Shortcuts

torchfunc

torchfunc is PyTorch oriented library with a goal to help you with:

  • Improving and analysing performance of your neural network

  • Record/analyse internal state of torch.nn.Module as data passes through it

  • Do the above based on external conditions (using single Callable to specify it)

  • Day-to-day neural network related duties (model size, seeding, performance measurements etc.)

  • Get information about your host operating system, CUDA devices and others

See related projects for other helpful libraries revolving around PyTorch.

Installation

Following installation methods are available:

pip:

To install latest release:

pip install --user torchfunc

To install nightly version:

pip install --user torchfunc-nightly

Docker:

Various torchfunc images are available both CPU and GPU-enabled. You can find them in Docker Cloud at szymonmaszke/torchfunc

CPU

CPU image is based on ubuntu:18.04 and official release can be pulled with:

docker pull szymonmaszke/torchfunc:18.04

For nightly release:

docker pull szymonmaszke/torchfunc:nightly_18.04

This image is significantly lighter due to lack of GPU support.

GPU

All images are based on nvidia/cuda Docker image. Each has corresponding CUDA version tag ( 10.1, 10 and 9.2) CUDNN7 support and base image ( ubuntu:18.04 ).

Following images are available:

  • 10.1-cudnn7-runtime-ubuntu18.04

  • 10.1-runtime-ubuntu18.04

  • 10.0-cudnn7-runtime-ubuntu18.04

  • 10.0-runtime-ubuntu18.04

  • 9.2-cudnn7-runtime-ubuntu18.04

  • 9.2-runtime-ubuntu18.04

Example pull:

docker pull szymonmaszke/torchfunc:10.1-cudnn7-runtime-ubuntu18.04

You can use nightly builds as well, just prefix the tag with nightly_, for example like this:

docker pull szymonmaszke/torchfunc:nightly_10.1-cudnn7-runtime-ubuntu18.04

conda:

TO BE ADDED

conda install -c conda-forge torchfunc