I am trying to run a pytorch code that was written in python 2.7.
So I setup anaconda2 and created a new python env within it. I check the python version.
I then go to the pytorch page and generate the conda install command using linux, conda, Python 2.7 and CUDA 8 as parameters. The command is conda install pytorch torchvision -c pytorch. After executing the command, I see that the python version has changed to 3.6 and which python is now pointing to the python within the newly created env.
TLDR; I can’t seem to have pytorch installed with python 2.7. Can some one help me navigate through this?
An observation: changing between python version to generate the conda install command on the pytorch home page does not change anything in the command. Not sure if anaconda will automatically decipher the python version that it should use? In any case, that does not seem to be working in my case.
@ptrblck, it says, “All packages are already installed” . I guess the problem is, the conda install commands always install PyTorch with Py3.6. What should I change in the conda install command so that I can install PyTorch with Py2.7 and not Py3.6?