Lossless tiling in convolution networks-graph metadata generation
A processing graph of an application with a sequence of processing nodes is obtained which processes an input and generates an intermediate representation a further intermediate representation, and an output representation of the input at stages in the sequence of processing nodes. Graph metadata is...
Gespeichert in:
| Hauptverfasser: | , , , , , , , , , |
|---|---|
| Format: | Patent |
| Sprache: | eng |
| Schlagworte: | |
| Online-Zugang: | Volltext bestellen |
| Tags: |
Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
|
| container_end_page | |
|---|---|
| container_issue | |
| container_start_page | |
| container_title | |
| container_volume | |
| creator | Chaphekar, Ruddhi Sivaramakrishnan, Ram Musaddiq, Matheen Prabhakar, Raghu Fuchs, Adi Wang, Junjue Sujeeth, Arvind Krishna Jairath, Sumti Nama, Tejas Nagendra Babu Liang, Kaizhao |
| description | A processing graph of an application with a sequence of processing nodes is obtained which processes an input and generates an intermediate representation a further intermediate representation, and an output representation of the input at stages in the sequence of processing nodes. Graph metadata is generated that specifies a non-overlapping target tiling configuration for the output representation, an overlapping tiling configuration for the input, an overlapping tiling configuration for the intermediate representation, and a third tiling configuration for the further intermediate representation. The processing graph is modified based on the graph metadata to conform to the parameters specified by the graph metadata. A set of computer instructions is then created to execute the modified processing graph on a target processing system. |
| format | Patent |
| fullrecord | <record><control><sourceid>epo_EVB</sourceid><recordid>TN_cdi_epo_espacenet_US12001936B2</recordid><sourceformat>XML</sourceformat><sourcesystem>PC</sourcesystem><sourcerecordid>US12001936B2</sourcerecordid><originalsourceid>FETCH-epo_espacenet_US12001936B23</originalsourceid><addsrcrecordid>eNrjZHD0yS8uzkktLlYoyczJzEtXyMxTSM7PK8vPKS3JzM9TyEstKc8vyi7WTS9KLMhQyE0tSUxJLElUSE_NSy1KBCnhYWBNS8wpTuWF0twMim6uIc4euqkF-fGpxQWJyUClJfGhwYZGBgaGlsZmTkbGxKgBAHDEMqc</addsrcrecordid><sourcetype>Open Access Repository</sourcetype><iscdi>true</iscdi><recordtype>patent</recordtype></control><display><type>patent</type><title>Lossless tiling in convolution networks-graph metadata generation</title><source>esp@cenet</source><creator>Chaphekar, Ruddhi ; Sivaramakrishnan, Ram ; Musaddiq, Matheen ; Prabhakar, Raghu ; Fuchs, Adi ; Wang, Junjue ; Sujeeth, Arvind Krishna ; Jairath, Sumti ; Nama, Tejas Nagendra Babu ; Liang, Kaizhao</creator><creatorcontrib>Chaphekar, Ruddhi ; Sivaramakrishnan, Ram ; Musaddiq, Matheen ; Prabhakar, Raghu ; Fuchs, Adi ; Wang, Junjue ; Sujeeth, Arvind Krishna ; Jairath, Sumti ; Nama, Tejas Nagendra Babu ; Liang, Kaizhao</creatorcontrib><description>A processing graph of an application with a sequence of processing nodes is obtained which processes an input and generates an intermediate representation a further intermediate representation, and an output representation of the input at stages in the sequence of processing nodes. Graph metadata is generated that specifies a non-overlapping target tiling configuration for the output representation, an overlapping tiling configuration for the input, an overlapping tiling configuration for the intermediate representation, and a third tiling configuration for the further intermediate representation. The processing graph is modified based on the graph metadata to conform to the parameters specified by the graph metadata. A set of computer instructions is then created to execute the modified processing graph on a target processing system.