Installing PyTorch

Hi everybody! I am having troubles installing PyTorch and fastai locally.

I am using Start Locally | PyTorch to get the commands to install, but I am getting the error:

“ERROR: Could not find a version that satisfies the requirement torchaudio (from versions: none)
ERROR: No matching distribution found for torchaudio”, when using the pip version, whereas when using the conda version I get:

"Solving environment: failed with initial frozen solve. Retrying with flexible solve.
Collecting package metadata (repodata.json): done
PackagesNotFoundError: The following packages are not available from current channels:

  • torchaudio
  • pytorch"

Even if I get the commands from Anaconda’s website the issues persist. Any clue as to what can I do?

Thanks in advance for your help!

Hi seba92,
I’m having a similar issue with PyTorch. When I pip install torch it says:

C:\Users\92175>pip install torch
Collecting torch
  Using cached torch-0.1.2.post2.tar.gz (128 kB)
Requirement already satisfied: pyyaml in c:\python\python37-32\lib\site-packages (from torch) (5.4.1)
WARNING: The candidate selected for download or install is a yanked version: 'torch' candidate (version 0.1.2.post2 at (from
Reason for being yanked: 0.1.2 is past it's support date and confuses users on unsupported platforms
Building wheels for collected packages: torch
  Building wheel for torch ( ... error

So it appears that I already have torch. Maybe(?)
If I try to install PyTorch I get:

Exception: You tried to install "pytorch". The package named for PyTorch is "torch"

When I click RUN in 01_intro.ipnb Running your first notebook I get a long error ending with

ModuleNotFoundError: No module named 'torch'

Hi Wes!

In Welcome to fastai | fastai there is a “Windows Support” section, there you will find out the installion command needs to change if you are running Windows. Hope its useful for you too!

1 Like