forked from 0oljyo0/TianChi
-
Notifications
You must be signed in to change notification settings - Fork 0
/
pool.py
332 lines (244 loc) · 13.3 KB
/
pool.py
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
# -*- coding: utf-8 -*-
import os
import csv
import tensorflow as tf
import numpy as np
import pandas as pd
# import matplotlib.pyplot as plt
import tensorflow as tf
# from sklearn.model_selection import train_test_split
# from sklearn.svm import SVC
from sklearn import preprocessing
# from sklearn.externals import joblib
# from sklearn import metrics
import seaborn
import matplotlib.pyplot as plt
def stop():
while 1:
pass
def data_classify(data_src, label_name, data_train_proportion): #把乱的数据根据标签分好类 变成有序数据 再分成训练集和测试集两部分 分为两个list返回 data_src存放从文件读来得原始数据 label_name存放的是用于分类的列的名字 data_train_proportion是训练集占总的比列
labels_class_num = data_src.loc[:,label_name].nunique()
labels_classes = data_src.loc[:,label_name].unique()
order=np.argsort(labels_classes)
labels_classes = labels_classes[order]
temp_train = []
temp_test = []
for lab in labels_classes:
classfy_pos = data_src.loc[data_src[label_name] == lab,:].index.values
data_classed = data_src.loc[classfy_pos,:]
data_classed = data_classed.reset_index() #重设索引
data_classed.drop(['index'],axis=1,inplace=True) #去除多余索引
row_num,col_num = data_classed.shape
train_row_num = row_num * data_train_proportion
train_data = data_classed.loc[0:(train_row_num)]
test_data = data_classed.loc[train_row_num:]
train_data = train_data.reset_index() #重设索引
train_data.drop(['index'],axis=1,inplace=True) #去除多余索引
test_data = test_data.reset_index() #重设索引
test_data.drop(['index'],axis=1,inplace=True) #去除多余索引
temp_train.append(train_data)
temp_test.append(test_data)
return temp_train,temp_test
def where_is_nan(data): #查数据中缺省的位置 pd的格式
temp = 0
print(sys._getframe().f_back.f_lineno)
for columname in data:
if(np.isnan(data.loc[:,columname]).any()):
temp = 1
print(columname)
print("\tsum = ",np.isnan(data.loc[:,columname]).sum(),"\n\tpos = ",np.where(np.isnan(data.loc[:,columname]) == True)[0])
if temp == 0 :
print("have no nan\n")
def drop_index_rule(data_src,rule):
drop_pos = data_src.loc[rule,:].index.values
data_src.drop(drop_pos,axis=0,inplace=True) #去除年月日
def output_all_col_dtype(data_src):
for col in data_src:
print(data_src[col].dtype)
def data_expand(data_list,row_expand_aim_num): #数据扩增 解决样本不平衡 待扩增的数据以list的新式放进去 row_expand_aim_num是想要扩增为多少行
# data_list[0] = pd.concat( [data_list[0], data_list[0]], axis = 0)
for count in range(len(data_list)):
row_num = data_list[count].shape[0]
if row_num < row_expand_aim_num:
err = row_expand_aim_num - row_num
temp = data_list[count].sample(n = err, replace=True)
data_list[count] = pd.concat( [data_list[count], temp], axis = 0)
data_list[count] = data_list[count].reset_index() #重设索引
data_list[count].drop(['index'],axis=1,inplace=True) #去除多余索引
def pack_data_list(data_list):
temp = data_list[0]
row_num = len(data_list)
for count in range(1,row_num):
temp = pd.concat( [temp, data_list[count]], axis = 0)
temp = temp.reset_index() #重设索引
temp.drop(['index'],axis=1,inplace=True) #去除多余索引
return temp
def get_next_batch(all_data,batch_size,step):
row_num = all_data.shape[0]
batch_num = row_num/batch_size
batch_count = step%batch_num
begin = int(batch_count * batch_size)
end = int((batch_count + 1) * batch_size)
return all_data[begin:end]
#np.set_printoptions(threshold=np.inf)
#显示所有列
# pd.set_option('display.max_columns', None)
# 显示所有行
# pd.set_option('display.max_rows', None)
# 设置value的显示长度为100,默认为50
# pd.set_option('max_colwidth',100)
data = pd.read_csv("./src-data/happiness_train_abbr.csv")
