PyTorch

type access

  • Operating System:

  • Terminal:

  • Shell:

  • Editor:

  • Package Manager:

  • Programming Language:

  • Database:

  • NVIDIA Libraries:

type access

  • Operating System:

  • Terminal:

  • Shell:

  • Editor:

  • Package Manager:

  • Programming Language:

  • Database:

  • NVIDIA Libraries:

type access

  • Operating System:

  • Terminal:

  • Shell:

  • Editor:

  • Package Manager:

  • Programming Language:

  • Database:

  • NVIDIA Libraries:

type access

  • Operating System:

  • Terminal:

  • Shell:

  • Editor:

  • Package Manager:

  • Programming Language:

  • Database:

  • NVIDIA Libraries:

PyTorch is a GPU accelerated tensor computational framework with a Python front end. Functionality can be easily extended with common Python libraries such as NumPy, SciPy, and Cython. Automatic differentiation is done with a tape-based system at both a functional and neural network layer level.

For basic usage of the Notebook environment, check the JupyterLab application.

For basic usage of the console environment, check the Terminal application.

Install new packages

Additional packages can be installed inside the application container using the Dependencies parameter. The user should provide the list of packages via a text file (.txt). The installation is done via the pip command line Python package manager. Alternatively, it is possible to load a Bash script (*.sh) with the list of shell commands to be used for the installation.

The example below shows two different ways to install the same packages:

numpy==1.18.1
pandas==1.0.2
keras==2.3.1
matplotlib==3.2.0
seaborn==0.10.0
plotly==4.5.4
#!/usr/bin/env bash

set -eux

pip install \
numpy==1.18.1 \
pandas==1.0.2 \
keras==2.3.1 \
matplotlib==3.2.0 \
seaborn==0.10.0 \
plotly==4.5.4

Batch mode

Use this option to submit a Bash script (*.sh), which will be executed after the job starts. The job will stop after the execution of the program.

Start a terminal session

Instead of the default Jupyter Notebook, the user can start a terminal browser via the Terminal Interface parameter.

Run on GPU nodes

GPU partitions can be selected from the application frontend page using the Machine type parameter.

The NVIDIA System Management Interface program, nvidia-smi, can be used to monitor the usage of the GPU resources:

$ watch -n 0.5 nvidia-smi