def run(config, codeRoot=''):
classifier = DecisionTreeClassifier([0, 1])
print " ========= %s ========= " % config['header']
pointsPerCase = int(config['pointsPerCase'])
args = {}
args['metric'] = metrics.error
args['maxdepth'] = 1
size = len(config['dataset'])
points = 0
first_error = ''
for i in xrange(size):
data = config['dataset'][i]
print "\tDataset: %s" % data
# Load data
trainingData = samples.loadDataFile(
os.path.join(codeRoot, "data/%s/training_data.csv" % data))
trainingLabels = samples.loadLabelsFile(
os.path.join(codeRoot, "data/%s/training_labels.csv" % data))
# Conduct training and testing
classifier.train(trainingData, trainingLabels, args)
tree = classifier.tree
best_split = int(config['best_split'][i])
test_passed = True
if tree.column != best_split:
error = 'WRONG SPLIT: Correct: %s , Returned: %s' % (best_split,
tree.column)
if not first_error: first_error = error
print '\t\t%s' % error
test_passed = False
if not has_one_level(tree):
error = 'The tree does not have one level'
if not first_error: first_error = error
print '\t\t%s' % error
test_passed = False
if test_passed:
points += pointsPerCase
print '\t\tOK'
needed_all_tests_passed = int(config['totalQuestion']) > 0
if needed_all_tests_passed and first_error:
points = 0
test_error = 'Your code does not pass all tests'
else:
test_error = ''
return (points, test_error)
def run( config , codeRoot = '' ) : classifier = DecisionTreeClassifier( [ 0 , 1 ] ) print " ========= %s ========= " % config[ 'header' ] pointsPerCase = int( config[ 'pointsPerCase' ] ) num_cases = len( config[ 'case' ] ) points = 0 first_error = '' for metric in config[ 'metric' ] : mymet = get_metric( metric ) print "\tMetric: %s" % metric results = [ float( val ) for val in config[ 'results_%s' % metric ] ] for i in xrange( num_cases ) : data = eval( config[ 'case' ][ i ] ) print "\t\tCase: %s" % data resp = mymet( data ) if compare( resp , results[ i ] ) == 0 : points += pointsPerCase print '\t\t\tOK' else : error = 'ERROR: Correct: %s, Returned: %s' % ( results[ i ] , resp ) print '\t\t\t%s' % error if not first_error : first_error = error # Check if need to pass all tests needed_all_tests_passed = int( config[ 'totalQuestion' ] ) > 0 if needed_all_tests_passed and first_error : points = 0 test_error = 'Your code does not pass all tests' else : test_error = '' return ( points , test_error )
def run(config, codeRoot=''):
classifier = DecisionTreeClassifier([0, 1])
print " ========= %s ========= " % config['header']
pointsPerCase = int(config['pointsPerCase'])
args = {}
args['metric'] = get_metric(config['metric'])
size = len(config['dataset'])
points = 0
first_error = ''
for i in xrange(size):
data = config['dataset'][i]
print "\tDataset: %s" % data
# Load data
trainingData = samples.loadDataFile(
os.path.join(codeRoot, "data/%s/training_data.csv" % data))
trainingLabels = samples.loadLabelsFile(
os.path.join(codeRoot, "data/%s/training_labels.csv" % data))
# Load test
testData = samples.loadDataFile(
os.path.join(codeRoot, "data/%s/test_data.csv" % data))
testLabels = samples.loadLabelsFile(
os.path.join(codeRoot, "data/%s/test_labels.csv" % data))
# Conduct training and testing
args['maxdepth'] = int(config['maxdepth'][i])
classifier.train(trainingData, trainingLabels, args)
guesses = classifier.classify(testData)
min_accuracy_required = float(config['accuracy'][i])
if is_good_classifier(guesses, testLabels, min_accuracy_required):
print '\t\tOK'
points += pointsPerCase
else:
points += (pointsPerCase / 2.)