data.drop(['survey_time'],axis=1,inplace=True) #去除年月日
data.drop([3129],axis=0,inplace=True) #去除缺省的行
data = data.dropna(axis=1, how='any') #去除有缺省列
err_index = data.loc[data.happiness<1,:].index.values #去除幸福感出错的行
data.drop(err_index,axis=0,inplace=True)
data.loc[data['income']>1200000,'income'] = 1200000 #工资太高的给他限制在范围内
data.loc[data['family_income']>2500000,'family_income'] = 2500000 #工资太高的给他限制在范围内
train_data, test_data = data_classify(data, 'happiness', 0.8)
data_expand(train_data,4000)
# data_expand(test_data,1000)
train_data = pack_data_list(train_data)
test_data = pack_data_list(test_data)
train_data = train_data.sample(frac=1).reset_index(drop=True)
train_data = train_data.sample(frac=1).reset_index(drop=True)
train_data = train_data.sample(frac=1).reset_index(drop=True)
test_data = test_data.sample(frac=1).reset_index(drop=True)
test_data = test_data.sample(frac=1).reset_index(drop=True)
test_data = test_data.sample(frac=1).reset_index(drop=True)
train_data.drop(['id'],axis=1,inplace=True) #删除id
test_data.drop(['id'],axis=1,inplace=True) #删除id
train_labels = train_data['happiness']
test_labels = test_data['happiness']
labels_test_src = test_labels
labels_train_src = train_labels
train_data.drop(['happiness'],axis=1,inplace=True) #删除id
test_data.drop(['happiness'],axis=1,inplace=True) #删除id
train_labels_temp = train_labels.values #获取原始的幸福感数据 numpy格式
train_labels = np.zeros((train_labels_temp.shape[0],5),dtype = np.int) #创建可以用于训练的np 标签 5分类问题
for i in range(train_labels_temp.shape[0]):
train_labels[i,train_labels_temp[i]-1] = 1
test_labels_temp = test_labels.values #获取原始的幸福感数据 numpy格式
test_labels = np.zeros((test_labels_temp.shape[0],5),dtype = np.int) #创建可以用于训练的np 标签 5分类问题
for i in range(test_labels_temp.shape[0]):
test_labels[i,test_labels_temp[i]-1] = 1
train_data = train_data.values
test_data = test_data.values
train_data = preprocessing.scale(train_data) #数据标准化 使数据的每列基本满足正态分布
test_data = preprocessing.scale(test_data)
print(train_data.shape,test_data.shape,train_labels.shape,test_labels.shape)
# stop()
# print(data.loc[data['happiness'] == 1,:].shape)
# data['happiness'].values
#以下是构建神经网络部分
ax = [] # 定义一个 x 轴的空列表用来接收动态的数据
ay = [] # 定义一个 y 轴的空列表用来接收动态的数据
ay2 = []
TRAIN_DATA_SIZE = train_data.shape[0]
BATCH_SIZE = 1000
LEARNING_RATE_BASE = 0.9
LEARNING_RATE_DECAY = 0.99
REGULARIZATION_RATE = 0.009
MOVING_AVERAGE_DECAY = 0.99
MODEL_SAVE_PATH="modules/"
MODEL_NAME="mnist_model"
TRAINING_STEPS = 900000000
input_num = 35
layer_node_num = [100, 50, 25]
output_num = 5
def get_weight_variable(name, shape, regularizer):
weights = tf.get_variable(name, shape, initializer=tf.truncated_normal_initializer(stddev=0.1))
if regularizer != None: tf.add_to_collection('losses', regularizer(weights))
return weights
#定义两层简单的网络
x = tf.placeholder(tf.float32, [None, input_num], name='x-input')
y_ = tf.placeholder(tf.float32, [None, output_num], name='y-input')
regularizer = tf.contrib.layers.l2_regularizer(REGULARIZATION_RATE)
weights1 = get_weight_variable("weights1",[input_num, layer_node_num[0]], regularizer)
biases1 = tf.get_variable("biases1", [layer_node_num[0]], initializer=tf.constant_initializer(0.0))
layer1 = tf.nn.relu(tf.matmul(x, weights1) + biases1)
weights2 = get_weight_variable("weights2", [layer_node_num[0], layer_node_num[1]],regularizer)
biases2 = tf.get_variable("biases2", [layer_node_num[1]], initializer=tf.constant_initializer(0.0))
layer2 = tf.nn.relu(tf.matmul(layer1, weights2) + biases2)
weights3 = get_weight_variable("weights3", [layer_node_num[1], layer_node_num[2]],regularizer)
