clf_5.fit(Xtrain, ytrain_casual) print 'Finished fitting' dt_casual = clf_1.predict(Xtest) ada_casual = clf_2.predict(Xtest) grad_casual = clf_3.predict(Xtest) rf_casual = clf_4.predict(Xtest) et_casual = clf_5.predict(Xtest) feature_imps = clf_4.feature_importances_ print "regular decision tree" print rmsle(ytest, dt_regular + dt_casual) print "boosted decision tree" print rmsle(ytest, ada_regular + ada_casual) print "gradient tree boosting" print rmsle(ytest, grad_regular + grad_casual) print "random forest classifier" print rmsle(ytest, rf_regular + rf_casual) print "extra trees classifier" print rmsle(ytest, et_casual + et_regular) print "feature importances" print feature_imps if __name__ == '__main__': (X,y1,y2,y3) = import_training_file(sys.argv[1], True) decision_tree(X, y1, y2, y3)
from sklearn import datasets
from sklearn.svm import SVC
from sklearn.svm import SVR
from sklearn.svm import LinearSVC
from sklearn.metrics import accuracy_score
from datetime import datetime
from import_train import rmsle
from import_train import import_training_file
from sklearn.grid_search import GridSearchCV
from sklearn.metrics import classification_report
from sklearn import preprocessing as pre
from scipy import sparse
import sys
if __name__ == '__main__':
(X, y_total, y_regis, y_casual) = import_training_file(sys.argv[1], True)
n,d = X.shape
nTrain = 0.5*n
Xtrain = X[:nTrain,:]
y_casual_train = y_casual[:nTrain]
y_regis_train = y_regis[:nTrain]
y_total_train = y_total[:nTrain]
Xtest = X[nTrain:,:]
y_casual_test = y_casual[nTrain:]
y_regis_test = y_regis[nTrain:]
y_total_test = y_total[nTrain:]
#linear
#param_grid = {'C': [1, 5, 10, 100],}
from sklearn import datasets
from sklearn.svm import SVC
from sklearn.svm import LinearSVC
from sklearn.svm import SVR
from sklearn.metrics import accuracy_score
from datetime import datetime
from import_train import rmsle
from import_train import import_training_file
from sklearn.grid_search import GridSearchCV
from sklearn.metrics import classification_report
from sklearn import preprocessing as pre
from scipy import sparse
import sys
if __name__ == '__main__':
(X, y_total, y_regis, y_casual) = import_training_file(sys.argv[1], True)
n, d = X.shape
nTrain = 0.5 * n
# shuffle the data
#idx = np.arange(n)
#np.random.seed(42)
#np.random.shuffle(idx)
#X = X[idx]
#y = y[idx]
Xtrain = X[:nTrain, :]
y_casual_train = y_casual[:nTrain]
y_regis_train = y_regis[:nTrain]
y_total_train = y_total[:nTrain]
clf_4.fit(Xtrain, ytrain_casual)
clf_5.fit(Xtrain, ytrain_casual)
print 'Finished fitting'
dt_casual = clf_1.predict(Xtest)
ada_casual = clf_2.predict(Xtest)
grad_casual = clf_3.predict(Xtest)
rf_casual = clf_4.predict(Xtest)
et_casual = clf_5.predict(Xtest)
feature_imps = clf_4.feature_importances_
print "regular decision tree"
print rmsle(ytest, dt_regular + dt_casual)
print "boosted decision tree"
print rmsle(ytest, ada_regular + ada_casual)
print "gradient tree boosting"
print rmsle(ytest, grad_regular + grad_casual)
print "random forest classifier"
print rmsle(ytest, rf_regular + rf_casual)
print "extra trees classifier"
print rmsle(ytest, et_casual + et_regular)
print "feature importances"
print feature_imps
if __name__ == '__main__':
