampledmldata1['cv'])
knn1 = NearestNeighbor(5, sampledmldata1['train'], True)
knn1.predict(sampledmldata1['test'])
#bestkexperiment('cars', mldata1)
''' Mushroom dataset '''
path2 = '/host/Users/vandana/Documents/Grad_Studies/Spring2012/Machine Learning/Projects/mushroom/agaricus-lepiota.data'
mldata2 = loadmushroom(path2)
sampledmldata2 = mldata2.sampledata({'train':65, 'test':15, 'cv':20})
#knn2 = NearestNeighbor(1, sampledmldata2['train'])
#knn2 = NearestNeighbor(1, sampledmldata2['train'], False, True, sampledmldata2['cv'])
knn2 = NearestNeighbor(1, sampledmldata2['train'], True)
knn2.predict(sampledmldata2['test'])
#bestkexperiment('mushroom', mldata2)
''' Voting dataset '''
path3 = '/host/Users/vandana/Documents/Grad_Studies/Spring2012/Machine Learning/Projects/voting/house-votes-84.data'
mldata3 = loadvoting(path3)
sampledmldata3 = mldata3.sampledata({'train':70, 'test':15, 'cv':15})
#knn3 = NearestNeighbor(1, sampledmldata3['train'])
#knn3 = NearestNeighbor(1, sampledmldata3['train'], False, True, sampledmldata3['cv'])
knn3 = NearestNeighbor(5, sampledmldata3['train'], True)
knn3.predict(sampledmldata3['test'])
#bestkexperiment('voting', mldata3)
''' Heart Disease dataset '''
path4 = '/host/Users/vandana/Documents/Grad_Studies/Spring2012/Machine Learning/Projects/heart_disease/processed.cleveland.data'
mldata4 = loadheartdisease(path4, True)
normaldata4 = mldata4.normalizeinput()
sampledmldata4 = normaldata4.sampledata({'train':70, 'test':15, 'cv':15})
#knn4 = NearestNeighbor(5, sampledmldata4['train'])
#knn4 = NearestNeighbor(5, sampledmldata4['train'], False, True, sampledmldata4['cv'])
knn4 = NearestNeighbor(5, sampledmldata4['train'], True)
knn4.predict(sampledmldata4['test'])
#bestkexperiment('heart disease', normaldata4)
''' Wine dataset '''
path5 = '/host/Users/vandana/Documents/Grad_Studies/Spring2012/Machine Learning/Projects/wine/wine.data'
mldata5 = loadwine(path5)
normaldata5 = mldata5.normalizeinput()
sampledmldata5 = normaldata5.sampledata({'train':70, 'test':15, 'cv':15})
#knn5 = NearestNeighbor(3, sampledmldata5['train'])
#knn5 = NearestNeighbor(3, sampledmldata5['train'], False, True, sampledmldata5['cv'])
knn5 = NearestNeighbor(5, sampledmldata5['train'], True)
knn5.predict(sampledmldata5['test'])
def main(argv=None):
if argv is None:
argv = sys.argv
#toy dataset
#mldata = loadtennis()
#sampledmldata = mldata.sampledata({'train':80, 'test':20, 'cv':0})
''' Iris dataset '''
path = '/host/Users/vandana/Documents/Grad_Studies/Spring2012/Machine Learning/Projects/iris/iris.data'
mldata = loadiris(path)
sampledmldata = mldata.sampledata({'train': 70, 'test': 10, 'cv': 20})
dt = dtree(sampledmldata['train'])
dt.generatetree(dt.root)
print "before pruning...."
dt.printtree()
dt.testtree(sampledmldata['test'], False, True)
#print "after pruning...."
