def test():
data_train = pd.read_csv('./data_set/iris_1_3.csv', header=0)
train_data = np.array(data_train ,'float')
#
X = train_data[:,:-1]
y = train_data[:, -1]
X_train, X_test, y_train, y_true = train_test_split(X, y,test_size=1 / 3., random_state= 6)
d_tree = DecisionTreeClassifier(criterion = 'gini')
#使用有放回抽样，抽样数据进行训练树，包外数据进行验证
X_subset, y_subset, out_of_bag_data = sampling_bagging(X_train,y_train)
d_tree.fit(X_subset, y_subset)
#使用袋外数据进行树的调整;
print('y_true : ', y_true.tolist())
pre_lab = d_tree.predict(X_test)
print('pre_lab: ',pre_lab.tolist())
# print('test_data\n',np.column_stack((X_test,y_true)))
print('The accuracy was ',100 * accuracy_score(y_true,pre_lab),'% on the test ')
def compare_performance(trees_num, max_depth, bootstrap):
dataset = datasets.load_digits()
X_train, X_test, y_train, y_test = train_test_split(dataset['data'], dataset['target'], test_size=0.3, random_state=0)
# 決定木
print('##### 決定木の性能 #####')
decision_tree = DecisionTreeClassifier(max_depth=max_depth)
dt_lr_start = time.time() # 学習開始時間を記録
decision_tree.fit(X_train, y_train)
dt_lr_time = time.time() - dt_lr_start # 学習時間
dt_est_start = time.time() # 推論開始時間を記録
y_est = decision_tree.predict(X_test)
dt_est_time = time.time() - dt_est_start # 推論時間
print('学習時間 : {:.6f} [sec] 推論時間 : {:.6f} [sec]'.format(dt_lr_time, dt_est_time))
dt_train_accuracy = decision_tree.accuracy_score(X_train, y_train)
dt_test_accuracy = decision_tree.accuracy_score(X_test, y_test)
print('train accuracy : {:.4f} test_accuracy : {:.4f}'.format(dt_train_accuracy, dt_test_accuracy))
# ランダムフォレスト
print('##### ランダムフォレストの性能 #####')
random_forest = RandomForestClassifier(trees_num=trees_num, max_depth=max_depth, bootstrap=bootstrap)
rf_lr_start = time.time() # 学習開始時間を記録
random_forest.fit(X_train, y_train)
rf_lr_time = time.time() - rf_lr_start # 学習時間
rf_est_start = time.time() # 推論開始時間を記録
y_est = random_forest.predict(X_test)
rf_est_time = time.time() - rf_est_start # 推論時間
print('学習時間 : {:.6f} [sec] 推論時間 : {:.6f} [sec]'.format(rf_lr_time, rf_est_time))
rf_train_accuracy = random_forest.accuracy_score(X_train, y_train)
rf_test_accuracy = random_forest.accuracy_score(X_test, y_test)
print('train accuracy : {:.4f} test_accuracy : {:.4f}'.format(rf_train_accuracy, rf_test_accuracy))
def main():
args = get_cmd_ln_arguments()
num_columns, num_rows, training_data = parse_txt_file(args.filename)
feature_names = get_feature_names(num_columns)
decision_tree = DecisionTreeClassifier()
print('Training a decision tree classifier...')
decision_tree.fit(feature_names, training_data)
print('Decision tree classifier trained:')
decision_tree.print()
print()
print(
'Entering a loop to query the decision tree. Press ctrl-c at anytime to exit.'
