def c45(X_train, y_train, X_test, y_test):
pyC45.train(X_train, y_train, "DecisionTree.xml")
# test the C45 decision tree
answer = []
testing_obs = []
for index, row in y_test.iteritems():
# testing_obs.append(row[:-1].tolist())
answer.append(str(row))
prediction = pyC45.predict("DecisionTree.xml", X_test)
return answer, prediction
def trainandpredict(self, trainX, trainY, testX, testY):
startTime = datetime.now()
pyC45.train(trainX, trainY, "DecisionTree.xml")
start_time = datetime.now()
# test the C45 decision tree
answer = []
testing_obs = []
for index, row in testY.iteritems():
answer.append(str(row))
startTime = datetime.now()
prediction = pyC45.predict("DecisionTree.xml", testX)
predictionTime = datetime.now()
dt = datetime.now() - start_time
ms = (dt.days * 24 * 60 * 60 + dt.seconds) * 1000 + dt.microseconds / 1000.0
print ('Time taken by this algo in millisec:' + str(ms))
return answer, prediction
import pyC45, csv
if __name__ == "__main__":
#train a C45 decision tree and save the tree as an XML file
reader = csv.reader(file('./data/training_set.csv'))
training_obs = []
training_cat = []
for line in reader:
training_obs.append(line[:-1])
training_cat.append(line[-1])
pyC45.train(training_obs, training_cat, "DecisionTree.xml")
#test the C45 decision tree
reader = csv.reader(file('./data/training_set.csv'))
answer = []
testing_obs = []
for line in reader:
testing_obs.append(line[:-1])
answer.append(line[-1])
answer.pop(0)
prediction = pyC45.predict("DecisionTree.xml", testing_obs)
err = 0
for i in range(len(answer)):
if not answer[i] == prediction[i]:
err = err + 1
print "error rate=", round(float(err) / len(prediction) * 100, 2), "%"
training_obs = []
training_cat = []
for line in reader:
training_obs.append(line[:-1])
training_cat.append(line[-1])
pyC45.train(
training_obs, training_cat,
"C:/Users/Luiz Felipe/Documents/TCC/assistenteX/plugin/dataDecisionTree.xml"
)
#test the C45 decision tree
reader = csv.reader(
open(
'C:/Users/Luiz Felipe/Documents/TCC/assistenteX/plugin/data/training_set.csv'
))
answer = []
testing_obs = []
for line in reader:
testing_obs.append(line[:-1])
answer.append(line[-1])
answer.pop(0)
prediction = pyC45.predict(
"C:/Users/Luiz Felipe/Documents/TCC/assistenteX/plugin/dataDecisionTree.xml",
testing_obs)
err = 0
for i in range(len(answer)):
if not answer[i] == prediction[i]:
err = err + 1
print("error rate=", round(float(err) / len(prediction) * 100, 2), "%")
import pyC45,csv
if __name__=="__main__":
#train a C45 decision tree and save the tree as an XML file
reader = csv.reader(file('./data/training_set.csv'))
training_obs=[]
training_cat=[]
for line in reader:
training_obs.append(line[:-1])
training_cat.append(line[-1])
pyC45.train(training_obs,training_cat,"DecisionTree.xml")
#test the C45 decision tree
reader = csv.reader(file('./data/training_set.csv'))
answer=[]
testing_obs=[]
for line in reader:
testing_obs.append(line[:-1])
answer.append(line[-1])
answer.pop(0)
prediction=pyC45.predict("DecisionTree.xml",testing_obs)
err=0
for i in range(len(answer)):
if not answer[i]==prediction[i]:
err=err+1
print "error rate=",round(float(err)/len(prediction)*100,2),"%"