FileNotFoundError: python setup.py develop

I try python setup.py develop and python setup.py build develop ,but both get the error

Traceback (most recent call last):
  File "setup.py", line 719, in <module>
    build_deps()
  File "setup.py", line 285, in build_deps
    build_dir='build')
  File "C:\Users\Jay\pytorch\tools\build_pytorch_libs.py", line 259, in build_caffe2
    my_env)
  File "C:\Users\Jay\pytorch\tools\build_pytorch_libs.py", line 239, in run_cmake
    check_call(cmake_args, cwd=build_dir, env=my_env)
  File "C:\Users\Jay\Anaconda3\lib\subprocess.py", line 306, in check_call
    retcode = call(*popenargs, **kwargs)
  File "C:\Users\Jay\Anaconda3\lib\subprocess.py", line 287, in call
    with Popen(*popenargs, **kwargs) as p:
  File "C:\Users\Jay\Anaconda3\lib\subprocess.py", line 729, in __init__
    restore_signals, start_new_session)
  File "C:\Users\Jay\Anaconda3\lib\subprocess.py", line 1017, in _execute_child
    startupinfo)
FileNotFoundError: [WinError 2] 系統找不到指定的檔案。

someone knows how to deal with this?
python version: 3.6.8
anaconda version: 1.7.2

It happened to me when I forgot to install some python packages.

pip install --upgrade setuptools

and

pip install astunparse numpy ninja pyyaml mkl mkl-include setuptools cmake cffi typing_extensions future six requests dataclasses

solved the issue. If you’re using Anaconda just replace pip with conda in the second line. More info here: Building PyTorch from source on Windows to work with an old GPU