)
while True:
sample = input(
'Enter a sample ({} numbers separated by a space): '.format(
num_columns))
try:
sample = line_to_int_list(sample)
except ValueError:
print(
'Input was not {} numbers separated by a space. Please try again. '
.format(num_columns))
continue
prediction = decision_tree.predict(sample)
print('Prediction: {}'.format(prediction))
def test_fit(self):
clf = DecisionTreeClassifier()
clf.fit(self.X_train, self.y_train, self.feature_names)
# verify the decision tree looks like this
#
# feature:
# outlook
# / | \
# / | \
# rainy / overcast \ sunny
# / | \
# / | \
# feature: class: feature:
# windy True humidity
# / \ / \
# False / \ True high / \ normal
# / \ / \
# class: class: class: class:
# True False False True
assert clf.root.feature == 'outlook'
rainy_node = clf.root.children_by_attribute['rainy']
overcast_node = clf.root.children_by_attribute['overcast']
sunny_node = clf.root.children_by_attribute['sunny']
assert rainy_node.feature == 'windy'
assert overcast_node.classification == True
assert sunny_node.feature == 'humidity'
assert rainy_node.children_by_attribute['False'].classification == True
assert rainy_node.children_by_attribute['True'].classification == False
assert sunny_node.children_by_attribute['high'].classification == False
assert sunny_node.children_by_attribute[
'normal'].classification == True
def test_benchmark_numerical():
print('\n** Numerical benchmark **')
df = pd.read_csv(DATA_PATH + 'dados_benchmark_v2.csv', sep=';')
dt = DecisionTreeClassifier(
target_attribute = 'Joga',
n_random_attributes=4
)
dt.fit(df)
dt.print_tree()
def main():
curr_dir = os.path.dirname(__file__)
csv_file = os.path.join(curr_dir, 'data/play.csv')
df = pd.read_csv(csv_file, index_col='Dia')
X, y = df.loc[:, df.columns != 'Jogar'], df['Jogar']
clf = DecisionTreeClassifier()
clf.fit(X, y)
print(clf.rules())
def test_decision_tree_classifier_numerical_split(self):
model = DecisionTreeClassifier()
# feature 3, 1 -> label 1
# feature 2, 0 -> label 0
features = np.array([[3, 0], [3, 0], [3, 0], [2, 0], [2, 0], [2, 0],
[1, 0], [1, 0], [1, 0], [0, 0], [0, 0]])
labels = np.array([1, 1, 1, 0, 0, 0, 1, 1, 1, 0, 0])
model.fit(features, labels)
"""
def __generate_trees(self, df, n_baggings):
'''
Given 'n' baggings, generate one Tree model for each of them
'''
for i in n_baggings:
clf = DecisionTreeClassifier(
self.target_attribute,
self.n_random_attributes
)
clf.fit(df.iloc[n_baggings[i],:])
self.trees[i] = clf
def test_decision_tree_classifier_exhaustive_categorical_split(self):
model = DecisionTreeClassifier(categorical_feature_indeces=[0, 1])
# feature[0] 3, 1 -> label 1
# feature[0] 2, 0 -> label 0, 2
features = np.array([[3, 0], [3, 0], [3, 0], [2, 0], [2, 0], [2, 0],
[1, 0], [1, 0], [1, 0], [0, 0], [0, 1]])
labels = np.array([1, 1, 1, 0, 0, 0, 1, 1, 1, 0, 2])
model.fit(features, labels)
"""
def compare_depth():
dataset = datasets.load_digits()
X_train, X_test, y_train, y_test = train_test_split(dataset['data'], dataset['target'], test_size=0.3, random_state=0)
# ランダムフォレストに関して、最良のハイパーパラメータを調べる
trees_num, bootstrap = grid_search_RF()
