Question

我有一个txt文件，可以显示为：

10 1:0.870137474304 2:0.722354071782 3:0.671913562758 
11 1:0.764133072717 2:0.4893616821 3:0.332713609364 
20 1:0.531732713984 2:0.0967819558321 3:0.169802773309

然后我想读取文件并以下列形式形成矩阵：

[[10 0.870137474304 0.722354071782 0.671913562758 ]
[11 0.764133072717 0.4893616821   0.332713609364 ]
[20 0.531732713984 0.0967819558321 0.169802773309]]

我知道如何拆分第一列以外的元素。如何处理第一栏？

matrix = []

lines = open("test.txt").read().split("\n")  # read all lines into an array
for line in lines:
    array [0] = line.split(" ")[0]
    # Split the line based on spaces and the sub-part on the colon
    array = [float(s.split(":")[1]) for s in line.split(" ")]  

    matrix.append(array)

print(matrix)

Answer 1

您可以使用正则表达式：

import re
data = [map(float, re.findall('(?<=:)[\d\.]+|^\d+', i.strip('\n'))) for i in open('filename.txt')]

输出：

[[10.0, 0.870137474304, 0.722354071782, 0.671913562758], [11.0, 0.764133072717, 0.4893616821, 0.332713609364], [20.0, 0.531732713984, 0.0967819558321, 0.169802773309]]

修改：使用numpy创建data数组：

import numpy as np
import re
data = [map(float, re.findall('(?<=:)[\d\.]+|^\d+', i.strip('\n'))) for i in open('filename.txt')]
new_data = np.array(data)

输出：

array([[ 10.        ,   0.87013747,   0.72235407,   0.67191356],
   [ 11.        ,   0.76413307,   0.48936168,   0.33271361],
   [ 20.        ,   0.53173271,   0.09678196,   0.16980277]])

Answer 2

以下是将数据提取为numpy数组的一种方法：

df = pd.read_csv('myfile.csv', header=None)

for col in range(1, 4):
    df[col] = df[col].apply(lambda x: float(x.split(':')[1]))

res = df.values

# [[ 10.           0.87013747   0.72235407   0.67191356]
#  [ 11.           0.76413307   0.48936168   0.33271361]
#  [ 20.           0.53173271   0.09678196   0.16980277]]

Answer 3

对于python中的初学者

富有表现力的版本：

import csv
matrix = []
with open('data.csv') as csvfile:
    reader = csv.reader(csvfile, delimiter=' ')
    for row in reader:
        cleaned_row = [col.split(':')[-1] for col in row]
        matrix.append(cleaned_row)
print matrix

使用列表理解

rows = [row for row in open('csvfile.csv').read().split('\n')]
matrix = [[col.split(':')[-1] for col in row.split(' ')] for row in rows]

从.txt文件中读取矩阵？

3 个答案: