def KNN_pred(self, is_total=0, combin_func='avg'):
names = locals()
r = Reader(rating_scale=(1, 5))
df = self.testdatas
total_test = np.array(df[['uid', 'iid', 'total']])
total_p = self.algos[0].test(total_test)
for i in range(1, self.no_of_criteria + 1):
# names['c' + str(i) + '_test'] = np.array(df[['uid','iid', 'c' + str(i)]])
names['c' + str(i) + '_test'] = Dataset.load_from_df(
df[['uid', 'iid', 'c' + str(i)]], reader=r)
names['c' + str(i) +
'_test'] = names.get('c' + str(i) +
'_test').build_full_trainset()
names['c' + str(i) + '_test'] = names.get('c' + str(i) +
'_test').build_
names['c' + str(i) + '_p'] = self.algos[i].test(
names.get('c' + str(i) + '_test'))
multi_p = []
if is_total == 0:
if combin_func == 'avg':
for i in range(len(total_p)):
s = 0
for j in range(1, self.no_of_criteria + 1):
s = s + names.get('c' + str(j) + '_p')[i].est
avg = s / self.no_of_criteria
p = predictions.Prediction(total_p[i].uid, total_p[i].iid,
total_p[i].r_ui, avg,
total_p[i].details)
multi_p.append(p)
elif combin_func == 'total_reg':
k = self.k
b = self.b
for i in range(len(total_p)):
s = 0
for j in range(self.no_of_criteria):
s = s + k[j] * names.get('c' + str(j + 1) +
'_p')[i].est
s = s + b
p = predictions.Prediction(total_p[i].uid, total_p[i].iid,
total_p[i].r_ui, s,
total_p[i].details)
multi_p.append(p)
else:
if combin_func == 'avg':
for i in range(len(total_p)):
s = 0
for j in range(1, self.no_of_criteria + 1):
s = s + names.get('c' + str(j) + '_p')[i].est
avg = (s + total_p[i].est) / (self.no_of_criteria + 1)
p = predictions.Prediction(total_p[i].uid, total_p[i].iid,
total_p[i].r_ui, avg,
total_p[i].details)
multi_p.append(p)
else:
print('总分作为准则不适合用于回归聚合函数')
s_mae = round(accuracy.mae(total_p), 4)
m_mae = round(accuracy.mae(multi_p), 4)
return s_mae, m_mae, total_p, multi_p
def transtoPrediction(multi, single):
P = []
for m in multi:
for s in single:
if m[0] == s[0] and m[1] == s[1]:
p = predictions.Prediction(m[0], m[1], s[2], m[2], s[4])
P.append(p)
return P
def avg(P, total_p):
multi_p = []
for i in range(len(P)):
pred = (P[i][3] + P[i][4] + P[i][5] + P[i][6] + P[i][7] +
P[i][2]) / 6 #整体评分作准则
# pred = (P[i][3] + P[i][4] + P[i][5] + P[i][6] + P[i][7] )/5 #整体评分不做准则
p = predictions.Prediction(P[i][0], P[i][1], total_p[i][2], pred,
total_p[i][4])
multi_p.append(p)
return multi_p
def totaldata_regModel(P, k, b, total_p):
multi_p = []
for i in range(len(P)):
#整体评分不做准则
pred = P[i][3] * k[0] + P[i][4] * k[1] + P[i][5] * k[2] + P[i][6] * k[
3] + P[i][7] * k[4] + b
p = predictions.Prediction(P[i][0], P[i][1], total_p[i][2], pred,
total_p[i][4])
multi_p.append(p)
return multi_p
def Predict(self,
min_or_more='min',
pred_options={
'is_total': False,
'combin_func': 'avg'
}):
names = locals()
r = Reader(rating_scale=(1, 5))
if min_or_more == 'min':
df = self.min_test.sort_values(by='uid')
else:
df = self.more_test.sort_values(by='uid')
# df = self.testDatas
total_test = np.array(df[['uid', 'iid', 'total']])
