{"id":18668101,"url":"https://github.com/fredhohman/summit-notebooks","last_synced_at":"2025-04-12T00:21:53.271Z","repository":{"id":149161867,"uuid":"173875970","full_name":"fredhohman/summit-notebooks","owner":"fredhohman","description":"Notebooks for Scaling Deep Learning Interpretability by Visualizing Activation and Attribution Summarizations","archived":false,"fork":false,"pushed_at":"2019-10-03T01:59:10.000Z","size":1466,"stargazers_count":15,"open_issues_count":5,"forks_count":4,"subscribers_count":4,"default_branch":"master","last_synced_at":"2025-03-25T20:06:40.538Z","etag":null,"topics":["deep-learning","deep-learning-visualization","interactive-interface","interactive-visualization","interpretability"],"latest_commit_sha":null,"homepage":"","language":"Jupyter Notebook","has_issues":true,"has_wiki":null,"has_pages":null,"mirror_url":null,"source_name":null,"license":"mit","status":null,"scm":"git","pull_requests_enabled":true,"icon_url":"https://github.com/fredhohman.png","metadata":{"files":{"readme":"README.md","changelog":null,"contributing":null,"funding":null,"license":"LICENSE.md","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":"2019-03-05T04:53:53.000Z","updated_at":"2024-11-06T14:40:21.000Z","dependencies_parsed_at":null,"dependency_job_id":"2af8b612-3de2-46c6-89c2-a8437c76cc68","html_url":"https://github.com/fredhohman/summit-notebooks","commit_stats":null,"previous_names":[],"tags_count":0,"template":false,"template_full_name":null,"repository_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/fredhohman%2Fsummit-notebooks","tags_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/fredhohman%2Fsummit-notebooks/tags","releases_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/fredhohman%2Fsummit-notebooks/releases","manifests_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/fredhohman%2Fsummit-notebooks/manifests","owner_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners/fredhohman","download_url":"https://codeload.github.com/fredhohman/summit-notebooks/tar.gz/refs/heads/master","host":{"name":"GitHub","url":"https://github.com","kind":"github","repositories_count":248498098,"owners_count":21114038,"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":["deep-learning","deep-learning-visualization","interactive-interface","interactive-visualization","interpretability"],"created_at":"2024-11-07T08:41:19.956Z","updated_at":"2025-04-12T00:21:53.260Z","avatar_url":"https://github.com/fredhohman.png","language":"Jupyter Notebook","funding_links":[],"categories":[],"sub_categories":[],"readme":"# Summit Notebooks\n\nSummit is an interactive system that scalably and systematically summarizes and visualizes what features a deep learning model has learned and how those features interact to make predictions.\nThis repository contains the python notebooks used to generate the data used in the [Summit visualization][summit].\n\nFor the main Summit repo, go to [https://github.com/fredhohman/summit][summit].\n\n\n### Main notebooks:\n\n* [`activation-matrices.ipynb`](activation-matrices.ipynb): generate aggregated activation matrices\n* [`influence.py`](activation-matrices.ipynb): generate aggregated influence matrices\n* [`activation-matrices-to-json.ipynb`](activation-matrices-to-json.ipynb): combine activation matrices per class into json format\n* [`attribution-graph.ipynb`](dag.ipynb): generating class attribution graphs\n* [`feature-vis-sprite-to-images.ipynb`](feature-vis-sprite-to-images.ipynb): split feature visualization sprites to single images\n\n### Experimental notebooks:\n\n* [`top-channels-used-per-layer.ipynb`](top-channels-used-per-layer.ipynb): analysis for determining which channels were used the most by all classes for all layers\n\n\n## Live Demo\n\nFor a live demo, visit: [fredhohman.com/summit][demo]\n\n\n## Resources\n\nWe used the following ImageNet metadata:\n\n* [https://github.com/google/inception/blob/master/synsets.txt](https://github.com/google/inception/blob/master/synsets.txt)\n* [https://gist.github.com/aaronpolhamus/964a4411c0906315deb9f4a3723aac57](https://gist.github.com/aaronpolhamus/964a4411c0906315deb9f4a3723aac57)\n* [https://gist.github.com/yrevar/942d3a0ac09ec9e5eb3a](https://gist.github.com/yrevar/942d3a0ac09ec9e5eb3a)\n\n\n## License\n\nMIT License. See [`LICENSE.md`](LICENSE.md).\n\n\n## Citation\n\n```\n@article{hohman2020summit,\n  title={Summit: Scaling Deep Learning Interpretability by Visualizing Activation and Attribution Summarizations},\n  author={Hohman, Fred and Park, Haekyu and Robinson, Caleb and Chau, Duen Horng},\n  journal={IEEE Transactions on Visualization and Computer Graphics (TVCG)},\n  year={2020},\n  publisher={IEEE},\n  url={https://fredhohman.com/summit/}\n}\n```\n\n\n## Contact\n\nFor questions or support [open an issue][issues] or contact [Fred Hohman][fred].\n\n[summit]: https://github.com/fredhohman/summit\n[fred]: https://fredhohman.com\n[demo]: https://fredhohman.com/summit/\n[issues]: https://github.com/fredhohman/summit-notebooks/issues\n","project_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Ffredhohman%2Fsummit-notebooks","html_url":"https://awesome.ecosyste.ms/projects/github.com%2Ffredhohman%2Fsummit-notebooks","lists_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Ffredhohman%2Fsummit-notebooks/lists"}