#besttree = dt.prune(2, sampledmldata['cv'], True)
#besttree.testtree(sampledmldata['test'], False, True)
#dt.prune(1, sampledmldata['cv'], True)
#dt.testtree(sampledmldata['test'], False, True)
''' Cars dataset '''
path1 = '/host/Users/vandana/Documents/Grad_Studies/Spring2012/Machine Learning/Projects/cars/car.data'
mldata1 = loadcars(path1)
sampledmldata1 = mldata1.sampledata({'train': 70, 'test': 10, 'cv': 20})
dt1 = dtree(sampledmldata1['train'])
dt1.generatetree(dt1.root)
dt1.printtree()
dt1.testtree(sampledmldata1['test'], False, True)
#print "after pruning...."
#besttree1 = dt1.prune(2, sampledmldata1['cv'], True)
#besttree1.testtree(sampledmldata1['test'], False, True)
#dt1.prune(1, sampledmldata1['cv'], True)
#dt1.testtree(sampledmldata1['test'], False, True)
''' Mushroom dataset '''
path2 = '/host/Users/vandana/Documents/Grad_Studies/Spring2012/Machine Learning/Projects/mushroom/agaricus-lepiota.data'
mldata2 = loadmushroom(path2)
sampledmldata2 = mldata2.sampledata({'train': 70, 'test': 10, 'cv': 20})
dt2 = dtree(sampledmldata2['train'])
dt2.generatetree(dt2.root)
print "before pruning...."
dt2.printtree()
dt2.testtree(sampledmldata2['test'], False, True)
#print "after pruning...."
#besttree2 = dt2.prune(2, sampledmldata2['cv'], True)
#besttree2.testtree(sampledmldata2['test'], False, True)
#dt2.prune(1, sampledmldata2['cv'], True)
#dt2.testtree(sampledmldata2['test'], False, True)
''' Voting dataset '''
path3 = '/host/Users/vandana/Documents/Grad_Studies/Spring2012/Machine Learning/Projects/voting/house-votes-84.data'
mldata3 = loadvoting(path3)
sampledmldata3 = mldata3.sampledata({'train': 70, 'test': 10, 'cv': 20})
dt3 = dtree(sampledmldata3['train'])
dt3.generatetree(dt3.root)
print "before pruning...."
dt3.printtree()
dt3.testtree(sampledmldata3['test'], False, True)
#print "after pruning...."
#besttree3 = dt3.prune(2, sampledmldata3['cv'], True)
#besttree3.testtree(sampledmldata3['test'], False, True)
#dt3.prune(1, sampledmldata3['cv'], True)
#dt3.testtree(sampledmldata3['test'], False, True)
''' Heart Disease dataset '''
path4 = '/host/Users/vandana/Documents/Grad_Studies/Spring2012/Machine Learning/Projects/heart_disease/processed.cleveland.data'
mldata4 = loadheartdisease(path4)
sampledmldata4 = mldata4.sampledata({'train': 70, 'test': 10, 'cv': 20})
dt4 = dtree(sampledmldata4['train'])
dt4.generatetree(dt4.root)
print "before pruning...."
dt4.printtree()
dt4.testtree(sampledmldata4['test'], False, True)
#print "after pruning...."
#besttree4 = dt4.prune(2, sampledmldata4['cv'], True)
#besttree4.testtree(sampledmldata4['test'], False, True)
#dt4.prune(1, sampledmldata4['cv'], True)
#dt4.testtree(sampledmldata4['test'], False, True)
return 0
def main(argv=None):
if argv is None:
argv = sys.argv
#toy dataset
#mldata = loadtennis()
#sampledmldata = mldata.sampledata({'train':80, 'test':20, 'cv':0})
''' Iris dataset '''
path = '/host/Users/vandana/Documents/Grad_Studies/Spring2012/Machine Learning/Projects/iris/iris.data'
mldata = loadiris(path)
sampledmldata = mldata.sampledata({'train':70, 'test':10, 'cv':20})
dt = dtree(sampledmldata['train'])
dt.generatetree(dt.root)
print "before pruning...."
dt.printtree()
dt.testtree(sampledmldata['test'], False, True)
#print "after pruning...."
