def cross_Valid(egs, numCross, B):
crossData = ut.dataCrossSplit(egs, numCross, False)
fold = 0
errors = {}
for x in B:
errors[x] = [0] * numCross
for i in range(numCross):
data = crossData[i]
print "# Fold", fold
for b in B:
print " ", b, "Bases:",
err = calcError(data[0], data[1], b)
print err
errors[b][fold] = err
fold = fold + 1
return errors
def cross_vad(examples, T, num_folds=10):
data = ut.dataCrossSplit(examples, num_folds, False)
errorRates = []
for i in range(num_folds):
egs = data[i]
classifier = LogitBoost(egs[0], T)
# calculate error rate
error = [0.0] * 2
for j in range(2):
for x in egs[j]:
if classifier(x) != x[0]:
error[j] = error[j] + 1
error[j] = error[j] / len(egs[j])
print "Fold ", i, " trainingData errorRate: ", error[0], " testData errorRate:", error[1]
errorRates.append(error)
arr = np.array(errorRates)
print "Train Mean ErrorRate:", np.mean(arr[:, 0]), " Test Mean ErrorRate:", np.mean(arr[:, 1])
print "Train StdVar ErrorRate:", np.sqrt(np.var(arr[:, 0])), " Test Mean ErrorRate:", np.sqrt(np.var(arr[:, 1]))
def cross_vad(examples, num_folds = 10):
data = ut.dataCrossSplit(examples, num_folds, False)
errorRates = []
for i in range(num_folds):
egs = data[i]
dt = DTree(SelectAtt)
dt.training(egs[0], 1)
# calculate error rate
error = [0.] * 2
for j in range(2):
for x in egs[j]:
if dt.predict(x) != x[0]:
error[j] = error[j] + 1
error[j] = error[j] / len(egs[j])
print "Fold ", i, " trainingData errorRate: ", error[0], " testData errorRate:", error[1]
errorRates.append(error)
arr = np.array(errorRates)
print "Train Mean ErrorRate:", np.mean(arr[:,0]), " Test Mean ErrorRate:", np.mean(arr[:,1])
print "Train StdVar ErrorRate:", np.sqrt(np.var(arr[:,0])), " Test Mean ErrorRate:", np.sqrt(np.var(arr[:, 1]))
# -*- coding: utf-8 -*-
"""
Created on Wed Nov 20 22:55:53 2013
@author: jiecaoc
"""
import utilities as ut
import classes as cls
egs = ut.importRawData('Ionosphere.csv')
raw = ut.dataCrossSplit(egs, 2, False)
dt = cls.DTree(cls.SelectAtt)
dt.training(raw[0][0])
c = 0
for eg in raw[0][1]:
if eg[0] != dt.predict(eg):
# print eg[0], dt.predict(eg)
c = c + 1
print (c + .0) / len(egs)
for k in range(1, len(crossData[i][1]) + 1):
for x in crossData[i][1][k]:
totNum = totNum + 1
if cl.classify(x) != k:
totError = totError + 1
return (totError + 0.) / totNum
# check input
if len(sys.argv) != 3:
print "input error"
exit()
fileName = sys.argv[1]
if fileName == "Iris.csv":
subDim = 3
else:
subDim = 9
num_cross = int(sys.argv[2])
# import data
# since the date in the file are in order
# need to make it random to run cross validation
rawData = ut.makeDataRandom(ut.importRawData(fileName))
rawData = ut.importRawData(fileName)
trainData = ut.dataCrossSplit(rawData, num_cross)
#print tmp[1]
print "Data: ", fileName
print "Error rate for cross_validation:", cross_validation(trainData, subDim)
def leastSquare(trainData, K = 3):
"""
give training data,
return f(x), defined as f(x) = W'x
"""
# construct Y
Y = np.array([classEncode(x[0]) for x in trainData])
# construct X
X = np.array([ x[1:] for x in trainData])
X_ = np.dot( np.linalg.inv( np.dot(X.transpose(), X) ), X.transpose() )
W_T = np.dot(X_, Y).transpose()
return lambda x: np.dot(W_T, x)
# check input
if len(sys.argv) != 3:
print "input error"
exit()
fileName = sys.argv[1]
num_cross = int(sys.argv[2])
rawData = ut.makeDataRandom(ut.importRawData(fileName))
crossData = ut.dataCrossSplit(rawData, num_cross, False)
print "Data: ", fileName
print "Error rate for cross_validation:", cross_validation(crossData)