{"about":{"site":"https://codewithpapers.app","non_affiliation":"Code with Papers and Syntology are not affiliated with, endorsed by, or sponsored by Papers with Code, Meta, or the pwc-archive mirror.","licence":"CC BY-SA 4.0","licence_url":"https://creativecommons.org/licenses/by-sa/4.0/legalcode","attribution":"https://codewithpapers.app/attribution","modified":"archive material modified by Syntology; see the attribution page"},"url":"/paper/purine-a-bi-graph-based-deep-learning","title":"Purine: A bi-graph based deep learning framework","arxiv_id":"1412.6249","date":"2014-12-19","proceeding":null,"authors":["Min Lin","Shuo Li","Xuan Luo","Shuicheng Yan"],"abstract":"In this paper, we introduce a novel deep learning framework, termed Purine.\nIn Purine, a deep network is expressed as a bipartite graph (bi-graph), which\nis composed of interconnected operators and data tensors. With the bi-graph\nabstraction, networks are easily solvable with event-driven task dispatcher. We\nthen demonstrate that different parallelism schemes over GPUs and/or CPUs on\nsingle or multiple PCs can be universally implemented by graph composition.\nThis eases researchers from coding for various parallelization schemes, and the\nsame dispatcher can be used for solving variant graphs. Scheduled by the task\ndispatcher, memory transfers are fully overlapped with other computations,\nwhich greatly reduce the communication overhead and help us achieve approximate\nlinear acceleration.","url_abs":"http://arxiv.org/abs/1412.6249v5","url_pdf":"http://arxiv.org/pdf/1412.6249v5.pdf","source":{"archive":"pwc-archive (Hugging Face), CC BY-SA 4.0","snapshot":"2025-07-28","licence_url":"https://creativecommons.org/licenses/by-sa/4.0/legalcode","row_kind":"abstracts"},"code_links":[{"paper_slug":"purine-a-bi-graph-based-deep-learning","repo_url":"https://github.com/purine/purine2","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"none","reach":null}],"tasks":[{"task_slug":"deep-learning","task_name":"Deep Learning"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"atlas_url":null,"mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}