# 決定木の深度の制限を変えて、調べる
depth_list = [i for i in range(21)] # 深さの制限0~20まで調べる
dt_train_acc_list = []
dt_test_acc_list = []
rf_train_acc_list = []
rf_test_acc_list = []
for depth in tqdm(depth_list):
print('***** max_depth = {} *****'.format(depth))
# 決定木
decision_tree = DecisionTreeClassifier(max_depth=depth)
decision_tree.fit(X_train, y_train)
dt_train_accuracy = decision_tree.accuracy_score(X_train, y_train)
dt_test_accuracy = decision_tree.accuracy_score(X_test, y_test)
# accuracyをリストに追加
dt_train_acc_list.append(dt_train_accuracy)
dt_test_acc_list.append(dt_test_accuracy)
print('決定木 train accuracy : {:.4f} test_accuracy : {:.4f}'.format(dt_train_accuracy, dt_test_accuracy))
# ランダムフォレスト
random_forest = RandomForestClassifier(trees_num=trees_num, max_depth=depth, bootstrap=bootstrap)
random_forest.fit(X_train, y_train)
rf_train_accuracy = random_forest.accuracy_score(X_train, y_train)
rf_test_accuracy = random_forest.accuracy_score(X_test, y_test)
# accuracyをリストに追加
rf_train_acc_list.append(rf_train_accuracy)
rf_test_acc_list.append(rf_test_accuracy)
print('ランダムフォレスト train accuracy : {:.4f} test_accuracy : {:.4f}'.format(rf_train_accuracy, rf_test_accuracy))
# グラフの描画
plt.plot(depth_list, dt_train_acc_list, label='Decision Tree - train accuracy', color='r')
plt.plot(depth_list, dt_test_acc_list, label='Decision Tree - test accuracy', color='g')
plt.plot(depth_list, rf_train_acc_list, label='Random Forest - train accuracy', color='y')
plt.plot(depth_list, rf_test_acc_list, label='Random Forest - test accuracy', color='b')
plt.xlabel('Max Depth')
plt.ylabel('Accuracy')
plt.xlim(0, 20)
plt.xticks(np.arange(0, 21, 2))
plt.ylim(0, 1.0)
plt.legend(loc='lower right')
plt.title('Max Depth of Decision Trees and Accuracy')
# グラフを保存
plt.savefig('figures/mnist/max_depth_&_accuracy.png')
def visualize(model, max_depth=None):
iris_dataset = datasets.load_iris()
petal_features = iris_dataset['data'][:, 2:]
targets = iris_dataset['target']
if max_depth is None:
# 決定木の最大深度は制限しない
# アヤメのデータセットの場合は、データ数やクラス数が少ないため、深度を制限しなくても計算時間はあまりかからない
if model == 'decision_tree':
clf = DecisionTreeClassifier()
else:
clf = RandomForestClassifier()
else:
if model == 'decision_tree':
clf = DecisionTreeClassifier(max_depth=max_depth)
else:
clf = RandomForestClassifier(max_depth=max_depth)
clf.fit(petal_features, targets)
# データの取りうる範囲 +-1 を計算する
x_min = max(0, petal_features[:, 0].min() - 1)
y_min = max(0, petal_features[:, 1].min() - 1)
x_max = petal_features[:, 0].max() + 1
y_max = petal_features[:, 1].max() + 1
# 教師データの取りうる範囲でメッシュ状の座標を作る
grid_interval = 0.2
xx, yy = np.meshgrid(np.arange(x_min, x_max, grid_interval),
np.arange(y_min, y_max, grid_interval))
# メッシュの座標を学習したモデルで判定させる
Z = clf.predict(np.c_[xx.ravel(), yy.ravel()])
# 各点の判定結果をグラフに描画する
plt.contourf(xx, yy, Z.reshape(xx.shape), cmap=plt.cm.rainbow, alpha=0.4)
# データもプロット
for c in np.unique(targets):
plt.scatter(petal_features[targets == c, 0],
petal_features[targets == c, 1])
feature_names = iris_dataset['feature_names']
plt.xlabel(feature_names[2])
plt.ylabel(feature_names[3])
if max_depth is None:
plt.title('Max Depth : No Limitation')
plt.savefig('figures/iris/{}_no_limit.png'.format(model))
else:
