PyTorch is a Python package that provides two high-level features:
- Tensor computation (like NumPy) with strong GPU acceleration
- Deep neural networks built on a tape-based autograd system
PyTorch has a distant connection with Torch, but for all practical purposes you can treat them as separate packages.
Latest available wheels
To see the latest version of PyTorch that we have built:
[name@server ~]$ avail_wheels "torch*"
For more information on listing wheels, see listing available wheels.
The preferred option is to install it using the python wheel that we compile, as follows:
- 1. Load a python module, either python/2.7, python/3.5, python/3.6 or python/3.7
- 2. Create and start a virtual environment.
- 3. Install PyTorch in the virtual environment with
(venv) [name@server ~] pip install numpy torch_gpu --no-index
(venv) [name@server ~] pip install numpy torch_cpu --no-index
Note: Do not install both torch_cpu and torch_gpu.
In addition to torch_cpu or torch_gpu, you can install torchvision, torchtext and torchaudio:
(venv) [name@server ~] pip install numpy six torch_cpu torchvision torchtext torchaudio --no-index
libtorch.so is included in the wheel. Once Pytorch is installed in a virtual environment, you can find it at: venv/lib/python3.6/site-packages/torch/lib/libtorch.so where venv is the virtual environment.
Once the setup is completed, you can submit a PyTorch job with
[name@server ~]$ sbatch pytorch-test.sh
Here is an example of a job submission script using the python wheel, with a virtual environment in $HOME/pytorch:
#!/bin/bash #SBATCH --gres=gpu:1 # Request GPU "generic resources" #SBATCH --cpus-per-task=6 # Cores proportional to GPUs: 6 on Cedar, 16 on Graham. #SBATCH --mem=32000M # Memory proportional to GPUs: 32000 Cedar, 64000 Graham. #SBATCH --time=0-03:00 #SBATCH --output=%N-%j.out module load python/3.6 source $HOME/pytorch/bin/activate python ./pytorch-test.py
The Python script
pytorch-test.py has the form
import torch x = torch.Tensor(5, 3) print(x) y = torch.rand(5, 3) print(y) # let us run the following only if CUDA is available if torch.cuda.is_available(): x = x.cuda() y = y.cuda() print(x + y)
Dependency torch not found
Other packages that depend on torch will fail to install; you can find instructions here to install such packages.