#besttree = dt.prune(2, sampledmldata['cv'], True)
#besttree.testtree(sampledmldata['test'], False, True)
#dt.prune(1, sampledmldata['cv'], True)
#dt.testtree(sampledmldata['test'], False, True)
''' Cars dataset '''
path1 = '/host/Users/vandana/Documents/Grad_Studies/Spring2012/Machine Learning/Projects/cars/car.data'
mldata1 = loadcars(path1)
sampledmldata1 = mldata1.sampledata({'train':70, 'test':10, 'cv':20})
dt1 = dtree(sampledmldata1['train'])
dt1.generatetree(dt1.root)
dt1.printtree()
dt1.testtree(sampledmldata1['test'], False, True)
#print "after pruning...."
#besttree1 = dt1.prune(2, sampledmldata1['cv'], True)
#besttree1.testtree(sampledmldata1['test'], False, True)
#dt1.prune(1, sampledmldata1['cv'], True)
#dt1.testtree(sampledmldata1['test'], False, True)
''' Mushroom dataset '''
path2 = '/host/Users/vandana/Documents/Grad_Studies/Spring2012/Machine Learning/Projects/mushroom/agaricus-lepiota.data'
mldata2 = loadmushroom(path2)
sampledmldata2 = mldata2.sampledata({'train':70, 'test':10, 'cv':20})
dt2 = dtree(sampledmldata2['train'])
dt2.generatetree(dt2.root)
print "before pruning...."
dt2.printtree()
dt2.testtree(sampledmldata2['test'], False, True)
#print "after pruning...."
#besttree2 = dt2.prune(2, sampledmldata2['cv'], True)
#besttree2.testtree(sampledmldata2['test'], False, True)
#dt2.prune(1, sampledmldata2['cv'], True)
#dt2.testtree(sampledmldata2['test'], False, True)
''' Voting dataset '''
path3 = '/host/Users/vandana/Documents/Grad_Studies/Spring2012/Machine Learning/Projects/voting/house-votes-84.data'
mldata3 = loadvoting(path3)
sampledmldata3 = mldata3.sampledata({'train':70, 'test':10, 'cv':20})
dt3 = dtree(sampledmldata3['train'])
dt3.generatetree(dt3.root)
print "before pruning...."
dt3.printtree()
dt3.testtree(sampledmldata3['test'], False, True)
#print "after pruning...."
#besttree3 = dt3.prune(2, sampledmldata3['cv'], True)
#besttree3.testtree(sampledmldata3['test'], False, True)
#dt3.prune(1, sampledmldata3['cv'], True)
#dt3.testtree(sampledmldata3['test'], False, True)
''' Heart Disease dataset '''
path4 = '/host/Users/vandana/Documents/Grad_Studies/Spring2012/Machine Learning/Projects/heart_disease/processed.cleveland.data'
mldata4 = loadheartdisease(path4)
sampledmldata4 = mldata4.sampledata({'train':70, 'test':10, 'cv':20})
dt4 = dtree(sampledmldata4['train'])
dt4.generatetree(dt4.root)
print "before pruning...."
dt4.printtree()
dt4.testtree(sampledmldata4['test'], False, True)
#print "after pruning...."
