机器学习算法训练时涉及到的文件读取问题汇总（更新中）

问题描述：

现有一个文件夹下多个CSV数据文件，他们的命名规则如下图

现在要循环读出他们的内部内容合并成为一个文件（或者一个矩阵）

代码：

import os
import numpy as np
from pandas import read_csv

path="/home/chen/PycharmProjects/avoidance/data"  #待读取的文件夹
path_list=os.listdir(path)
path_list.sort()    #对读取的路径进行排序，注意这样的文件最开始命名只能是 00 01 这样

f1 = open("/home/chen/PycharmProjects/avoidance/data/ImData_B00.csv", encoding='UTF-8')
data1 = np.array(read_csv(f1))
X = data1[:, 0:10]  # 数据 
y = data1[:, 10]    # label

for filename in path_list:
	f = open(os.path.join(path,filename), encoding='UTF-8')
	"""
	data = np.array(read_csv(f))
	X1 = data[:, 0:10]
	y1 = data[:, 10]
	X = np.r_[X,X1]     # np.r_[] 的用法就是把两个矩阵按列拼接
	y = np.r_[y,y1]     # 相应的按行拼接是 np.c_[]
	"""
	print(os.path.join(path,filename))

输出结果：

/home/chen/anaconda3/bin/python /home/chen/PycharmProjects/avoidance/tryagain.py
/home/chen/PycharmProjects/avoidance/data/ImData_B00.csv
/home/chen/PycharmProjects/avoidance/data/ImData_B01.csv
/home/chen/PycharmProjects/avoidance/data/ImData_B02.csv
/home/chen/PycharmProjects/avoidance/data/ImData_B03.csv
/home/chen/PycharmProjects/avoidance/data/ImData_B04.csv
/home/chen/PycharmProjects/avoidance/data/ImData_B05.csv
/home/chen/PycharmProjects/avoidance/data/ImData_B06.csv
/home/chen/PycharmProjects/avoidance/data/ImData_B07.csv
/home/chen/PycharmProjects/avoidance/data/ImData_B08.csv
/home/chen/PycharmProjects/avoidance/data/ImData_B09.csv
/home/chen/PycharmProjects/avoidance/data/ImData_B10.csv
/home/chen/PycharmProjects/avoidance/data/ImData_B11.csv
/home/chen/PycharmProjects/avoidance/data/ImData_B12.csv
/home/chen/PycharmProjects/avoidance/data/ImData_B13.csv
/home/chen/PycharmProjects/avoidance/data/ImData_B14.csv
/home/chen/PycharmProjects/avoidance/data/ImData_B15.csv
/home/chen/PycharmProjects/avoidance/data/ImData_B16.csv
/home/chen/PycharmProjects/avoidance/data/ImData_B17.csv
/home/chen/PycharmProjects/avoidance/data/ImData_B18.csv
/home/chen/PycharmProjects/avoidance/data/ImData_B19.csv
/home/chen/PycharmProjects/avoidance/data/ImData_B20.csv
/home/chen/PycharmProjects/avoidance/data/ImData_B21.csv
/home/chen/PycharmProjects/avoidance/data/ImData_B22.csv
/home/chen/PycharmProjects/avoidance/data/ImData_B23.csv
/home/chen/PycharmProjects/avoidance/data/ImData_B24.csv
/home/chen/PycharmProjects/avoidance/data/ImData_B25.csv
/home/chen/PycharmProjects/avoidance/data/ImData_B26.csv
/home/chen/PycharmProjects/avoidance/data/ImData_B27.csv
/home/chen/PycharmProjects/avoidance/data/ImData_B28.csv
/home/chen/PycharmProjects/avoidance/data/ImData_B29.csv
/home/chen/PycharmProjects/avoidance/data/ImData_B30.csv
/home/chen/PycharmProjects/avoidance/data/ImData_B31.csv
/home/chen/PycharmProjects/avoidance/data/ImData_B32.csv
/home/chen/PycharmProjects/avoidance/data/ImData_B33.csv
/home/chen/PycharmProjects/avoidance/data/ImData_B34.csv

Process finished with exit code 0

关于numpy的一点总结：

np.r_是按行连接两个矩阵，就是把两矩阵上下相加，要求列数相等，类似于pandas中的concat()

np.c_是按列连接两个矩阵，就是把两矩阵左右相加，要求行数相等，类似于pandas中的merge()