Question

我有一个包含以下数据的txt文件：

43,70,90,12,98,54,23
80,100,21,67
23,45
30

我想对数据进行填充，以便数据具有相同的长度，例如我想使用数字0填充，以便输出应该是：

43,70,90,12,98,54,23
80,100,21,67,0,0,0
23,45,0,0,0,0,0
30,0,0,0,0,0,0

在Python中执行此操作的最佳方法是什么？

Answer 1

这是一种方法：

确定文本/文件的最大列数
将其他列附加到没有最大数量的行

从文本文件中读取的示例代码：

#!/usr/bin/env python3

output = ""
with open("1.txt") as f:
    # Determine number of columns
    cols = max([len(line.split(",")) for line in f])
    print("Maximum number of columns: %d" % cols)

    # Build output string, starting from beginning of file
    f.seek(0)
    for line in f:
        output += line.strip() + ",0" * (cols - len(line.split(","))) + "\n"
print("Output:", output)

从字符串文字中读取的示例代码：

#!/usr/bin/env python3

# You could read text from file. Here we use a literal string
text = """43,70,90,12,98,54,23
80,100,21,67
23,45
30
"""

# Split text to lines
lines = text.splitlines(False)

# Determine number of columns
cols = max([len(line.split(",")) for line in lines])
print("Maximum number of columns: %d" % cols)

output = ""
for line in lines:
    output += line + ",0" * (cols - len(line.split(","))) + "\n"

print("Output:", output)

这两个例子都使用列表推导。

要详细了解它们的工作原理，请阅读以下内容：

Python Documentation - 5.1.4. List Comprehensions

Python: List Comprehensions

Answer 2

你去了：

import numpy as np
import csv

input_list = []  # list to keep track of input
max_number_of_columns = 0  # maximum number of columns
number_of_rows = 0

with open("test.txt") as testfile:
    for line in testfile:
        # read line, strip end of line and split at the comma
        line = line.replace("\n", "") 
        values = line.split(",")

        number_of_rows += 1
        if len(values) > max_number_of_columns:
            max_number_of_columns = len(values)
        print values
        input_list.append(values)  # add new values to the list
# initialize all zero numpy array
np_array = np.zeros((number_of_rows, max_number_of_columns))
for i, row in enumerate(input_list):
    np_array[i,:len(row)] = row  # write entries into the array

在python中填充数据

2 个答案: