def main():
gnb = GNB()
with open('train.json', 'rb') as f:
j = json.load(f)
#print j.keys()
X = j['states']
Y = j['labels']
#print len(X), len(Y)
print "-----start training-------"
gnb.train(X, Y)
with open('test.json', 'rb') as f:
j = json.load(f)
X = j['states']
Y = j['labels']
score = 0
for coords, label in zip(X,Y):
predicted = gnb.predict(coords)
if predicted == label:
score += 1
fraction_correct = float(score) / len(X)
print "You got {} percent correct".format(100 * fraction_correct)
def person_test():
# Create an empty dataframe
data = pd.DataFrame()
# Create our target variable
data['Gender'] = [
'male', 'male', 'male', 'male', 'female', 'female', 'female', 'female'
]
# Create our feature variables
data['Height'] = [6, 5.92, 5.58, 5.92, 5, 5.5, 5.42, 5.75]
data['Weight'] = [180, 190, 170, 165, 100, 150, 130, 150]
data['Foot_Size'] = [12, 11, 12, 10, 6, 8, 7, 9]
# Create an empty dataframe
person = pd.DataFrame()
# Create some feature values for this single row
person['Height'] = [6]
person['Weight'] = [130]
person['Foot_Size'] = [8]
gnb = GNB()
gnb.train(data, "Gender")
for idx, row in person.iterrows():
prob_label, prob = gnb.predict(row)
print("Probability Label: %s --> Probability: %s" % (prob_label, prob))
def main():
gnb = GNB()
with open('train.json', 'r') as f:
j = json.load(f)
print(j.keys())
X = j['states']
Y = j['labels']
gnb.train(X, Y)
skl_nb = GaussianNB()
LE = LabelEncoder()
skl_nb.fit(X, LE.fit_transform(Y))
with open('test.json', 'r') as f:
j = json.load(f)
X = j['states']
Y = j['labels']
score = 0
for coords, label in zip(X, Y):
predicted = gnb.predict(coords)
if predicted == label:
score += 1
fraction_correct = float(score) / len(X)
print("You got {0:3.2f} percent correct".format(100 * fraction_correct))
Y_test = LE.transform(Y)
Y_pred = skl_nb.predict(X)
skl_fraction_correct = sum([1 for a, b in zip(Y_test, Y_pred) if a == b
]) / float(len(Y_test))
print("Sklearn got {0:3.2f} percent correct".format(100 *
skl_fraction_correct))
def main():
# Load training data
with open('train.json', 'r') as f:
j = json.load(f)
X = j['states']
Y = j['labels']
# Train classifier
gnb = GNB()
gnb.train(X, Y)
# Open test data
with open('test.json', 'r') as f:
j = json.load(f)
X = j['states']
Y = j['labels']
# Evaluate classifier
score = 0
for coords, label in zip(X, Y):
predicted = gnb.predict(coords)
if predicted == label:
score += 1
fraction_correct = float(score) / len(X)
print("You got {} percent correct".format(100 * fraction_correct))
def main():
gnb = GNB()
with open('train.json', encoding='utf-8') as f:
j = json.load(f)
print(j.keys())
X = j['states']
Y = j['labels']
gnb.train(X, Y)
with open('test.json', encoding='utf-8') as f:
j = json.load(f)
X = j['states']
Y = j['labels']
score = 0
for coords, label in zip(X,Y):
predicted = gnb.predict(coords)
if predicted == label:
score += 1
percent_correct = 100 * float(score) / len(X)
print(str(percent_correct) + " percent correct.")
def main():
gnb = GNB()
train_df = convert_to_labels(pd.read_json('train.json'))
test_df = convert_to_labels(pd.read_json('test.json'))
gnb.train(train_df, 'labels')
for idx, row in test_df.iterrows():
prediction_label, prob = gnb.predict(row)
print("Predicted: %s Actual: %s" % (prediction_label, row['labels']))
def main():
gnb = GNB()
with open('train.json', 'r') as f:
j = json.load(f)
X = j['states']
Y = j['labels']
gnb.train(X, Y)
with open('test.json', 'r') as f:
j = json.load(f)
X = j['states']
Y = j['labels']
score = 0
for coords, label in zip(X, Y):
predicted = gnb.predict(coords)
if predicted == label:
score += 1
fraction_correct = float(score) / len(X)
print("You got %.2f percent correct" % (100 * fraction_correct))
def main():
# TRAINING
gnb = GNB()
with open('train.json', 'r') as f:
j = json.load(f)
#print(j.keys())
'''
X - array of N observations
- Each observation is a tuple with 4 values: s, d,
s_dot and d_dot.
'''
X = j['states']
X = np.array(X)
Y = j['labels']
Y = np.array(Y)
# d-values seem biased where 0 is the center of the left lane.
X[:, 1] = X[:, 1] + 2
# just feed delta d to classifier
#X = np.array([X[:,3]]).T
# feed d and delta_d into classifier
X = np.vstack((X[:, 2], X[:, 3])).T
gnb.train(X, Y)
# CLASSIFICATION
with open('test.json', 'r') as f:
j = json.load(f)
X = j['states']
X = np.array(X)
Y = j['labels']
Y = np.array(Y)
X[:, 1] = X[:, 1] + 2
#X = np.array([X[:,3]]).T
X = np.vstack((X[:, 2], X[:, 3])).T
score = 0
for coords, label in zip(X, Y):
predicted = gnb.predict(coords)
if predicted == label:
score += 1
fraction_correct = float(score) / len(X)
print("You got {} percent correct".format(100 * fraction_correct))
def main():
clf = GNB()
with open('train.json', encoding='utf-8') as f:
j = json.load(f)
print(j.keys())
X = j['states']
Y = j['labels']
clf.train(X, Y)
with open('test.json', encoding='utf-8') as f:
j = json.load(f)
X = j['states']
Y = j['labels']
score = 0
for coords, label in zip(X, Y):
predicted = clf.predict(coords)
if predicted == label:
score += 1
fraction_correct = float(score) / len(X)
print("You got {} percent correct".format(100 * fraction_correct))
def main():
gnb = GNB()
with open('train.json', 'rb') as f:
j = json.load(f)
print("JSON file keys =", j.keys())
X = j['states']
Y = j['labels']
gnb.train(X, Y)
with open('test.json', 'rb') as f:
j = json.load(f)
X = j['states']
Y = j['labels']
score = 0
for coords, label in zip(X,Y):
predicted = gnb.predict(coords)
print('pred=',predicted,' label=',label)
if predicted == label:
score += 1
fraction_correct = float(score) / len(X)
print("You got {} percent correct".format(100 * fraction_correct))
