kaggle_getstarted/Digit_Recognizer/nn_sk.py at master · zx-code123/kaggle_getstarted · GitHub

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
'''
Created on 2018-5-31
Author: zx-code123
Github: https://github.com/zx-code123
'''
import os
import csv
import time
import numpy as np
import pandas as pd
from sklearn.decomposition import PCA
from sklearn.neural_network import MLPClassifier
import matplotlib.pyplot as plt

dataset_dir = "digit_data"

# 读取csv文件
def opencsv():
    # 使用pandas读取
    data_train = pd.read_csv(os.path.join(dataset_dir,"train.csv"))
    data_test = pd.read_csv(os.path.join(dataset_dir,"test.csv"))

    train_df = data_train.values[0:,1:] #读取训练数据
    test_df = data_test.values[0:,0:] # 读取测试数据
    label_df = data_train.values[0:,0] # 读取训练标签
    return train_df,label_df,test_df

# 保存方法一
def saveResult(result,outputName):
    # 创建记录输出结果的文件（w 和 wb 使用的时候有问题）
    with open(outputName,"w") as f:
        # python3里面对 str和bytes类型做了严格的区分，不像python2里面某些函数里可以混用。所以用python3来写wirterow时，打开文件不要用wb模式，只需要使用w模式，然后带上newline=''
        output = csv.writer(f)
        output.writerow(["ImageId","Label"])
        index = 0
        for r in result:
            index += 1
            output.writerow([index,int(r)])
        print("[INFO] save succcessfully")

# PCA降维
def reduction(train_data,test_data):
    pca = PCA(n_components=35, whiten=True)
    train_data = pca.fit_transform(train_data)
    test_data = pca.transform(test_data)
    return train_data,test_data

def nnClassify(train_df,label_df):
    nnClf = clf = MLPClassifier(
        hidden_layer_sizes=(100, ),
        activation='relu',
        alpha=0.0001,
        learning_rate='constant',
        learning_rate_init=0.001,
        max_iter=200,
        shuffle=True,
        random_state=34)
    nnClf.fit(train_df,label_df)
    return nnClf
# 可视化最后10张结果
def showResult(X_test,submit):
    fig = plt.figure(figsize=(10,4))
    # loop over the first 10 images of the test set
    for i in range(10):
        ax = plt.subplot(2, 5, i+1)
        ax.set_title('Prediction: {}'.format(submit.iloc[i].Label))
        ax.axis('off')
        # X_test = pd.DataFrame(X_test)
        image = X_test.iloc[i, :].values.reshape(28, 28)
        imgplot = plt.imshow(image, cmap='binary')

    plt.show()

def main():
    start_time = time.time()
    # 读取数据
    train_df,label_df,test_df = opencsv()
    # 降维
    train_df,test_df = reduction(train_df,test_df)
    s_time1 = time.time()
    print("[INFO] load data cost:{}".format(s_time1-start_time))

    # 训练模型
    nnClf = nnClassify(train_df,label_df)
    # 预测结果
    test_label = nnClf.predict(test_df)
    # 保存方法一
    # saveResult(test_label,"result.csv")
    # 保存方法二
    submit = pd.DataFrame({"ImageId": range(1,len(test_label)+1), "Label": test_label})
    data_test = pd.read_csv(os.path.join(dataset_dir,"test.csv"))
    showResult(data_test,submit)
    submit.to_csv('result.csv', index=False, header=True)
    s_time2 = time.time()
    print("[INFO] classify test_data cost:{}".format(s_time2-s_time1))

if __name__ =='__main__':
    main()