torch中文教程报错：ModuleNotFoundError: No module named 'sgmllib'

Question

pytorch的官方中文教程，执行到下面的代码就报错：
我进行安装transformer：

!pip install transforms

代码部分：

import transforms as T
def get_transform(train):
    transforms = []
    transforms.append(T.ToTensor())
    if train:
        transforms.append(T.RandomHorizontalFlip(0.5))
    return T.Compose(transforms)

报错代码：

---------------------------------------------------------------------------
ModuleNotFoundError                       Traceback (most recent call last)
<ipython-input-11-8f09a7ff1c98> in <module>
----> 1 import transforms as T
      2 
      3 def get_transform(train):
      4     transforms = []
      5     transforms.append(T.ToTensor())

~/.local/lib/python3.6/site-packages/transforms/__init__.py in <module>
----> 1 from .safe_html import safe_html, bodyfinder
      2 
      3 # modules = [
      4 #     'st',             # zopish
      5 #     'rest',           # docutils

~/.local/lib/python3.6/site-packages/transforms/safe_html.py in <module>
----> 1 from sgmllib import SGMLParser, SGMLParseError
      2 import re
      3 from cgi import escape
      4 
      5 from .utils import safeToInt

ModuleNotFoundError: No module named 'sgmllib'

Answer 1

pytorch的torchvision自带有transforms不用，为什么要用别的呢？安装对应的torchvision就行了