plt.title('Max Depth : ' + str(max_depth))
plt.savefig('figures/iris/{}_depth_{}.png'.format(model, max_depth))
plt.close()
def test_predict(self):
clf = DecisionTreeClassifier()
clf.fit(self.X_train, self.y_train, self.feature_names)
expected_for_x = [
(np.array(['sunny', 'hot', 'high', False]),
False), # sunny outlook + high humidity -> don't play
(np.array(['sunny', 'hot', 'normal', False]),
True), # sunny outlook + normal humidity -> play
(np.array(['overcast', 'hot', 'high',
False]), True), # overcast outlook -> don't play
]
for x, expected in expected_for_x:
output = clf.predict(x)
assert output == expected
def fit(self, X, y):
self.forest = []
N = len(y)
N_sub_data = int(N * self.bootstrap)
for i in range(self.trees_num):
self.shuffle(X, y)
X_sub = X[:N_sub_data]
y_sub = y[:N_sub_data]
decision_tree = DecisionTreeClassifier(self.features_num,
self.max_depth)
decision_tree.fit(X_sub, y_sub)
# 得られた決定木をforestのリストに追加
self.forest.append(decision_tree)
def test_decision_tree_classifier_numerical_split_hard(self):
model = DecisionTreeClassifier()
# feature 3, 1 -> label 1
# feature 2, 0 -> label 0
features = np.array([[0, 0], [0, 0], [1, 0], [1, 0], [2, 0], [2, 0], [3, 0], [3, 0], [4, 0], [4, 0], [5, 0], [5, 0], \
[6, 0], [6, 0], [7, 0], [7, 0], [8, 0], [8, 0], [9, 0], [9, 0], [10, 0], [10, 0], [11, 0], [11, 0]])
labels = np.array([
1,
1,
0,
0,
1,
1,
0,
0,
1,
1,
0,
0,
1,
1,
0,
0,
1,
1,
0,
0,
1,
1,
0,
0,
])
model.fit(features, labels)
"""
print("test_decision_tree_classifier_numerical_split_hard")
from pprint import pprint
pprint(model._node)
"""
predictions = model.predict(
np.array([[0, 0], [1, 0], [2, 0], [3, 0], [4, 0], [5, 0], [6, 0],
[7, 0], [8, 0], [9, 0], [10, 0], [11, 0]]))
self.assertEqual(predictions.tolist(),
[1, 0, 1, 0, 1, 0, 1, 0, 1, 0, 1, 0])
def main():
dataset = datasets.load_iris()
X_train, X_test, y_train, y_test = train_test_split(dataset['data'],
dataset['target'],
test_size=0.3,
random_state=0)
# 決定木の深度の制限が1~3、制限なしの各場合について調べる
depth_list = [1, 2, 3, None]
for depth in depth_list:
print('######### max_depth = {} #########'.format(depth))
# 全ての特徴量を使用したときの精度、学習時間、推論時間、汎化性能を調べる
# 決定木
decision_tree = DecisionTreeClassifier(max_depth=depth)
dt_lr_start = time.time() # 学習開始時間を記録
decision_tree.fit(X_train, y_train)
dt_lr_time = time.time() - dt_lr_start # 学習時間
dt_est_start = time.time() # 推論開始時間を記録
y_est = decision_tree.predict(X_test)
dt_est_time = time.time() - dt_est_start # 推論時間
print('決定木 学習時間 : {:.6f} [sec] 推論時間 : {:.6f} [sec]'.format(
dt_lr_time, dt_est_time))
dt_train_accuracy = decision_tree.accuracy_score(X_train, y_train)
dt_test_accuracy = decision_tree.accuracy_score(X_test, y_test)
print('決定木 train accuracy : {:.4f} test_accuracy : {:.4f}'.
format(dt_train_accuracy, dt_test_accuracy))
# ランダムフォレスト
random_forest = RandomForestClassifier(max_depth=depth)
rf_lr_start = time.time() # 学習開始時間を記録
random_forest.fit(X_train, y_train)
rf_lr_time = time.time() - rf_lr_start # 学習時間
rf_est_start = time.time() # 推論開始時間を記録
y_est = random_forest.predict(X_test)
rf_est_time = time.time() - rf_est_start # 推論時間
print('ランダムフォレスト 学習時間 : {:.6f} [sec] 推論時間 : {:.6f} [sec]'.
