{"id":13819319,"url":"https://github.com/heheda12345/MagPy","last_synced_at":"2025-05-16T04:33:08.658Z","repository":{"id":239908303,"uuid":"650218409","full_name":"heheda12345/MagPy","owner":"heheda12345","description":null,"archived":false,"fork":false,"pushed_at":"2024-06-05T08:15:33.000Z","size":1398,"stargazers_count":13,"open_issues_count":0,"forks_count":1,"subscribers_count":1,"default_branch":"master","last_synced_at":"2024-08-04T08:01:58.470Z","etag":null,"topics":[],"latest_commit_sha":null,"homepage":null,"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/heheda12345.png","metadata":{"files":{"readme":"README.md","changelog":null,"contributing":null,"funding":null,"license":"LICENSE","code_of_conduct":null,"threat_model":null,"audit":null,"citation":null,"codeowners":null,"security":null,"support":null,"governance":null,"roadmap":null,"authors":null,"dei":null,"publiccode":null,"codemeta":null}},"created_at":"2023-06-06T15:41:25.000Z","updated_at":"2024-08-04T08:01:59.468Z","dependencies_parsed_at":"2024-05-23T03:22:14.312Z","dependency_job_id":"cfbb3910-53cc-463b-9db9-2acea584925b","html_url":"https://github.com/heheda12345/MagPy","commit_stats":null,"previous_names":["heheda12345/frontend"],"tags_count":0,"template":false,"template_full_name":null,"repository_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/heheda12345%2FMagPy","tags_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/heheda12345%2FMagPy/tags","releases_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/heheda12345%2FMagPy/releases","manifests_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/heheda12345%2FMagPy/manifests","owner_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners/heheda12345","download_url":"https://codeload.github.com/heheda12345/MagPy/tar.gz/refs/heads/master","host":{"name":"GitHub","url":"https://github.com","kind":"github","repositories_count":225405583,"owners_count":17469372,"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":[],"created_at":"2024-08-04T08:00:45.012Z","updated_at":"2024-11-19T18:31:38.195Z","avatar_url":"https://github.com/heheda12345.png","language":"Python","funding_links":[],"categories":["Python","其他_机器学习与深度学习"],"sub_categories":[],"readme":"# MagPy\nMagPy is a JIT compiler for PyTorch programs. It can extract the operator graph from PyTorch programs and optimize the graph with a wide range of deep learning graph compilers.\n\n# Installation\nMagPy now supports Python 3.9. The support of other Python versions is working in progress.\n\n1. Install CUDA. CUDA 11.8 is recommended.\n2. Install dependencies:\n    ```bash\n    pip install -r requirements.txt -f https://download.pytorch.org/whl/torch_stable.html\n    ```\n3. Install MagPy:\n    ```bash\n    pip install -e .\n    ```\n4. Compile a shared library to disable Python integer cache by LD_PRELOAD. This script will generates a ``ldlong.v3.9.12.so'' file in build/ directory. You need to set the LD_PRELOAD environment variable to this file when running the PyTorch program.\n    ```bash\n    cd scripts\n    ./compile_longobj.sh\n    ```\n\n# Example Usage\n\nThe following script compiles and runs a simple PyTorch program with MagPy.\n\n```python\nLD_PRELOAD=build/ldlong.v3.9.12.so python test/example.py\n```\n\n# Citation\nIf you find MagPy useful in your research, please consider citing the following paper:\n\n\u003e MagPy: Effective Operator Graph Instantiation for Deep Learning by Execution State Monitoring; Chen Zhang, Rongchao Dong, Haojie Wang, Runxin Zhong, Jike Chen, and Jidong Zhai, Tsinghua University; will be appeared in USENIX ATC'24.\n\n","project_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fheheda12345%2FMagPy","html_url":"https://awesome.ecosyste.ms/projects/github.com%2Fheheda12345%2FMagPy","lists_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fheheda12345%2FMagPy/lists"}