Skip to main content

OmniOpt2 - Hyperparameter Optimizer for SLURM-based Systems

OmniOpt2 is a tool designed to assist researchers, engineers, and data scientists with hyperparameter optimization on SLURM-based clusters, even though it works without it as well. It simplifies large-scale optimization tasks with built-in fault tolerance and flexibility. A graphical user interface (GUI) is available for command creation, accessible at OmniOpt2 GUI. For tutorials on configuration, exit codes, and debugging, visit OmniOpt2 Tutorials.

Main program

omniopt --partition=alpha --experiment_name=example --mem_gb=1 --time=60 \
    --worker_timeout=60 --max_eval=500 --num_parallel_jobs=500 --gpus=1 \
    --follow --run_program=$(echo 'echo "RESULT: %(param)"' | base64 -w0) \
    --parameter param range 0 1000 float

This command initiates OmniOpt2 and installs dependencies if not already installed. The parameter --run_program uses a Base64-encoded string to specify commands. It is recommended to use the GUI, though.

Plot Results

Generates visualizations, such as scatter and hex scatter plots. --min and --max adjust the plotted result value range.

Or, with --min and --max:

omniopt_plot --run_dir runs/example/0
omniopt_plot --run_dir runs/example/0 --min 0 --max 100

Use Optuna directly

Every Optuna sampler (TPE, CMA-ES, GP, Random, Grid, NSGA-II, NSGA-III, MOTPE, ...) is reachable as a special case of OmniOpt through the omniopt_optuna entry point:

omniopt_optuna --max-eval=20 --parameter 'x range -10 10 int' \
    --run-program 'echo "RESULT: $x"'
omniopt_optuna --model=OPTUNA_NSGAII --max-eval=30 \
    --result-names RESULT1=min RESULT2=max \
    --parameter 'x range 0 5 float'

Under the hood this is just ./omniopt --model=OPTUNA_TPE ...; the entry point adds sensible defaults for every --optuna_* flag and forwards the rest verbatim. See the Optuna integration tutorial for the full surface (pruners, persistent storage, multi-objective, remote control via omniopt_optuna study ...).

Using live-share

Use --live_share (also enablable via GUI) to automatically share the job. You will get a URL where your job data is hosted publically for 30 days, meaning everyone can access your results, and you can see all kinds of visualizations and export them.

Run Tests (Developer Use Only)

The test suite simulates various scenarios, including handling faulty jobs and ensuring program resilience.

./tests/main

See the automated tests tutorial page for more details.

Install from pypi

This may not use the bleeding-edge version, but if you get the version from here it means, the test suite has completely tested it properly.

pip3 install omniopt2

Install from repo (bleeding edge, may contain untested changes)

pip3 install -e git+https://github.com/NormanTUD/OmniOpt.git#egg=OmniOpt2

Alternatively, it can be executed directly, as OmniOpt2 will install its dependencies automatically if required.

Error Codes

For common issues and exit codes, see the exit codes tutorial-page.

Autocompletions

Autocomplete files for zsh and bash are in .shells. Run

bash .shells/install

to install them.

Contributions

I'd be glad to see your contributions!

Issues

If you experience any problems, please write issues at my Github Issues page.

Old OmniOpt

The old OmniOpt version, based on HyperOpt, is not supported anymore. It is still available, though, at https://github.com/NormanTUD/LegacyOmniOpt.

Metadata

Release files for omniopt2 9938

For a detailed explanation of source distributions (sdists) and built distributions (wheels), please see the package formats documentation.

Source distribution (sdist)

Source distribution for omniopt2 9938
File Size Uploaded Provenance Actions
omniopt2-9938.tar.gz 258.9 kB Without provenance Details for omniopt2-9938.tar.gz

Built distribution (wheel)

Table of built distributions (wheels) for omniopt2 9938
File Interpreter ABI Platform Actions
omniopt2-9938-py3-none-any.whl Python 3 none any Details for omniopt2-9938-py3-none-any.whl

Total release size: 830.4 kB

Release files / omniopt2-9938.tar.gz

Download URL omniopt2-9938.tar.gz
Size 258.9 kB
Tags Source
SHA-256 checksum
How to use checksums
dcb80595b5b3df68917ac8ad1d27614d7c42aba0f001c32c33a6a1de3e0cfed6
BLAKE2b-256 checksum
How to use checksums
fba9d3f8a098196235c1427bdf141dd08cd4b16f39df88f407b8459ae2798a21
Upload date
Uploaded using Trusted Publishing?
What is trusted publishing?
No
Uploaded via twine/6.2.0 CPython/3.9.25

Release files / omniopt2-9938-py3-none-any.whl

Download URL omniopt2-9938-py3-none-any.whl
Size 571.4 kB
Tags Python 3
SHA-256 checksum
How to use checksums
41a2d3f91426a442d0ba889d6943718eb1f958c1cdcbff6f7a44132d6fcdbc61
BLAKE2b-256 checksum
How to use checksums
b2e7d8e22927eec8feb462cb415f1acd987b187d8b6741d7afcccda3a15c5937
Upload date
Uploaded using Trusted Publishing?
What is trusted publishing?
No
Uploaded via twine/6.2.0 CPython/3.9.25

Security for omniopt2 9938

Provenance

Provenance describes where a file came from. On PyPI, provenance is shared via attestations, which provide a verifiable record of the build or publishing details. View details, limitations and caveats.

No provenance found

None of the files in this release were uploaded with provenance metadata.

A note on package security

PyPI is a repository for distributing packages, not a service for auditing them. While third-party security researchers and our staff do their best to detect and remove malicious packages, PyPI cannot guarantee the quality, security or suitability of any given package.

To learn more about protecting yourself from malicious actors, visit our package security documentation.

Release history Release notifications | RSS feed

This release

9938 This release

2 release files No provenance

Anthropic, PBC Visionary sponsor Bloomberg Visionary sponsor Hudson River Trading Visionary sponsor Meta Visionary sponsor NVIDIA Visionary sponsor Microsoft Sustainability sponsor Depot Continuous Integration AWS Cloud computing and Security Sponsor Datadog Monitoring Fastly CDN Google Download Analytics Sentry Error logging StatusPage Status page