print '\t\tMost frequent classifier is better'
needed_all_tests_passed = int(config['totalQuestion']) > 0
if needed_all_tests_passed and first_error:
points = 0
test_error = 'Your code does not pass all tests'
else:
test_error = ''
return (points, test_error)
def run(config, codeRoot=''):
classifier = DecisionTreeClassifier([0, 1])
print " ========= %s ========= " % config['header']
args = {}
args['metric'] = metrics.error
args['maxdepth'] = 0
size = len(config['dataset'])
points = 0
first_error = ''
for i in xrange(size):
data = config['dataset'][i]
print "\tDataset: %s" % data
# Load data
trainingData = samples.loadDataFile(
os.path.join(codeRoot, "data/%s/training_data.csv" % data))
trainingLabels = samples.loadLabelsFile(
os.path.join(codeRoot, "data/%s/training_labels.csv" % data))
# Conduct training and testing
classifier.train(trainingData, trainingLabels, args)
tree = classifier.tree
if not tree.leftchild and not tree.rightchild:
label = int(config['label'][i])
if tree.label != label:
error = 'The tree is not classifying by the most frequent label: %s' % label
if not first_error: first_error = error
print '\t\t%s' % error
else:
points += int(config['pointsPerCase'])
print '\t\tOK'
else:
error = 'The tree has one or more levels'
if not first_error: first_error = error
print '\t\t%s' % error
needed_all_tests_passed = int(config['totalQuestion']) > 0
if needed_all_tests_passed and first_error:
points = 0
test_error = 'Your code does not pass all tests'
else:
test_error = ''
return (points, test_error)
def run(config, codeRoot=''):
classifier = DecisionTreeClassifier([0, 1])
print " ========= %s ========= " % config['header']
pointsPerCase = int(config['pointsPerCase'])
args = {}
args['metric'] = get_metric(config['metric'])
args['maxdepth'] = int(config['maxdepth'])
size = len(config['dataset'])
points = 0
first_error = ''
for i in xrange(size):
data = config['dataset'][i]
print "\tDataset: %s" % data
# Load data
trainingData = samples.loadDataFile(
os.path.join(codeRoot, "data/%s/training_data.csv" % data))
trainingLabels = samples.loadLabelsFile(
os.path.join(codeRoot, "data/%s/training_labels.csv" % data))
# Conduct training and testing
classifier.train(trainingData, trainingLabels, args)
tree = classifier.tree
status, error = compare(tree, get_tree(data, codeRoot))
if status != 0:
if not first_error: first_error = error
print '\t\t%s' % error
else:
points += pointsPerCase
print '\t\tOK'
needed_all_tests_passed = int(config['totalQuestion']) > 0
if needed_all_tests_passed and first_error:
points = 0
test_error = 'Your code does not pass all tests'
else:
test_error = ''
return (points, test_error)
def run(config, codeRoot=''):
classifier = DecisionTreeClassifier([0, 1])
print " ========= %s ========= " % config['header']
pointsPerCase = int(config['pointsPerCase'])
size = len(config['data'])
points = 0
first_error = ''
''' Check if it is empty '''
data = eval(config['data'][0])
labels = eval(config['label'][0])
maxdepth = int(config['maxdepth'][0])
classifier.maxdepth = maxdepth
print '\tCase #1: DATA(%s), LABELS(%s)' % (data, labels)
if not classifier.isLeaf(data, labels, 2):
error = 'WRONG: Check if dataset is empty'
if not first_error: first_error = error
print '\t\t%s' % error
else:
points += pointsPerCase
print '\t\tOK'
''' Check if all instances have same label '''
data = eval(config['data'][1])
labels = eval(config['label'][1])
maxdepth = int(config['maxdepth'][1])
classifier.maxdepth = maxdepth
print '\tCase #2: DATA(%s), LABELS(%s)' % (data, labels)
if not classifier.isLeaf(data, labels, 2):
error = 'WRONG: Check if all instances have same label'
if not first_error: first_error = error
print '\t\t%s' % error
else:
points += pointsPerCase
print '\t\tOK'
''' Check if all instances are the same '''
data = eval(config['data'][2])
labels = eval(config['label'][2])
maxdepth = int(config['maxdepth'][2])
classifier.maxdepth = maxdepth
print '\tCase #3: DATA(%s), LABELS(%s)' % (data, labels)
if not classifier.isLeaf(data, labels, 2):
error = 'WRONG: Check if all instances are the same'
if not first_error: first_error = error
print '\t\t%s' % error
else:
points += pointsPerCase
print '\t\tOK'
''' Check maximum depth '''
data = eval(config['data'][3])
labels = eval(config['label'][3])
maxdepth = int(config['maxdepth'][3])
classifier.maxdepth = maxdepth
print '\tCase #4: DATA(%s), LABELS(%s)' % (data, labels)
if not classifier.isLeaf(data, labels, 3):
error = 'WRONG: Check maximum depth parameter. Maximum: %s, Current: %s' % (
maxdepth, 3)
if not first_error: first_error = error
print '\t\t%s' % error
else:
points += pointsPerCase
print '\t\tOK'
needed_all_tests_passed = int(config['totalQuestion']) > 0
if needed_all_tests_passed and first_error:
points = 0
test_error = 'Your code does not pass all tests'
else:
test_error = ''
return (points, test_error)
from sklearn.datasets import load_iris
from sklearn.model_selection import train_test_split
from decisionTree import DecisionTreeClassifier
from pprint import pprint
import numpy as np
def accuracy(y_true, y_pred):
accuracy = np.sum(y_true == y_pred) / len(y_true)
return accuracy
iris = load_iris()
X = iris.data
y = iris.target
X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=0.1)
clf = DecisionTreeClassifier(max_depth=7, feature_names=iris.feature_names)
m = clf.fit(X_train, y_train)
pprint(m)
predictions = clf.predict(X_test)
print('Test accuracy:', accuracy(y_test, predictions))