</description><language>eng</language><subject>CALCULATING ; COMPUTER SYSTEMS BASED ON SPECIFIC COMPUTATIONAL MODELS ; COMPUTING ; COUNTING ; PHYSICS</subject><creationdate>2024</creationdate><oa>free_for_read</oa><woscitedreferencessubscribed>false</woscitedreferencessubscribed></display><links><openurl>$$Topenurl_article</openurl><openurlfulltext>$$Topenurlfull_article</openurlfulltext><thumbnail>$$Tsyndetics_thumb_exl</thumbnail><linktohtml>$$Uhttps://worldwide.espacenet.com/publicationDetails/biblio?FT=D&date=20240604&DB=EPODOC&CC=US&NR=12001936B2$$EHTML$$P50$$Gepo$$Hfree_for_read</linktohtml><link.rule.ids>230,309,782,887,25573,76557</link.rule.ids><linktorsrc>$$Uhttps://worldwide.espacenet.com/publicationDetails/biblio?FT=D&date=20240604&DB=EPODOC&CC=US&NR=12001936B2$$EView_record_in_European_Patent_Office$$FView_record_in_$$GEuropean_Patent_Office$$Hfree_for_read</linktorsrc></links><search><creatorcontrib>Chaphekar, Ruddhi</creatorcontrib><creatorcontrib>Sivaramakrishnan, Ram</creatorcontrib><creatorcontrib>Musaddiq, Matheen</creatorcontrib><creatorcontrib>Prabhakar, Raghu</creatorcontrib><creatorcontrib>Fuchs, Adi</creatorcontrib><creatorcontrib>Wang, Junjue</creatorcontrib><creatorcontrib>Sujeeth, Arvind Krishna</creatorcontrib><creatorcontrib>Jairath, Sumti</creatorcontrib><creatorcontrib>Nama, Tejas Nagendra Babu</creatorcontrib><creatorcontrib>Liang, Kaizhao</creatorcontrib><title>Lossless tiling in convolution networks-graph metadata generation</title><description>A processing graph of an application with a sequence of processing nodes is obtained which processes an input and generates an intermediate representation a further intermediate representation, and an output representation of the input at stages in the sequence of processing nodes. Graph metadata is generated that specifies a non-overlapping target tiling configuration for the output representation, an overlapping tiling configuration for the input, an overlapping tiling configuration for the intermediate representation, and a third tiling configuration for the further intermediate representation. The processing graph is modified based on the graph metadata to conform to the parameters specified by the graph metadata. A set of computer instructions is then created to execute the modified processing graph on a target processing system.</description><subject>CALCULATING</subject><subject>COMPUTER SYSTEMS BASED ON SPECIFIC COMPUTATIONAL MODELS</subject><subject>COMPUTING</subject><subject>COUNTING</subject><subject>PHYSICS</subject><fulltext>true</fulltext><rsrctype>patent</rsrctype><creationdate>2024</creationdate><recordtype>patent</recordtype><sourceid>EVB</sourceid><recordid>eNrjZHD0yS8uzkktLlYoyczJzEtXyMxTSM7PK8vPKS3JzM9TyEstKc8vyi7WTS9KLMhQyE0tSUxJLElUSE_NSy1KBCnhYWBNS8wpTuWF0twMim6uIc4euqkF-fGpxQWJyUClJfGhwYZGBgaGlsZmTkbGxKgBAHDEMqc</recordid><startdate>20240604</startdate><enddate>20240604</enddate><creator>Chaphekar, Ruddhi</creator><creator>Sivaramakrishnan, Ram</creator><creator>Musaddiq, Matheen</creator><creator>Prabhakar, Raghu</creator><creator>Fuchs, Adi</creator><creator>Wang, Junjue</creator><creator>Sujeeth, Arvind Krishna</creator><creator>Jairath, Sumti</creator><creator>Nama, Tejas Nagendra Babu</creator><creator>Liang, Kaizhao</creator><scope>EVB</scope></search><sort><creationdate>20240604</creationdate><title>Lossless