Skip to main content

PyTorch functions to improve performance, analyse models and make your life easier.

Project description

torchfunc Logo


Version Docs Tests Coverage Style PyPI Python PyTorch Docker Roadmap
Version Documentation Tests codebeat PyPI Python PyTorch Docker Roadmap

torchfunc is library revolving around PyTorch with a goal to help you with:

  • Improving and analysing performance of your neural network
  • Daily neural network duties (model size, seeding, performance measurements etc.)
  • Plotting and visualizing modules
  • Record neuron activity and tailor it to your specific task or target
  • Get information about your host operating system, CUDA devices and others

Quick examples

  • Seed globaly, Freeze weights, check inference time and model size
# Inb4 MNIST, you can use any module with those functions
model = torch.nn.Linear(784, 10)
frozen = torchfunc.module.freeze(model, bias=False)

with torchfunc.Timer() as timer:
  frozen(torch.randn(32, 784)
  print(timer.checkpoint()) # Time since the beginning
  frozen(torch.randn(128, 784)
  print(timer.checkpoint()) # Since last checkpoint

print(f"Overall time {timer}; Model size: {torchfunc.sizeof(frozen)}")
  • Recorder and sum per-layer activation statistics as data passes through network:
# MNIST classifier
model = torch.nn.Sequential(
    torch.nn.Linear(784, 100),
    torch.nn.ReLU(),
    torch.nn.Linear(100, 50),
    torch.nn.ReLU(),
    torch.nn.Linear(50, 10),
)
# Recorder which sums layer inputs from consecutive forward calls
recorder = torchfunc.record.ForwardPreRecorder(reduction=lambda x, y: x+y)
# Record inputs going into Linear(100, 50) and Linear(50, 10)
recorder.children(model, indices=(2, 3))
# Train your network normally (or pass data through it)
...
# Save tensors (of shape 100 and 50) in folder, each named 1.pt and 2.pt respectively
recorder.save(pathlib.Path("./analysis"))

For performance tips, plotting and other check torchfunc documentation.

Installation

pip

Latest release:

pip install --user torchfunc

Nightly:

pip install --user torchfunc-nightly

Docker

CPU standalone and various versions of GPU enabled images are available at dockerhub.

For CPU quickstart, issue:

docker pull szymonmaszke/torchfunc:18.04

Nightly builds are also available, just prefix tag with nightly_. If you are going for GPU image make sure you have nvidia/docker installed and it's runtime set.

Contributing

If you find any issue or you think some functionality may be useful to others and fits this library, please open new Issue or create Pull Request.

To get an overview of something which one can done to help this project, see Roadmap

Project details


Release history Release notifications | RSS feed

Download files

Download the file for your platform. If you're not sure which to choose, learn more about installing packages.

Source Distribution

torchfunc-nightly-1568790130.tar.gz (19.9 kB view details)

Uploaded Source

Built Distribution

torchfunc_nightly-1568790130-py3-none-any.whl (25.0 kB view details)

Uploaded Python 3

File details

Details for the file torchfunc-nightly-1568790130.tar.gz.

File metadata

  • Download URL: torchfunc-nightly-1568790130.tar.gz
  • Upload date:
  • Size: 19.9 kB
  • Tags: Source
  • Uploaded using Trusted Publishing? No
  • Uploaded via: twine/1.15.0 pkginfo/1.5.0.1 requests/2.22.0 setuptools/41.2.0 requests-toolbelt/0.9.1 tqdm/4.36.0 CPython/3.7.4

File hashes

Hashes for torchfunc-nightly-1568790130.tar.gz
Algorithm Hash digest
SHA256 a49f85f0456df3e3649f8f5cc0aa8963a31e05a170cbeb029d6b81e2468c8b8f
MD5 fa470c111f216e9ad19f42450f04ba0c
BLAKE2b-256 7c87db921151f9094c7546276711f3bdfe972156dd1d732f9316b518f2f96a23

See more details on using hashes here.

File details

Details for the file torchfunc_nightly-1568790130-py3-none-any.whl.

File metadata

  • Download URL: torchfunc_nightly-1568790130-py3-none-any.whl
  • Upload date:
  • Size: 25.0 kB
  • Tags: Python 3
  • Uploaded using Trusted Publishing? No
  • Uploaded via: twine/1.15.0 pkginfo/1.5.0.1 requests/2.22.0 setuptools/41.2.0 requests-toolbelt/0.9.1 tqdm/4.36.0 CPython/3.7.4

File hashes

Hashes for torchfunc_nightly-1568790130-py3-none-any.whl
Algorithm Hash digest
SHA256 498284e83f3ffa62dfd05f20e21fa16d28d8a676a05a47bb7977bd97e4f6e76d
MD5 52a1e90635fa3e5d497717c0c042446a
BLAKE2b-256 ea82df976ee04f2ad0b348579a0173517791a07bb11adccc9f362b39d5e6063e

See more details on using hashes here.

Supported by

AWS AWS Cloud computing and Security Sponsor Datadog Datadog Monitoring Fastly Fastly CDN Google Google Download Analytics Microsoft Microsoft PSF Sponsor Pingdom Pingdom Monitoring Sentry Sentry Error logging StatusPage StatusPage Status page