def oversamp():
x, y = dataing.data_make() ##返回的都是数值类型
#print(x[317],y[317])
x_set = np.array(x, dtype=object) ##由于是不定长数据，需要带上dtype
y_set = np.array(y)
'''print(x_set[0])
print(x_set.shape) # （516，）
print(y_set.shape) # （516，10）'''
max_length = 193
X = keras.preprocessing.sequence.pad_sequences(x_set,
maxlen=max_length,
dtype='float64',
padding='post',
value=[0.0, 0.0]).tolist()
x = []
for i in X:
k = [n for a in i for n in a]
x.append(k)
#print(x[0])
y = y_set[:, 0].tolist() ###设置单标签
#print('y_set',len(y_set))
print(Counter(y))
##平衡数据：0第一类{1: 330, 0: 187}，1第二类{1: 393, 0: 124}，2第三类{1: 347, 0: 170}
########： 3第四类{0: 430, 1: 87}，4第五类{1: 316, 0: 201}，5第六类{0: 482, 1: 35}
########： 6第七类{0: 444, 1: 73}，7第八类{0: 432, 1: 85}，8第九类{0: 422, 1: 95}，9第十类{0: 502, 1: 15}
from imblearn.over_sampling import RandomOverSampler
ros = RandomOverSampler(random_state=10)
x_resampled, y_resampled = ros.fit_resample(x, y)
print(sorted(Counter(y_resampled).items()))
#print(x_resampled[0])
#print(len(y_resampled))
result = []
for i in x_resampled:
b = np.array(i).reshape(193, 2).tolist() # reshape(列的长度，行的长度)
result.append(b)
#print(type(result),type(y_resampled))##返回的都是等长的列表类型
return result, y_resampled
import numpy as np
import pandas as pd
from tensorflow.keras.layers import Dense
import numpy as np
import pandas as pd
import random
from sklearn import metrics
import tensorflow.keras.backend as K
#在自己的库函数
import dataing ##读取数据
import data_same_length ##处理成定长数据
import loss_function
from tensorflow import keras
#1.读入数据,打乱顺序
x, y = dataing.data_make()
train_set = []
for i in range(len(x)):
train_set.append([x[i], y[i]])
random.shuffle(train_set) ###数据重排
x_set = [e[0] for e in train_set] #特征数据
y_set = [f[1] for f in train_set] #标签
##训练数据最大长度,可以在dataing程序中查看，目前是276
max_length = 276
##keras.preprocessing.sequence.pad_sequences将多个序列截断或补齐为相同长度，返回numpy数组
X = keras.preprocessing.sequence.pad_sequences(x_set,
maxlen=max_length,
dtype='float64',
padding='post',
import pandas as pd
import random
from sklearn import metrics
import tensorflow.keras.backend as K
#在自己的库函数
import dataing ##读取数据
import data_same_length ##处理成定长数据
import loss_function
##平衡数据：0第一类{1: 330, 0: 187}，1第二类{1: 393, 0: 124}，2第三类{1: 347, 0: 170}
########： 3第四类{0: 430, 1: 87}，4第五类{1: 316, 0: 201}，5第六类{0: 482, 1: 35}
########： 6第七类{0: 444, 1: 73}，7第八类{0: 432, 1: 85}，8第九类{0: 422, 1: 95}，9第十类{0: 502, 1: 15}
#读取数据
x, y = dataing.data_make() ##不定长特征
#x,y = dataing.data_make_samelen()##定长特征
train_set = []
for i in range(len(x)):
random.shuffle(x[i])
#print(x[i])
#a =x[i].copy()
train_set.append([x[i], y[i]])
random.shuffle(train_set) ###数据重排
x_set1 = [e[0] for e in train_set] #特征数据
y_set1 = [f[1] for f in train_set] #标签
'''x1 = x.copy()
y1 = np.array(y)[:,0].tolist()
print(len(x))
# -*- coding: utf-8 -*-
#@Author : lynch
# 导入本项目所需要的包
import dataing ##读取数据
import data_same_length ##处理成定长数据
import tensorflow as tf
from tensorflow.keras.layers import Dense
import numpy as np
import pandas as pd
import random
#加载数据
x, y = dataing.data_make(
) ##分别读取读取‘training_set.csv’和‘training_label.csv’文件，不定长
train_set = []
for i in range(len(x)):
train_set.append([x[i], y[i]])
random.shuffle(train_set) ###数据重排
x_1 = [e[0] for e in train_set] #特征数据
y_1 = [f[1] for f in train_set] #标签
'''#数据长度归一化
length_x = []
for i in x_set:
length_x.append(len(i))
max_length = max(length_x)
print('训练数据最大长度是：',max_length)
x_set = data_same_length.same_length(x_set,max_length)
print(x_set[0])
print(len(x_set[0]))'''