Data

Tools

Indexes

Applications

Experiments

Open Source Science

torchode

A parallel ODE solver for PyTorch

https://github.com/martenlienen/torchode

Science Score: 44.0%

This score indicates how likely this project is to be science-related based on various indicators:

  • ✓
    CITATION.cff file
    Found CITATION.cff file
  • ✓
    codemeta.json file
    Found codemeta.json file
  • ✓
    .zenodo.json file
    Found .zenodo.json file
  • ○
    DOI references
  • ○
    Academic publication links
  • ○
    Academic email domains
  • ○
    Institutional organization owner
  • ○
    JOSS paper metadata
  • ○
    Scientific vocabulary similarity
    Low similarity (12.2%) to scientific vocabulary

Keywords

deep-learning neural-ode pytorch
Last synced: 10 months ago · JSON representation ·

Repository

A parallel ODE solver for PyTorch

Basic Info
Statistics
  • Stars: 264
  • Watchers: 1
  • Forks: 20
  • Open Issues: 0
  • Releases: 0
Topics
deep-learning neural-ode pytorch
Created over 3 years ago · Last pushed over 1 year ago
Metadata Files
Readme Changelog License Citation

README.md

A Parallel ODE Solver for PyTorch

pytest

torchode is a suite of single-step ODE solvers such as dopri5 or tsit5 that are compatible with PyTorch's JIT compiler and parallelized across a batch. JIT compilation often gives a performance boost, especially for code with many small operations such as an ODE solver, while batch-parallelization means that the solver can take a step of 0.1 for one sample and 0.33 for another, depending on each sample's difficulty. This can avoid performance traps for models of varying stiffness and ensures that the model's predictions are independent from the compisition of the batch. See the paper for details.

If you get stuck at some point, you think the library should have an example on x or you want to suggest some other type of improvement, please open an issue on github.

Installation

You can get the latest released version from PyPI with

sh pip install torchode

To install a development version, clone the repository and install in editable mode:

sh git clone https://github.com/martenlienen/torchode cd torchode pip install -e .

