数据处理（过采样）

# coding: utf-8
import os
import sys
import matplotlib.pyplot as plt
#将全部数据索引至字典中
path = "/media/dell/dell/data/remote_sensing/remote/train_image"
dirs = sorted(os.listdir(path))
print(dirs)
files = {}
for index, dir in enumerate(dirs):
    path_ = path + "/" + dir + "/"
    files[int(dir)] = []
    for file in os.listdir(path_):
        files[int(dir)].append(path_+file)
    sys.stdout.write('\r>> Loading data %d/%d'%(index+1, 9))
    sys.stdout.flush()
sys.stdout.write("\n")    
#print(len(files))
#查看各种数据所占比例
file_num = []
for i in range(1, 10):
    file_num.append(len(files[i]))
print(file_num)
plt.bar(range(len(file_num)),file_num)
plt.xlabel('class_id')
plt.ylabel('amount')
plt.ylim(0, 11000)
for x,y in zip(range(len(file_num)),file_num):
    plt.text(x, y+100, '%d' % y, ha='center', va= 'bottom')
plt.show()
#写入valid数据
import rand