format(rf_lr_time, rf_est_time))
rf_train_accuracy = random_forest.accuracy_score(X_train, y_train)
rf_test_accuracy = random_forest.accuracy_score(X_test, y_test)
print(
'ランダムフォレスト train accuracy : {:.4f} test_accuracy : {:.4f}'
.format(rf_train_accuracy, rf_test_accuracy))
# 使用する特徴量を2つ(Petal legth, Petal width)に絞って、二次元で可視化
visualize('decision_tree', max_depth=depth)
visualize('random_forest', max_depth=depth)
def _compare_sklearn_dataset(self, dataset):
dataset = load_iris()
features = dataset.data
labels = dataset.target
model_sklearn = DecisionTreeClassifierSklearn()
model_sklearn.fit(features, labels)
predictions_sklearn = model_sklearn.predict(features)
model = DecisionTreeClassifier()
model.fit(features, labels)
predictions = model.predict(features)
self.assertEqual(predictions.tolist(), predictions_sklearn.tolist())
def test_decision_tree_classifier_fit(self):
model = DecisionTreeClassifier()
# XOR problem
features = np.array([[0, 0], [0, 0], [0, 1], [0, 1], [0, 1], [1, 0],
[1, 0], [1, 0], [1, 1], [1, 1], [1, 1]])
labels = np.array([0, 0, 1, 1, 1, 1, 1, 1, 0, 0, 0])
model.fit(features, labels)
"""
from pprint import pprint
pprint(model._node)
"""
predictions = model.predict(np.array([[0, 0], [0, 1], [1, 0], [1, 1]]))
self.assertEqual(predictions.tolist(), [0, 1, 1, 0])
def fit(self, X, y):
# print('===', sys._getframe().f_code.co_filename, sys._getframe().f_code.co_name, sys._getframe().f_lineno,"===")
self.forest = []
self.n_calss_num = len(set(y)) #有几个类
self.n_calss = list(set(y)) #类标签集合
for i in range(self.n_estimators):
#随机的取数据 self.bootstrap 比率 ，表示抽取样本集的比例
X_subset, y_subset = sampling_with_reset(X, y, self.bootstrap)
###########################################
tree = DecisionTreeClassifier(self.max_features, self.criterion,
self.max_depth,
self.min_samples_split,
self.min_impurity_split)
#打印树的信息
print('tree_' + str(i))
tree.fit(X_subset, y_subset)
self.forest.append(tree) #树的集合
from sklearn import datasets, cross_validation
from decision_tree import DecisionTreeClassifier
iris = datasets.load_iris()
X, Y = iris.data, iris.target
clf = DecisionTreeClassifier()
clf.fit(X, Y)
print cross_validation.cross_val_score(clf, X, Y)
clf.draw_tree('decision_tree_example.png')
from sklearn.datasets import load_iris
from decision_tree import DecisionTreeClassifier
from sklearn import tree
# load the iris dataset
dataset = load_iris()
# set X and y variables
X, y = dataset.data, dataset.target
print(':::::::::::::::::::::::::::::::::::::::::::::')
print(f'APPROPRIATE X, y DATATYPES: {type(X)}')
print(':::::::::::::::::::::::::::::::::::::::::::::')
# create a new isntance of the DecisionTreeClassifier object
clf = DecisionTreeClassifier(max_depth=5)
# call the fit method on that object
clf.fit(X, y)
print('')
print(':::::::::::::PREDICTIONS:::::::::::::::::::::')
print('')
print(':::::::::::::::::::::::::::::::::::::::::::::')
inputs = [[1, 1.5, 5, 1.5]]
print(f'INPUTS: {inputs}')
print(f'OUR MODEL PREDICTION: {clf.predict(inputs)}')
clf2 = tree.DecisionTreeClassifier(max_depth=5)
clf2.fit(X, y)
print(f'SCIKITLEARN MODEL PREDICTION: {clf2.predict(inputs)}')
print(':::::::::::::::::::::::::::::::::::::::::::::')
image_list = dorsal_images
image_list.extend(palmar_images)
random.shuffle(image_list)
X = []
y = [0] * len(image_list)
for i in range(0, len(image_list)):
image = image_list[i]
if image in dorsal_features:
y[i] = 0
else:
y[i] = 1
X.append(label_folder_features[image])
X = np.array(X)
y = np.array(y)
decisiontree.fit(X, y)
for image_id, feature in unlabelled_features.items():
val = decisiontree.predict([feature])
# print(image_id, val)
if val[0] == 0:
result[image_id] = 'dorsal'
else:
result[image_id] = 'palmar'
elif classifier == 'SVM':
dorsal_images = list(dorsal_features.keys())
palmar_images = list(palmar_features.keys())
image_list = dorsal_images
image_list.extend(palmar_images)
random.shuffle(image_list)
X = []