(X, y1, y2, y3) = import_training_file(sys.argv[1], True)
decision_tree(X, y1, y2, y3)
train_pred - train_given,
c='b',
s=40,
alpha=0.25,
label='Training Residuals')
plt.plot([0, 600], [0, 0], c='r')
plt.xlim(0, 600)
plt.legend()
plt.title('Residuals vs Bike Share Count')
plt.xlabel("Predicted count values")
plt.ylabel("Residuals")
plt.show()
if __name__ == '__main__':
X_train, y_total, y_reg, y_casual = import_training_file(sys.argv[1])
X_test, datetime = import_testing_file(sys.argv[2])
n, _ = X_train.shape
nTrain = int(0.5 * n) #training on 50% of the data
sub_Xtrain = X_train[:nTrain, :]
sub_ytrain = y_total[:nTrain]
sub_ytrain_registered = y_reg[:nTrain]
sub_ytest_registered = y_reg[nTrain:]
sub_ytrain_casual = y_casual[:nTrain]
sub_ytest_casual = y_casual[nTrain:]
sub_Xtest = X_train[nTrain:, :]
sub_ytest = y_total[nTrain:]
rf_opt = RandomForestRegressor(bootstrap=True,
compute_importances=None,
if row[5] >= 80:
row[5] = 4
elif row[5] >= 60:
row[5] = 3
elif row[5] >= 40:
row[5] = 2
elif row[5] >= 20:
row[5] = 1
else:
row[5] = 0
# reclassify wind speed
if row[6] >= 30.0:
row[6] = 3
elif row[6] >= 20.0:
row[6] = 2
elif row[6] >= 10.0:
row[6] = 1
else:
row[6] = 0
print X[0]
print y[0]
return simple_naive_bayes(X, y)
if __name__ == '__main__':
(X, y) = import_training_file(sys.argv[1], True)
simple_naive_bayes(X, y)
print '------------------'
harder_naive_bayes(X, y)
for i, C in enumerate(10. ** np.arange(1, 6)):
clf_l1_LR = LogisticRegression(C=C, penalty='l1', tol=0.01)
clf_l2_LR = LogisticRegression(C=C, penalty='l2', tol=0.01)
clf_l1_LR.fit(Xtrain, ytrain)
clf_l2_LR.fit(Xtrain, ytrain)
y1 = clf_l1_LR.predict(Xtest)
y2 = clf_l2_LR.predict(Xtest)
#L1 penalty
print "L1 Penalty with C=" + str(C)
print rmsle(ytest, y1)
print "L2 Penalty with C=" + str(C)
#L2 penalty
print rmsle(ytest, y2)
logreg = LinearRegression()
logreg.fit(Xtrain, ytrain)
y3 = logreg.predict(Xtest)
print "Linear Regression"
print y3
print rmsle(ytest,y3)
if __name__ == '__main__':
(X, y) = import_training_file(sys.argv[1], True)
logistic_regression(X, y)
temp_dict = {'datetime' : datetime[idx], 'count' : np.rint(y_total_pred[idx])}
writer.writerow(temp_dict)
def plot_residuals(train_pred, train_given, test_pred, test_given):
plt.scatter(test_pred, test_pred - test_given, c='g', s=40, alpha=0.8, label='Testing Residuals')
plt.scatter(train_pred, train_pred - train_given, c='b', s=40, alpha=0.25, label='Training Residuals')
plt.plot([0,600],[0,0], c='r')
plt.xlim(0,600)
plt.legend()
plt.title('Residuals vs Bike Share Count')
plt.xlabel("Predicted count values")
plt.ylabel("Residuals")
plt.show()
if __name__ == '__main__':
X_train, y_total, y_reg, y_casual = import_training_file(sys.argv[1])
X_test, datetime = import_testing_file(sys.argv[2])
n, _ = X_train.shape
nTrain = int(0.5*n) #training on 50% of the data
sub_Xtrain = X_train[:nTrain,:]
sub_ytrain = y_total[:nTrain]
sub_ytrain_registered = y_reg[:nTrain]
sub_ytest_registered = y_reg[nTrain:]
sub_ytrain_casual = y_casual[:nTrain]
sub_ytest_casual = y_casual[nTrain:]
sub_Xtest = X_train[nTrain:,:]
sub_ytest = y_total[nTrain:]
rf_opt = RandomForestRegressor(bootstrap=True, compute_importances=None,
criterion='mse', max_depth=None, max_features='auto',