tiling in convolution networks-graph metadata generation</title><author>Chaphekar, Ruddhi ; Sivaramakrishnan, Ram ; Musaddiq, Matheen ; Prabhakar, Raghu ; Fuchs, Adi ; Wang, Junjue ; Sujeeth, Arvind Krishna ; Jairath, Sumti ; Nama, Tejas Nagendra Babu ; Liang, Kaizhao</author></sort><facets><frbrtype>5</frbrtype><frbrgroupid>cdi_FETCH-epo_espacenet_US12001936B23</frbrgroupid><rsrctype>patents</rsrctype><prefilter>patents</prefilter><language>eng</language><creationdate>2024</creationdate><topic>CALCULATING</topic><topic>COMPUTER SYSTEMS BASED ON SPECIFIC COMPUTATIONAL MODELS</topic><topic>COMPUTING</topic><topic>COUNTING</topic><topic>PHYSICS</topic><toplevel>online_resources</toplevel><creatorcontrib>Chaphekar, Ruddhi</creatorcontrib><creatorcontrib>Sivaramakrishnan, Ram</creatorcontrib><creatorcontrib>Musaddiq, Matheen</creatorcontrib><creatorcontrib>Prabhakar, Raghu</creatorcontrib><creatorcontrib>Fuchs, Adi</creatorcontrib><creatorcontrib>Wang, Junjue</creatorcontrib><creatorcontrib>Sujeeth, Arvind Krishna</creatorcontrib><creatorcontrib>Jairath, Sumti</creatorcontrib><creatorcontrib>Nama, Tejas Nagendra Babu</creatorcontrib><creatorcontrib>Liang, Kaizhao</creatorcontrib><collection>esp@cenet</collection></facets><delivery><delcategory>Remote Search Resource</delcategory><fulltext>fulltext_linktorsrc</fulltext></delivery><addata><au>Chaphekar, Ruddhi</au><au>Sivaramakrishnan, Ram</au><au>Musaddiq, Matheen</au><au>Prabhakar, Raghu</au><au>Fuchs, Adi</au><au>Wang, Junjue</au><au>Sujeeth, Arvind Krishna</au><au>Jairath, Sumti</au><au>Nama, Tejas Nagendra Babu</au><au>Liang, Kaizhao</au><format>patent</format><genre>patent</genre><ristype>GEN</ristype><title>Lossless tiling in convolution networks-graph metadata generation</title><date>2024-06-04</date><risdate>2024</risdate><abstract>A processing graph of an application with a sequence of processing nodes is obtained which processes an input and generates an intermediate representation a further intermediate representation, and an output representation of the input at stages in the sequence of processing nodes. Graph metadata is generated that specifies a non-overlapping target tiling configuration for the output representation, an overlapping tiling configuration for the input, an overlapping tiling configuration for the intermediate representation, and a third tiling configuration for the further intermediate representation. The processing graph is modified based on the graph metadata to conform to the parameters specified by the graph metadata. A set of computer instructions is then created to execute the modified processing graph on a target processing system.</abstract><oa>free_for_read</oa></addata></record> |
| fulltext | fulltext_linktorsrc |
| identifier | |
| ispartof | |
| issn | |
| language | eng |
| recordid | cdi_epo_espacenet_US12001936B2 |
| source | esp@cenet |
| subjects | CALCULATING COMPUTER SYSTEMS BASED ON SPECIFIC COMPUTATIONAL MODELS COMPUTING COUNTING PHYSICS |
| title | Lossless tiling in convolution networks-graph metadata generation |
| url | https://sfx.bib-bvb.de/sfx_tum?ctx_ver=Z39.88-2004&ctx_enc=info:ofi/enc:UTF-8&ctx_tim=2024-12-01T01%3A34%3A11IST&url_ver=Z39.88-2004&url_ctx_fmt=infofi/fmt:kev:mtx:ctx&rfr_id=info:sid/primo.exlibrisgroup.com:primo3-Article-epo_EVB&rft_val_fmt=info:ofi/fmt:kev:mtx:patent&rft.genre=patent&rft.au=Chaphekar,%20Ruddhi&rft.date=2024-06-04&rft_id=info:doi/&rft_dat=%3Cepo_EVB%3EUS12001936B2%3C/epo_EVB%3E%3Curl%3E%3C/url%3E&disable_directlink=true&sfx.directlink=off&sfx.report_link=0&rft_id=info:oai/&rft_id=info:pmid/&rfr_iscdi=true |