{"id":18129139,"url":"https://github.com/maximedebarbat/dolphin","last_synced_at":"2025-07-07T06:38:46.321Z","repository":{"id":172218551,"uuid":"594360051","full_name":"MaximeDebarbat/Dolphin","owner":"MaximeDebarbat","description":"Dolphin is a python toolkit meant to speed up inference of TensorRT by providing CUDA-Accelerated processing.","archived":false,"fork":false,"pushed_at":"2023-07-20T13:28:41.000Z","size":602,"stargazers_count":3,"open_issues_count":1,"forks_count":0,"subscribers_count":1,"default_branch":"main","last_synced_at":"2025-06-22T09:47:24.291Z","etag":null,"topics":["cuda","python","tensorrt-inference"],"latest_commit_sha":null,"homepage":"https://dolphin-python.readthedocs.io/","language":"Python","has_issues":true,"has_wiki":null,"has_pages":null,"mirror_url":null,"source_name":null,"license":"apache-2.0","status":null,"scm":"git","pull_requests_enabled":true,"icon_url":"https://github.com/MaximeDebarbat.png","metadata":{"files":{"readme":"README.md","changelog":null,"contributing":null,"funding":null,"license":"LICENSE","code_of_conduct":"CODE_OF_CONDUCT.md","threat_model":null,"audit":null,"citation":null,"codeowners":null,"security":null,"support":null,"governance":null,"roadmap":null,"authors":null}},"created_at":"2023-01-28T09:55:49.000Z","updated_at":"2024-02-04T22:19:49.000Z","dependencies_parsed_at":"2023-12-24T00:16:08.251Z","dependency_job_id":"fbbfce42-221f-4f1f-b374-0a7efcc71a53","html_url":"https://github.com/MaximeDebarbat/Dolphin","commit_stats":null,"previous_names":["maximedebarbat/dolphin"],"tags_count":0,"template":false,"template_full_name":null,"purl":"pkg:github/MaximeDebarbat/Dolphin","repository_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/MaximeDebarbat%2FDolphin","tags_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/MaximeDebarbat%2FDolphin/tags","releases_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/MaximeDebarbat%2FDolphin/releases","manifests_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/MaximeDebarbat%2FDolphin/manifests","owner_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners/MaximeDebarbat","download_url":"https://codeload.github.com/MaximeDebarbat/Dolphin/tar.gz/refs/heads/main","sbom_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/MaximeDebarbat%2FDolphin/sbom","host":{"name":"GitHub","url":"https://github.com","kind":"github","repositories_count":264029705,"owners_count":23546520,"icon_url":"https://github.com/github.png","version":null,"created_at":"2022-05-30T11:31:42.601Z","updated_at":"2022-07-04T15:15:14.044Z","host_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub","repositories_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories","repository_names_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repository_names","owners_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners"}},"keywords":["cuda","python","tensorrt-inference"],"created_at":"2024-11-01T10:06:11.809Z","updated_at":"2025-07-07T06:38:46.292Z","avatar_url":"https://github.com/MaximeDebarbat.png","language":"Python","funding_links":[],"categories":[],"sub_categories":[],"readme":"# Dolphin\n\n![Banner](misc/banner.png)\n\nGeneral python package for CUDA accelerated deep learning inference.\n\n- **Documentation** : [ReadTheDoc](https://dolphin-python.readthedocs.io/en/latest/index.html)\n- **Source code** : [https://github.com/MaximeDebarbat/Dolphin](https://github.com/MaximeDebarbat/Dolphin)\n- **Bug reports** : [https://github.com/MaximeDebarbat/Dolphin/issues](https://github.com/MaximeDebarbat/Dolphin/issues)\n- **Getting Starterd** : \u003ca href=\"https://colab.research.google.com/drive/1RTZI9hJ6a33NtVUYM0esvSg8nzh2MlP2?usp=sharing\"\u003e\u003cimg src=\"https://colab.research.google.com/assets/colab-badge.svg\" alt=\"Open In Colab\"\u003e\u003c/a\u003e\n\nIt provides :\n\n- A set of common image processing functions\n- A TensorRT wrapper for easy inference\n- Speeds up the inference with CUDA and TensorRT\n- An easy to use API with Numpy\n- A fast N-Dimensional array\n\nTesting :\n\nIn order to test the package, you will need the library `pytest` which you can run from the root of the project :\n```\npytest\n```\n\n## Install\n\n```\npip install dolphin-python\n```\n\n## Build\n\nDolphin can be installed with Pypi (coming soon) or built with Docker which is the recommended way to use it :\n\n```\ndocker build  -f Dockerfile \\\n              --rm \\\n              -t dolphin:latest \\\n              .\n```\n\n## Docker run\n\nEnsure that you have the `nvidia-docker` package installed and run the following command :\n\n```\ndocker run \\\n        -it \\\n        --rm \\\n        --gpus all \\\n        -v \"$( cd \"$( dirname \"${BASH_SOURCE[0]}\" )\" \u003e/dev/null 2\u003e\u00261 \u0026\u0026 pwd )\":\"/app\" \\\n        dolphin:latest \\\n        bash\n```\n\nPlease note that Dolphin might not work without the `--gpus all` flag or `--runtime nvidia`.\n\n## Acknowledgements\n\nThis project could not have been possible without [PyCuda](https://github.com/inducer/pycuda):\n\n\u003e Andreas Klöckner, Nicolas Pinto, Yunsup Lee, Bryan Catanzaro, Paul Ivanov, Ahmed Fasih, PyCUDA and PyOpenCL: A scripting-based approach to GPU run-time code generation, \u003e Parallel Computing, Volume 38, Issue 3, March 2012, Pages 157-174.\n\n## TODOs\n\n- [ ] Improve `Engine` class in order to support *int8*\n- [ ] Use Cython to speed up the code","project_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fmaximedebarbat%2Fdolphin","html_url":"https://awesome.ecosyste.ms/projects/github.com%2Fmaximedebarbat%2Fdolphin","lists_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fmaximedebarbat%2Fdolphin/lists"}