Python 进阶学习

1、VsCode Debug 配置文件

配置当前路径、命令行参数

"configurations": [
        {
            "name": "Python: 当前文件",
            "type": "python",
            "request": "launch",
            "program": "${file}",
            "console": "integratedTerminal",
            "args": [
                "--data", "./data/", 
                "--output", "./output/", 
                "--lang", "en"],
            "cwd": "${fileDirname}"
        }
    ]

2、生成requirements.txt

1 2	python -m pip install pipreqs pipreqs ./ --encoding=utf-8

3、单星号、双星号

单星号（*）：*agrs，将所以参数以元组(tuple)的形式导入

双星号（**）：**kwargs，将参数以字典的形式导入

def foo(param1, *param2):
    print(param1)
    print(param2)
    print(param2[2])
foo(1, 2, 3, 4, 5)

def bar(param1, **param2):
    print(param1)
    print(param2)
    print(param2["a"])
bar(1, a=2, b=3)

"""
输出：
1
(2, 3, 4, 5)
4
1
{'a': 2, 'b': 3}
2
"""

4、list, np.array, torch.tensor 相互转换

ndarray = np.array(list)  # list 转 numpy数组
list = ndarray.tolist()  # numpy 转 list

tensor=torch.Tensor(list)  # list 转 torch.Tensor
list = tensor.numpy().tolist()  # torch.Tensor 转 list  先转numpy，后转list

ndarray = tensor.cpu().numpy()  # torch.Tensor 转 numpy  *gpu上的tensor不能直接转为numpy
tensor = torch.from_numpy(ndarray)  # numpy 转 torch.Tensor

5、np.array, torch.tensor 数组嵌套

import numpy as np
a = [1, 2, 3]
b = [0, 1, 0]
a = np.array(a)
b = np.array(b)
c = a[b]
print(c)
# 输出：[1 2 1]


import torch
a = [1, 2, 3]
b = [0, 1, 0]
a = torch.Tensor(a)
b = torch.Tensor(b)
b = b.gt(0)
c = a[b]
print(c)
# 输出：tensor([2.])

6、Python 执行 shell

import os
import subprocess

x1 = os.system('pwd')  # 直接输出shell结果，不会赋值给x1
print('x1', x1)  # 打印结果：0, 表示命令执行成功

x2 = os.popen('pwd').read()  # 不直接输出shell结果，会赋值给x2
print('x2', x2)  # 打印结果：/home/zzk, os.popen() 返回的是一个【文件对象】

x3 = subprocess.call('pwd', shell=True)  # 直接输出shell结果
print('x3', x3)  # 打印结果：0

subprocess.call('echo 123', shell=True)
subprocess.call(['echo', '456'], shell=False)
x4 = subprocess.Popen('cd ../ && pwd', stdout=subprocess.PIPE, shell=True)
print('x4', x4.communicate())  # 打印结果

7、卸载所有三方包

1
2
3

pip list
pip freeze>python_modules.txt
pip uninstall -r python_modules.txt -y