total_p = self.algos[0].test(total_test)
for i in range(1, self.no_of_criteria + 1):
# names['c' + str(i) + '_test'] = np.array(df[['uid','iid', 'c' + str(i)]])
names['c' + str(i) + '_test'] = np.array(
df[['uid', 'iid', 'c' + str(i)]])
names['c' + str(i) + '_p'] = self.algos[i].test(
names.get('c' + str(i) + '_test'))
multi_p = []
if pred_options['is_total']:
if pred_options['combin_func'] == 'avg':
for i in range(len(total_p)):
s = 0
for j in range(1, self.no_of_criteria + 1):
s = s + names.get('c' + str(j) + '_p')[i].est
avg = (s + total_p[i].est) / (self.no_of_criteria + 1)
p = predictions.Prediction(total_p[i].uid, total_p[i].iid,
total_p[i].r_ui, avg,
total_p[i].details)
multi_p.append(p)
else:
print('总分作为准则不适合用于回归聚合函数')
else:
if pred_options['combin_func'] == 'avg':
for i in range(len(total_p)):
s = 0
for j in range(1, self.no_of_criteria + 1):
s = s + names.get('c' + str(j) + '_p')[i].est
avg = s / self.no_of_criteria
p = predictions.Prediction(total_p[i].uid, total_p[i].iid,
total_p[i].r_ui, avg,
total_p[i].details)
multi_p.append(p)
elif pred_options['combin_func'] == 'total_reg':
k = self.k
b = self.b
for i in range(len(total_p)):
s = 0
for j in range(self.no_of_criteria):
s = s + k[j] * names.get('c' + str(j + 1) +
'_p')[i].est
s = s + b
p = predictions.Prediction(total_p[i].uid, total_p[i].iid,
total_p[i].r_ui, s,
total_p[i].details)
multi_p.append(p)
elif pred_options['combin_func'] == 'info_entropy':
self.Info_Entropy()
H = np.array(self.H)
for i in range(len(total_p)):
s = 0
if len(np.argwhere(H[:, 0] == total_p[i].uid)):
h = H[np.argwhere(H[:, 0] == total_p[i].uid)][0][0]
for j in range(1, self.no_of_criteria + 1):
s = s + h[j] * names.get('c' + str(
j) + '_p')[i].est / h[self.no_of_criteria + 1]
p = predictions.Prediction(total_p[i].uid,
total_p[i].iid,
total_p[i].r_ui, s,
total_p[i].details)
multi_p.append(p)
else:
s = 0
for j in range(1, self.no_of_criteria + 1):
s = s + names.get('c' + str(j) + '_p')[i].est
s = s / self.no_of_criteria
p = predictions.Prediction(total_p[i].uid,
total_p[i].iid,
total_p[i].r_ui, s,
total_p[i].details)
multi_p.append(p)
s_mae = round(accuracy.mae(total_p), 4)
m_mae = round(accuracy.mae(multi_p), 4)
return s_mae, m_mae
def SVD(self, seed=3, n_factor=20, n_epoch=20):
'''
seed：int-3划分训练集测试集的随机种子
k：int-40，最大邻居数量
options：dict-{'name': 'pearson', 'user_based': False}，算法的选项，默认为Pearson相似度，基于项目的方法
'''
df = self.datas
names = locals()
r = Reader(rating_scale=(1, 5))
# 读取、划分数据;训练预测数据
total = Dataset.load_from_df(df[['uid', 'iid', 'total']], reader=r)
total_train, total_test = train_test_split(total, random_state=seed)
total_algo = SVD(n_factors=n_factor, n_epochs=n_epoch, verbose=True)
total_algo.fit(total_train)
total_p = total_algo.test(total_test)
for i in range(1, self.no_of_criteria + 1):
names['c' + str(i)] = Dataset.load_from_df(
df[['uid', 'iid', 'c' + str(i)]], reader=r)
names['c' + str(i) + '_train'], names['c' + str(i) +
'_test'] = train_test_split(
names.get('c' + str(i)),
random_state=seed)
names['algo_c' + str(i)] = SVD(n_factors=n_factor,
n_epochs=n_epoch,
verbose=True)
names.get('algo_c' + str(i)).fit(names.get('c' + str(i) +
'_train'))
names['c' + str(i) + '_p'] = names.get('algo_c' + str(i)).test(
names.get('c' + str(i) + '_test'))
#
multi_p = []
if self.is_total == 0:
if self.combin_func == 'avg':
for i in range(len(total_p)):
s = 0
for j in range(1, self.no_of_criteria + 1):
s = s + names.get('c' + str(j) + '_p')[i].est
avg = s / self.no_of_criteria
p = predictions.Prediction(total_p[i].uid, total_p[i].iid,
total_p[i].r_ui, avg,
total_p[i].details)
multi_p.append(p)
elif self.combin_func == 'total_reg':
for i in range(1, self.no_of_criteria + 1):
names['x' + str(i)] = np.array(
names.get('c' + str(i) + '_train').build_testset())[:,
2]
if i == 1:
x = names.get('x' + str(i))
else:
x = np.vstack((x, names.get('x' + str(i))))
x = x.T
y = np.array(total_train.build_testset())[:, 2]
reg = LinearRegression()
reg.fit(x, y)
k = reg.coef_
b = reg.intercept_
for i in range(len(total_p)):
s = 0
for j in range(self.no_of_criteria):
s = s + k[j] * names.get('c' + str(j + 1) +
'_p')[i].est
s = s + b
p = predictions.Prediction(total_p[i].uid, total_p[i].iid,
total_p[i].r_ui, s,
total_p[i].details)
multi_p.append(p)
else:
if self.combin_func == 'avg':
for i in range(len(total_p)):
s = 0
for j in range(1, self.no_of_criteria + 1):
s = s + names.get('c' + str(j) + '_p')[i].est
avg = (s + total_p[i].est) / (self.no_of_criteria + 1)
p = predictions.Prediction(total_p[i].uid, total_p[i].iid,
total_p[i].r_ui, avg,
total_p[i].details)
multi_p.append(p)
else:
print('总分作为准则不适合用于回归聚合函数')
s_mae = round(accuracy.mae(total_p), 4)
m_mae = round(accuracy.mae(multi_p), 4)
return s_mae, m_mae
def transtoPredictions(multi_P, total_P, P):
for m in multi_P:
for t in total_P:
if m[0] == t[0] and m[1] == t[1]:
p = predictions.Prediction(m[0], m[1], t[2], m[2], t[4])
P.append(p)
'_test'] = train_test_split(
names.get('c' + str(i)),
random_state=3)
names['algo_c' + str(i)] = SVD(n_factors=20, n_epochs=20, verbose=True)
names.get('algo_c' + str(i)).fit(names.get('c' + str(i) + '_train'))
names['c' + str(i) + '_p'] = names.get('algo_c' + str(i)).test(
names.get('c' + str(i) + '_test'))
#
multi_p = []
for i in range(1, 6):
names['x' + str(i)] = np.array(
names.get('c' + str(i) + '_train').build_testset())[:, 2]
if i == 1:
x = names.get('x' + str(i))
else:
x = np.vstack((x, names.get('x' + str(i))))
x = x.T
y = np.array(total_train.build_testset())[:, 2]
reg = LinearRegression()
reg.fit(x, y)
k = reg.coef_
b = reg.intercept_
for i in range(len(total_p)):
s = 0
for j in range(5):
s = s + k[j] * names.get('c' + str(j + 1) + '_p')[i].est
s = s + b
p = predictions.Prediction(total_p[i].uid, total_p[i].iid, total_p[i].r_ui,
s, total_p[i].details)
multi_p.append(p)