Skip to main content

gpuastar-experiments

gpuastar-experiments contains the domain and experiment code that is kept out of the base gpuastar package:

  • Example domains in src/gpuastar_experiments/environments/: Rubik's Cube (cube3), sliding tile puzzles (puzzle15, puzzle24, puzzle35, puzzle48), Lights Out (lightsout7), and Sokoban (sokoban), together with their tensor-based adapters.
  • Test datasets and trained model checkpoints in src/gpuastar_experiments/data/ and src/gpuastar_experiments/saved_models/.
  • Hydra default overlays in src/gpuastar_experiments/configs/hydra/command_defaults/, made visible to the gpuastar CLI by the search-path plugin in src/hydra_plugins/gpuastar_experiments_searchpath.py. With this package installed, command=astar_search and command=aoti_export receive Rubik's Cube example defaults for the environment, dataset, model, and results paths.
  • The experiment suite in src/gpuastar_experiments/experiments/, driven by the gpuastar-experiment console script and the presets in experiments/conf/experiment/. The shell wrappers in scripts/ call the presets.
  • Benchmark runners: the standalone CUDA solver examples in src/gpuastar_experiments/program/.
  • Compiled example domains for the C++ solver in cpp/, packaged separately as gpuastar-experiments-cpp.

Installation

python -m pip install "gpuastar[experiments]"

The CUDA examples also require the CUDA solver:

python -m pip install "gpuastar[cu,experiments]"

Running from a source checkout

The Pixi environments defined at the repository root include this package. Two examples:

CUDA_VISIBLE_DEVICES=1 pixi run -e cu-experiments cu-example-program
CUDA_VISIBLE_DEVICES=1 pixi run -e cu-experiments cu-cube3-smoke

To run an experiment preset:

pixi run -e dev gpuastar-experiment experiment=quick_validation

The available presets and the output layout are documented in the repository README and in docs/benchmarks.md.

Metadata

Release files for gpuastar-experiments 0.1.0

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

Source distribution (sdist)

Source distribution for gpuastar-experiments 0.1.0
File Size Uploaded
gpuastar_experiments-0.1.0.tar.gz 55.2 MB Details

Built distribution (wheel)

Table of built distributions (wheels) for gpuastar-experiments 0.1.0
File Interpreter ABI Platform
gpuastar_experiments-0.1.0-py3-none-any.whl Python 3 none any Details

Total release size: 110.4 MB

Release files / gpuastar_experiments-0.1.0.tar.gz

Download URL gpuastar_experiments-0.1.0.tar.gz
Size 55.2 MB
Tags Source
SHA-256 checksum
How to use checksums
c0c2b61e9359fd153e769a71134cbe8c2ed33d57bf110b96f3aa499589292716
BLAKE2b-256 checksum
How to use checksums
8e2a30742fcd237296b4053e4fad64bd334cec3e03f5c7f6f6fd916e9d09ac84
Upload date
Uploaded using Trusted Publishing?
What is trusted publishing?
Yes
Uploaded via twine/6.1.0 CPython/3.13.13

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.

PyPI Publish Attestation

PyPI verified that this artifact, at this checksum, originated from the publisher listed below.

Signed by GitHub Actions, verified by PyPI on Aug 5, 2026.

Transparency log

Release files / gpuastar_experiments-0.1.0-py3-none-any.whl

Download URL gpuastar_experiments-0.1.0-py3-none-any.whl
Size 55.2 MB
Tags Python 3
SHA-256 checksum
How to use checksums
3d598649f1f3e703e98e03f35cbbbfc0a5321755f24258523c959b18a9af8c08
BLAKE2b-256 checksum
How to use checksums
5950c295c68ac79ee638e198c09aa4db499084bbd55b16aabd3f6e45dd5d3b52
Upload date
Uploaded using Trusted Publishing?
What is trusted publishing?
Yes
Uploaded via twine/6.1.0 CPython/3.13.13

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.

PyPI Publish Attestation

PyPI verified that this artifact, at this checksum, originated from the publisher listed below.

Signed by GitHub Actions, verified by PyPI on Aug 5, 2026.

Transparency log

Release history Release notifications | RSS feed

This release

0.1.0 This release

2 release files

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