biases3 = tf.get_variable("biases3", [layer_node_num[2]], initializer=tf.constant_initializer(0.0))
layer3 = tf.nn.tanh(tf.matmul(layer2, weights3) + biases3)
weights_out = get_weight_variable("weights_out",[layer_node_num[2], output_num], regularizer)
biases_out = tf.get_variable("biases_out", [output_num], initializer=tf.constant_initializer(0.0))
layer_out = tf.matmul(layer3, weights_out) + biases_out
y = layer_out
global_step = tf.Variable(0, trainable=False)
variable_averages = tf.train.ExponentialMovingAverage(MOVING_AVERAGE_DECAY, global_step)
variables_averages_op = variable_averages.apply(tf.trainable_variables())
cross_entropy = tf.nn.sparse_softmax_cross_entropy_with_logits(logits=y, labels=tf.argmax(y_, 1))
cross_entropy_mean = tf.reduce_mean(cross_entropy)
loss = cross_entropy_mean + tf.add_n(tf.get_collection('losses'))
learning_rate = tf.train.exponential_decay(
LEARNING_RATE_BASE,
global_step,
TRAIN_DATA_SIZE / BATCH_SIZE, LEARNING_RATE_DECAY,
staircase=True)
# Optimizer
# GradientDescentOptimizer
# AdagradOptimizer
# AdagradDAOptimizer
# MomentumOptimizer
# AdamOptimizer
# FtrlOptimizer
# RMSPropOptimizer
# train_step = tf.train.GradientDescentOptimizer(learning_rate).minimize(loss, global_step=global_step)
train_step = tf.train.GradientDescentOptimizer(learning_rate).minimize(loss, global_step=global_step)
with tf.control_dependencies([train_step, variables_averages_op]):
train_op = tf.no_op(name='train')
saver = tf.train.Saver()
with tf.Session() as sess:
tf.global_variables_initializer().run()
plt.ion() # 开启一个画图的窗口
for step_count in range(TRAINING_STEPS):
xs = get_next_batch(train_data,BATCH_SIZE,step_count)
ys = get_next_batch(train_labels,BATCH_SIZE,step_count)
xt = test_data
yt = test_labels
_, loss_value, step ,np_output_labels= sess.run([train_op, loss, global_step, y], feed_dict={x: xs, y_: ys})
# print("loss :", loss_value)
np_test_output_labels,_ = sess.run([y,loss], feed_dict={x: xt, y_: yt})
output_labels_size,_ = np_output_labels.shape #计算训练集的错误率
result = np.zeros((output_labels_size))
for i in range(output_labels_size):
maxarg = np.argmax(np_output_labels[i])
np_output_labels[i] = 0
np_output_labels[i][maxarg] = 1
result[i] = np.logical_not(np.logical_not(np_output_labels[i] == ys[i]).any())
#pass
right_sum = result.sum()
err_sum = np.logical_not(result).sum()
print("训练集:\n","错误率:",err_sum/result.shape[0]," 正确率:",right_sum/result.shape[0])
ay2.append(right_sum/result.shape[0]) # 添加 i 的平方到 y 轴的数据中
output_labels_size,_ = np_test_output_labels.shape #计算测试集的错误率
result = np.zeros((output_labels_size))
for i in range(output_labels_size):
maxarg = np.argmax(np_test_output_labels[i])
np_test_output_labels[i] = 0
np_test_output_labels[i][maxarg] = 1
result[i] = np.logical_not(np.logical_not(np_test_output_labels[i] == yt[i]).any())
pass
right_sum = result.sum()
err_sum = np.logical_not(result).sum()
# print("score :",((labels_test_src.values - result)**2).sum()/result.shape[0])
print("\t测试集\n","\t错误率:",err_sum/result.shape[0]," 正确率:",right_sum/result.shape[0]," score :",((labels_test_src.values - result)**2).sum()/result.shape[0])
saver.save(sess, os.path.join(MODEL_SAVE_PATH, MODEL_NAME))
ax.append(step_count) # 添加 i 到 x 轴的数据中
ay.append(right_sum/result.shape[0]) # 添加 i 的平方到 y 轴的数据中
plt.clf() # 清除之前画的图
plt.plot(ax,ay,color = 'red') # 画出当前 ax 列表和 ay 列表中的值的图形
plt.plot(ax,ay2) # 画出当前 ax 列表和 ay 列表中的值的图形
plt.pause(0.01) # 暂停一秒
plt.ioff() # 关闭画图的窗口
#训练完成后,通过模型得到预测的y值
# predict_y=sess.run(predict,feed_dict={x:x_data})
# plt.figure()
# plt.scatter(x_data,y_data)
# plt.plot(x_data,predict_y,'r',lw=5)
# plt.show()
# data[:].family_income.plot()
# plt.show()