def init_W(self, mode='normal', decomposition='ova'):
self.W = {}
if (self.status != 'load_train_data') and (self.status != 'train'):
print("Please load train data first.")
return self.W
self.status = 'init'
self.data_num = len(self.train_Y)
self.data_demension = len(self.train_X[0])
self.decomposition = decomposition
if self.decomposition == 'ovo':
self.class_list = list(
itertools.combinations(np.unique(self.train_Y), 2))
elif self.decomposition == 'ova':
self.class_list = np.unique(self.train_Y)
for class_item in self.class_list:
self.W[class_item] = np.zeros(self.data_demension)
if mode == 'linear_regression_accelerator':
accelerator = linear_regression.Accelerator()
for class_item in self.class_list:
if self.decomposition == 'ovo':
modify_X, modify_Y = utility.DatasetLoader.modify_XY(
self.train_X, self.train_Y, class_item)
self.temp_train_X = self.train_X
self.temp_train_Y = self.train_Y
self.train_X = modify_X
self.train_Y = modify_Y
self.temp_data_num = self.data_num
self.data_num = len(self.train_Y)
self.temp_W = self.W
self.W = self.temp_W[class_item]
self.temp_W[class_item] = accelerator.init_W(self)
self.train_X = self.temp_train_X
self.train_Y = self.temp_train_Y
self.temp_train_X = []
self.temp_train_Y = []
self.data_num = self.temp_data_num
self.temp_data_num = 0
self.W = self.temp_W
self.temp_W = {}
elif self.decomposition == 'ova':
modify_Y = utility.DatasetLoader.modify_Y(
self.train_Y, class_item)
self.temp_train_Y = self.train_Y
self.train_Y = modify_Y
self.temp_W = self.W
self.W = self.temp_W[class_item]
self.temp_W[class_item] = accelerator.init_W(self)
self.train_Y = self.temp_train_Y
self.temp_train_Y = []
self.W = self.temp_W
self.temp_W = {}
return self.W
def calculate_alpha_s(self, weak_learner, s):
alpha = 0.0
new_s = s
data_num = len(weak_learner.train_Y)
X = []
for i in range(data_num):
predict_string = np.array(list(map(str, weak_learner.train_X[i])))
predict_string = ' '.join(predict_string[1:]) + ' ' + str(weak_learner.train_Y[i])
prediction = weak_learner.prediction(predict_string, 'test_data')
X.append([float(prediction['prediction'])])
X = np.array(X)
linear = linear_regression.LinearRegression()
linear.status = 'load_train_data'
linear.train_X = X
linear.train_Y = weak_learner.train_Y - s
linear.set_param()
linear.init_W()
linear.train()
alpha = linear.W[0]
new_s = s + alpha * np.ravel(X)
return alpha, new_s
def train(self):
predict_X = []
for x in self.models[0].train_X:
x_string = ' '.join(map(str, x.tolist()[1:]))
transform_x = self.feature_transform(x_string)
transform_x.insert(0, 1)
predict_X.append(transform_x)
self.linear = linear_regression.LinearRegression()
self.linear.load_train_data()
self.linear.load_test_data()
self.linear.train_X = np.array(predict_X)
self.linear.train_Y = self.models[0].train_Y
self.linear.set_param()
self.linear.init_W()
self.linear.train()
def init_W(self, mode='normal'):
'''
Init the W
Simple way is init W all zeros
'''
if (self.status != 'load_train_data') and (self.status != 'train'):
print("Please load train data first.")
return self.W
self.status = 'init'
self.data_num = len(self.train_Y)
self.data_demension = len(self.train_X[0])
self.W = np.zeros(self.data_demension)
if mode == 'linear_regression_accelerator':
accelerator = linear_regression.Accelerator()
self.W = accelerator.init_W(self)
return self.W