#besttree4 = dt4.prune(2, sampledmldata4['cv'], True)
#besttree4.testtree(sampledmldata4['test'], False, True)
#dt4.prune(1, sampledmldata4['cv'], True)
#dt4.testtree(sampledmldata4['test'], False, True)
return 0
def main(argv=None):
''' Iris dataset '''
path = '/host/Users/vandana/Documents/Grad_Studies/Spring2012/Machine Learning/Projects/iris/iris.data'
mldata = loadiris(path)
normaldata = mldata.normalizeinput()
sampledmldata = normaldata.sampledata({'train': 70, 'test': 15, 'cv': 15})
#knn = NearestNeighbor(3, sampledmldata['train'])
#knn = NearestNeighbor(3, sampledmldata['train'], False, True, sampledmldata['cv'])
knn = NearestNeighbor(3, sampledmldata['train'], True)
knn.predict(sampledmldata['test'])
#knn.predict(sampledmldata['test'])
#knn.predictdistweighted(sampledmldata['test'])
#bestkexperiment('iris', normaldata)
''' Cars dataset '''
path1 = '/host/Users/vandana/Documents/Grad_Studies/Spring2012/Machine Learning/Projects/cars/car.data'
mldata1 = loadcars(path1)
sampledmldata1 = mldata1.sampledata({'train': 70, 'test': 15, 'cv': 15})
#knn1 = NearestNeighbor(5, sampledmldata1['train'])
#knn1 = NearestNeighbor(5, sampledmldata1['train'], False, True, sampledmldata1['cv'])
knn1 = NearestNeighbor(5, sampledmldata1['train'], True)
knn1.predict(sampledmldata1['test'])
#bestkexperiment('cars', mldata1)
''' Mushroom dataset '''
path2 = '/host/Users/vandana/Documents/Grad_Studies/Spring2012/Machine Learning/Projects/mushroom/agaricus-lepiota.data'
mldata2 = loadmushroom(path2)
sampledmldata2 = mldata2.sampledata({'train': 65, 'test': 15, 'cv': 20})
#knn2 = NearestNeighbor(1, sampledmldata2['train'])
#knn2 = NearestNeighbor(1, sampledmldata2['train'], False, True, sampledmldata2['cv'])
knn2 = NearestNeighbor(1, sampledmldata2['train'], True)
knn2.predict(sampledmldata2['test'])
#bestkexperiment('mushroom', mldata2)
''' Voting dataset '''
path3 = '/host/Users/vandana/Documents/Grad_Studies/Spring2012/Machine Learning/Projects/voting/house-votes-84.data'
mldata3 = loadvoting(path3)
sampledmldata3 = mldata3.sampledata({'train': 70, 'test': 15, 'cv': 15})
#knn3 = NearestNeighbor(1, sampledmldata3['train'])
#knn3 = NearestNeighbor(1, sampledmldata3['train'], False, True, sampledmldata3['cv'])
knn3 = NearestNeighbor(5, sampledmldata3['train'], True)
knn3.predict(sampledmldata3['test'])
#bestkexperiment('voting', mldata3)
''' Heart Disease dataset '''
path4 = '/host/Users/vandana/Documents/Grad_Studies/Spring2012/Machine Learning/Projects/heart_disease/processed.cleveland.data'
mldata4 = loadheartdisease(path4, True)
normaldata4 = mldata4.normalizeinput()
sampledmldata4 = normaldata4.sampledata({
'train': 70,
'test': 15,
'cv': 15
})
#knn4 = NearestNeighbor(5, sampledmldata4['train'])
#knn4 = NearestNeighbor(5, sampledmldata4['train'], False, True, sampledmldata4['cv'])
knn4 = NearestNeighbor(5, sampledmldata4['train'], True)
knn4.predict(sampledmldata4['test'])
#bestkexperiment('heart disease', normaldata4)
''' Wine dataset '''
path5 = '/host/Users/vandana/Documents/Grad_Studies/Spring2012/Machine Learning/Projects/wine/wine.data'
mldata5 = loadwine(path5)
normaldata5 = mldata5.normalizeinput()
sampledmldata5 = normaldata5.sampledata({
'train': 70,
'test': 15,
'cv': 15
})
#knn5 = NearestNeighbor(3, sampledmldata5['train'])
#knn5 = NearestNeighbor(3, sampledmldata5['train'], False, True, sampledmldata5['cv'])
knn5 = NearestNeighbor(5, sampledmldata5['train'], True)
knn5.predict(sampledmldata5['test'])