data = pd.read_csv("D:/Sai/JavaDoc/Cousera/5/2/gbm-data.csv")
y = data[data.columns[1]].values
x = data[data.columns[1:]].values
x_train, x_test, y_train, y_test = train_test_split(x,
y,
test_size=0.33,
random_state=42)
ls = [1, 0.5, 0.3, 0.2, 0.1]
for i in ls:
clf = GradientBoostingClassifier(n_estimators=250,
verbose=True,
random_state=241,
learning_rate=i)
clf.fit(x_train, y_train)
qual_test = clf.staged_decision_function(x_test)
qual_train = clf.staged_decision_function(x_train)
predict = clf.predict(x_test)
pred_trans = 1 / (1 + math.exp(-predict))
plt.figure()
plt.plot(test_loss, 'r', linewidth=2)
plt.plot(train_loss, 'g', linewidth=2)
plt.legend(['test', 'train'])
print("MSE: %.4f" % mse)
print
# <codecell>
params = clf.get_params()
params
# <codecell>
test_score = np.zeros((params['n_estimators'],), dtype=np.float64)
test_score
# <codecell>
for i, y_pred in enumerate(clf.staged_decision_function(X_test)):
test_score[i] = clf.loss_(y_test, y_pred)
test_score
# <codecell>
plt.figure(figsize=(12, 6))
plt.subplot(1, 2, 1)
plt.title('Deviance')
plt.plot(np.arange(params['n_estimators']) + 1, clf.train_score_, 'b-',
label='Training Set Deviance')
plt.plot(np.arange(params['n_estimators']) + 1, test_score, 'r-',
label='Test Set Deviance')
plt.legend(loc='upper right')
plt.xlabel('Boosting Iterations')