Usage

```python import matplotlib.pyplot as pp import torch import torchode as to

def f(t, y): return -0.5 * y

y0 = torch.tensor([[1.2], [5.0]]) nsteps = 10 teval = torch.stack((torch.linspace(0, 5, nsteps), torch.linspace(3, 4, nsteps)))

term = to.ODETerm(f) stepmethod = to.Dopri5(term=term) stepsizecontroller = to.IntegralController(atol=1e-6, rtol=1e-3, term=term) solver = to.AutoDiffAdjoint(stepmethod, stepsizecontroller) jit_solver = torch.compile(solver)

sol = jitsolver.solve(to.InitialValueProblem(y0=y0, teval=t_eval)) print(sol.stats)

=> {'nfevals': tensor([26, 26]), 'n_steps': tensor([4, 2]),

=> 'naccepted': tensor([4, 2]), 'ninitialized': tensor([10, 10])}

pp.plot(sol.ts[0], sol.ys[0]) pp.plot(sol.ts[1], sol.ys[1]) ```

Citation

If you build upon this work, please cite the following paper.

@inproceedings{lienen2022torchode, title = {torchode: A Parallel {ODE} Solver for PyTorch}, author = {Marten Lienen and Stephan G{\"u}nnemann}, booktitle = {The Symbiosis of Deep Learning and Differential Equations II, NeurIPS}, year = {2022}, url = {https://openreview.net/forum?id=uiKVKTiUYB0} }

Owner

  • Name: Marten Lienen
  • Login: martenlienen
  • Kind: user
  • Location: Germany
  • Company: TUM

Citation (CITATION.cff) 

cff-version: 1.2.0
title: torchode
message: >-
  If you use this software, please cite it using the
  metadata from this file.
type: software
authors:
  - given-names: Marten
    family-names: Lienen
    email: m.lienen@tum.de
  - given-names: Stephan
    family-names: Günnemann
    email: s.guennemann@tum.de
repository-code: "https://github.com/martenlienen/torchode"
license: MIT
preferred-citation:
  type: conference-paper
  title: "torchode: A Parallel ODE Solver for PyTorch"
  authors:
    - given-names: Marten
      family-names: Lienen
      email: m.lienen@tum.de
    - given-names: Stephan
      family-names: Günnemann
      email: s.guennemann@tum.de
  collection-title: "The Symbiosis of Deep Learning and Differential Equations II, NeurIPS"
  year: 2022
  url: "https://openreview.net/forum?id=uiKVKTiUYB0"

GitHub Events

Total
  • Issues event: 6
  • Watch event: 33
  • Issue comment event: 12
  • Fork event: 8
Last Year
  • Issues event: 6
  • Watch event: 33
  • Issue comment event: 12
  • Fork event: 8

Issues and Pull Requests

Last synced: 10 months ago

All Time
  • Total issues: 3
  • Total pull requests: 0
  • Average time to close issues: about 15 hours
  • Average time to close pull requests: N/A
  • Total issue authors: 2
  • Total pull request authors: 0
  • Average comments per issue: 2.67
  • Average comments per pull request: 0
  • Merged pull requests: 0
  • Bot issues: 0
  • Bot pull requests: 0
Past Year
  • Issues: 3
  • Pull requests: 0
  • Average time to close issues: about 15 hours
  • Average time to close pull requests: N/A
  • Issue authors: 2
  • Pull request authors: 0
  • Average comments per issue: 2.67
  • Average comments per pull request: 0
  • Merged pull requests: 0
  • Bot issues: 0
  • Bot pull requests: 0
View more stats
Top Authors
Issue Authors
  • Ilykuleshov (3)
  • timnotavailable (2)
  • 1ho0jin1 (1)
  • zjowowen (1)
  • wangmiaowei (1)
  • jucor (1)
  • gisilvs (1)
  • m13ammed (1)
  • kschischo (1)
  • chansigit (1)
  • janrohleff (1)
  • tuchris (1)
Pull Request Authors
  • Ilykuleshov (2)
  • wootwootwootwoot (1)
Top Labels
Issue Labels
Pull Request Labels

Packages

  • Total packages: 1
  • Total downloads:
    • pypi 2,685 last-month
  • Total dependent packages: 3
  • Total dependent repositories: 2
  • Total versions: 14
  • Total maintainers: 1
pypi.org: torchode

A parallel ODE solver for PyTorch

  • Versions: 14
  • Dependent Packages: 3
  • Dependent Repositories: 2
  • Downloads: 2,685 Last month
Rankings
Dependent packages count: 4.8%
Average: 8.5%
Downloads: 9.3%
Dependent repos count: 11.5%
Maintainers (1)
cqql
Last synced: over 1 year ago

Dependencies

.github/workflows/python-package.yml actions
  • actions/checkout v3 composite
  • actions/setup-python v3 composite
docs/requirements.in pypi
  • mkdocs ==1.4.1
  • mkdocs-jupyter *
docs/requirements.txt pypi
  • attrs ==22.1.0
  • beautifulsoup4 ==4.11.1
  • bleach ==5.0.1
  • certifi ==2022.12.7
  • charset-normalizer ==2.1.1
  • click ==8.1.3
  • defusedxml ==0.7.1
  • entrypoints ==0.4
  • fastjsonschema ==2.16.2
  • ghp-import ==2.1.0
  • idna ==3.4
  • jinja2 ==3.1.2
  • jsonschema ==4.17.0
  • jupyter-client ==7.4.4
  • jupyter-core ==4.11.2
  • jupyterlab-pygments ==0.2.2
  • jupytext ==1.14.1
  • lxml ==4.9.1
  • markdown ==3.3.7
  • markdown-it-py ==2.1.0
  • markupsafe ==2.1.1
  • mdit-py-plugins ==0.3.1
  • mdurl ==0.1.2
  • mergedeep ==1.3.4
  • mistune ==0.8.4
  • mkdocs ==1.4.1
  • mkdocs-jupyter ==0.22.0
  • mkdocs-material ==8.5.7
  • mkdocs-material-extensions ==1.1
  • nbclient ==0.7.0
  • nbconvert ==6.5.4
  • nbformat ==5.7.0
  • nest-asyncio ==1.5.6
  • packaging ==21.3
  • pandocfilters ==1.5.0
  • pygments ==2.13.0
  • pymdown-extensions ==9.7
  • pyparsing ==3.0.9
  • pyrsistent ==0.19.1
  • python-dateutil ==2.8.2
  • pyyaml ==6.0
  • pyyaml-env-tag ==0.1
  • pyzmq ==24.0.1
  • requests ==2.28.1
  • six ==1.16.0
  • soupsieve ==2.3.2.post1
  • tinycss2 ==1.2.1
  • toml ==0.10.2
  • tornado ==6.2
  • traitlets ==5.5.0
  • urllib3 ==1.26.12
  • watchdog ==2.1.9
  • webencodings ==0.5.1
pyproject.toml pypi
  • functorch *
  • sympy ~= 1.10
  • torch ~= 1.11
  • torchtyping ~= 0.1.4