kaggle之手写体识别

kaggle地址

数据预览

首先载入数据集

import pandas as pd
import numpy as np

train = pd.read_csv('/Users/frank/Documents/workspace/kaggle/dataset/digit_recognizer/train.csv')
test = pd.read_csv('/Users/frank/Documents/workspace/kaggle/dataset/digit_recognizer/test.csv')
print train.head()
print test.head()

   label  pixel0  pixel1  pixel2  pixel3  pixel4  pixel5  pixel6  pixel7  \
0      1       0       0       0       0       0       0       0       0
1      0       0       0       0       0       0       0       0       0
2      1       0       0       0       0       0       0       0       0
3      4       0       0       0       0       0       0       0       0
4      0       0       0       0       0       0       0       0       0   

   pixel8    ...     pixel774  pixel775  pixel776  pixel777  pixel778  \
0       0    ...            0         0         0         0         0
1       0    ...            0         0         0         0         0
2       0    ...            0         0         0         0         0
3       0    ...            0         0         0         0         0
4       0    ...            0         0         0         0         0   

   pixel779  pixel780  pixel781  pixel782  pixel783
0         0         0         0         0         0
1         0         0         0         0         0
2         0         0         0         0         0
3         0         0         0         0         0
4         0         0         0         0         0  

[5 rows x 785 columns]
   pixel0  pixel1  pixel2  pixel3  pixel4  pixel5  pixel6  pixel7  pixel8  \
0       0       0       0       0       0       0       0       0       0
1       0       0       0       0       0       0       0       0       0
2       0       0       0       0       0       0       0       0       0
3       0       0       0       0       0       0       0       0       0
4       0       0       0       0       0       0       0       0       0   

   pixel9    ...     pixel774  pixel775  pixel776  pixel777  pixel778  \
0       0    ...            0         0         0         0         0
1       0    ...            0         0         0         0         0
2       0    ...            0         0         0         0         0
3       0    ...            0         0         0         0         0
4       0    ...            0         0         0         0         0   

   pixel779  pixel780  pixel781  pixel782  pixel783
0         0         0         0         0         0
1         0         0         0         0         0
2         0         0         0         0         0
3         0         0         0         0         0
4         0         0         0         0         0  

[5 rows x 784 columns]

分离训练数据和标签：

train_data = train.values[:,1:]
label = train.ix[:,0]
test_data = test.values

使用PCA来降维：PCA文档

使用SVM来训练：SVM文档

降维

from sklearn.decomposition import PCA
from sklearn.svm import SVC
pca = PCA(n_components=0.8, whiten=True)
# pca.fit(train_data)
train_data = pca.fit_transform(train_data)
# pca.fit(test_data)
test_data = pca.transform(test_data)

SVM训练

print('使用SVM进行训练...')
svc = SVC(kernel='rbf',C=2)
svc.fit(train_data, label)
print('训练结束.')

使用SVM进行训练...
训练结束.

print('对测试集进行预测...')
predict = svc.predict(test_data)
print('预测结束.')

对测试集进行预测...
预测结束.

保存结果：

pd.DataFrame(
    {"ImageId": range(1, len(predict) + 1), "Label": predict}
).to_csv('output.csv', index=False, header=True)

print 'done.'

done.