[V&N2020 公开赛] 记一道MachineLearning题

  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

$ nc node3.buuoj.cn 25115
hello,whats your name?
frieggs
hi,frieggs
There is a func and some data to train,please guess my func and answer my question:
When you need help,please take this num to ask for help:1
Now I will give you 70 data:[ENTER]

x=0.23293131229477126,y=-0.6546346007667923;
x=0.9141812803441588,y=-1.923420422937762;
x=0.5978186731303873,y=-1.4330594279625735;
x=0.9081244761918499,y=-1.9252163352184513;
x=0.6397127738411532,y=-1.5098826967326122;
x=0.4126789955389717,y=-1.0610931549067264;
x=0.4723890788951174,y=-1.164921025969659;
x=0.14372339309751103,y=-0.4077362577217497;
x=0.9225541186759473,y=-1.9643974463302152;
x=0.7711563780945728,y=-1.702248722890862;
x=0.2038953054962851,y=-0.5997436441843649;
x=0.6263053110417236,y=-1.4816287831706654;
x=0.32514168647239805,y=-0.8664563573379703;
x=0.6249863260140601,y=-1.4632787128261502;
x=0.2706873892963526,y=-0.667958062941054;
x=0.18529895655996909,y=-0.5153729906817134;
x=0.43797173474171236,y=-1.1505276391445383;
x=0.09206362654732536,y=-0.22902442460467892;
x=0.28674442651700094,y=-0.7568451882263891;
x=0.8473679877638907,y=-1.7847257038214273;
x=0.5199051988574164,y=-1.3132961377962167;
x=0.23662738696874386,y=-0.6775411613309307;
x=0.5105118320718761,y=-1.2921978879411098;
x=0.5022470922340977,y=-1.3112019941405897;
x=0.7752705900700582,y=-1.695869752384335;
x=0.24518941046122678,y=-0.6728831661626552;
x=0.5476635719996427,y=-1.3350604937270167;
x=0.14555038196250714,y=-0.4502351420894147;
x=0.5402691798964878,y=-1.2881121079051063;
x=0.060376524271612886,y=-0.22952293246414246;
x=0.6685815075710536,y=-1.5457709729242959;
x=0.3891074774923087,y=-1.0099949516551094;
x=0.8497091497091634,y=-1.833082162965835;
x=0.5308089077776695,y=-1.3000807761787088;
x=0.5319205111100348,y=-1.3027475694565198;
x=0.7460198741302583,y=-1.6791168349347692;
x=0.4361965526637247,y=-1.0805461179967266;
x=0.8381685198758803,y=-1.802818419639961;
x=0.49170712019766427,y=-1.2995792517865328;
x=0.09375206370396205,y=-0.3210610110426531;
x=0.23191284838776027,y=-0.6098419736980532;
x=0.3861293210803447,y=-1.050336512975188;
x=0.6393526599126003,y=-1.5280334701843157;
x=0.6831673785649255,y=-1.6218118230629652;
x=0.7607856571980453,y=-1.6892705354914876;
x=0.6556675697457789,y=-1.540235578785005;
x=0.25976364113975015,y=-0.672125184716639;
x=0.6126008719466577,y=-1.5310814384436477;
x=0.9694970227101206,y=-1.9325120143495362;
x=0.6949118217585081,y=-1.5717537157195178;
x=0.08188742975790797,y=-0.21082472094809357;
x=0.7870408155959038,y=-1.7347261736185102;
x=0.25292479781129673,y=-0.7236291459828087;
x=0.25519691318721127,y=-0.6465866313592334;
x=0.022113229287526015,y=0.0013596887514838174;
x=0.7467069251250199,y=-1.6701056178020084;
x=0.2420825936331218,y=-0.7123246046772168;
x=0.703562175733554,y=-1.6269793627781495;
x=0.4336818991326158,y=-1.0629238800984522;
x=0.21475120663954939,y=-0.5480934889951207;
x=0.7213355828205726,y=-1.63253228990537;
x=0.7794141621830423,y=-1.755053624663293;
x=0.2777075023004413,y=-0.7901581455824338;
x=0.3568414571627093,y=-0.9602735787303361;
x=0.513602580825322,y=-1.3143973009694436;
x=0.6108141481891833,y=-1.4581997442253065;
x=0.9668204533936676,y=-1.972803885674858;
x=0.7890379052688813,y=-1.8087162106504862;
x=0.5277245729370588,y=-1.308632554138398;
x=0.44539342462609965,y=-1.1234586310392338;
okay,do you guess my func?I will ask you 10 questions.If your loss less than 0.06,i will give you the FLAG[ENTER]

When x=0.5389179669047208,y=?
1
When x=0.4984903405356802,y=?
1
When x=0.8770009254960851,y=?
1
When x=0.45269982284910104,y=?
1
When x=0.6850424168303579,y=?
1
When x=0.6730912681959299,y=?
1
When x=0.928994296997916,y=?
1
When x=0.5356037645753124,y=?
1
When x=0.035392897170888005,y=?
1
When x=0.8092257545967552,y=?
1
NOnono~~~

 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

# Author: frieggs
# Blog: https://blog.frieggs.com

from tensorflow.keras.models import Sequential
from tensorflow.keras.layers import Dense, Activation
from tensorflow import keras
import matplotlib.pyplot as plt
import numpy as np
import pandas as pd
import socket
import re
import time

def getMessage():
    time.sleep(0.1)
    content = s.recv(1024).decode()
    print(content)
    return content

def sendMessage(text=""):
    time.sleep(0.1)
    print(text)
    s.send((str(text) + "\n").encode())

def plot_history(history):
  hist = pd.DataFrame(history.history)
  hist['epoch'] = history.epoch
  plt.figure()
  plt.xlabel('Epoch')
  plt.ylabel('Mean Abs Error [MPG]')
  plt.plot(hist['epoch'], hist['mae'],
           label='Train Error')
  plt.ylim([0,2])
  plt.legend()
  plt.show()

s = socket.socket()
s.connect(('node3.buuoj.cn', 25465))
getMessage()
sendMessage("frieggs")
getMessage()
sendMessage()
content = ''
content += getMessage()
content += getMessage()
content += getMessage()
content += getMessage()
print(content)
x = re.findall(r'x=(.*?),y=.*?;', content)
y = re.findall(r'x=.*?,y=(.*?);', content)
x = [float(i) for i in x]
y = [float(i) for i in y]
X_train = np.array(x)
Y_train = np.array(y)

# 构建一个顺序模型
model = Sequential()

# 在模型中添加64单元的全连接层
model.add(Dense(64, input_shape=(1,)))
# 用relu作为激活函数
model.add(Activation("relu"))

model.add(Dense(64))
model.add(Activation("relu"))

model.add(Dense(1))

'''
编译模型，使用adam加速训练
采用均方误差（Mean Square Error，MSE）作为损失函数训练模型
采用平均绝对误差（Mean Absolute Error，MAE）评价模型
'''
model.compile(optimizer='adam', loss='mse', metrics=['mae'])

# 打印模型结构摘要
print(model.summary())

# 训练300轮
history = model.fit(X_train, Y_train, epochs=300)
# 可视化训练过程
plot_history(history)
sendMessage()
# 记录训练集数据的离散点
plt.scatter(X_train, Y_train)
for i in range(10):
    res = re.findall('When x=(.*?),y=?', getMessage())
    X_p = np.array([float(res[0])])
    # 用X预测Y的值
    Y_p = model.predict(X_p)[0][0]
    sendMessage(Y_p)
    # 红色加号标注预测点
    plt.plot(X_p, Y_p, 'r+')
plt.show()
getMessage()
s.close()