Task processing method and device, electronic equipment and storage medium
The invention provides a task processing method, and relates to the technical field of artificial intelligence, in particular to the technical fields of distributed training, large model, pre-training and the like. According to the specific implementation scheme, in response to the fact that it is d...
Gespeichert in:
Hauptverfasser: | , , , , |
---|---|
Format: | Patent |
Sprache: | chi ; 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 | SHEN LIANG GONG WEIBAO YU DIANHAI ZENG JINLE HAO HONGXIANG |
description | The invention provides a task processing method, and relates to the technical field of artificial intelligence, in particular to the technical fields of distributed training, large model, pre-training and the like. According to the specific implementation scheme, in response to the fact that it is determined that an ith forward calculation task is executed and reverse input data of a second processing layer is received, a target calculation unit executes a reverse calculation task according to the reverse input data and a forward calculation result corresponding to the jth batch of reverse input data to obtain a reverse calculation result of the second processing layer, the target calculation unit is any calculation unit in the plurality of calculation units; and in response to determining that the execution of the K reverse calculation tasks of the second processing layer related to the current batch of training data is completed, the target calculation unit sends K batch reverse calculation results of the s |
format | Patent |
fullrecord | <record><control><sourceid>epo_EVB</sourceid><recordid>TN_cdi_epo_espacenet_CN116934571A</recordid><sourceformat>XML</sourceformat><sourcesystem>PC</sourcesystem><sourcerecordid>CN116934571A</sourcerecordid><originalsourceid>FETCH-epo_espacenet_CN116934571A3</originalsourceid><addsrcrecordid>eNrjZPAKSSzOVigoyk9OLS7OzEtXyE0tychPUUjMS1FISS3LTE7VUUjNSU0uKcrPy0xWSC0szSzITc0rASsoLskvSkxPBepJySzN5WFgTUvMKU7lhdLcDIpuriHOHrqpBfnxqcUFicmpeakl8c5-hoZmlsYmpuaGjsbEqAEAJEk1JA</addsrcrecordid><sourcetype>Open Access Repository</sourcetype><iscdi>true</iscdi><recordtype>patent</recordtype></control><display><type>patent</type><title>Task processing method and device, electronic equipment and storage medium</title><source>esp@cenet</source><creator>SHEN LIANG ; GONG WEIBAO ; YU DIANHAI ; ZENG JINLE ; HAO HONGXIANG</creator><creatorcontrib>SHEN LIANG ; GONG WEIBAO ; YU DIANHAI ; ZENG JINLE ; HAO HONGXIANG</creatorcontrib><description>The invention provides a task processing method, and relates to the technical field of artificial intelligence, in particular to the technical fields of distributed training, large model, pre-training and the like. According to the specific implementation scheme, in response to the fact that it is determined that an ith forward calculation task is executed and reverse input data of a second processing layer is received, a target calculation unit executes a reverse calculation task according to the reverse input data and a forward calculation result corresponding to the jth batch of reverse input data to obtain a reverse calculation result of the second processing layer, the target calculation unit is any calculation unit in the plurality of calculation units; and in response to determining that the execution of the K reverse calculation tasks of the second processing layer related to the current batch of training data is completed, the target calculation unit sends K batch reverse calculation results of the s</description><language>chi ; eng</language><subject>CALCULATING ; COMPUTER SYSTEMS BASED ON SPECIFIC COMPUTATIONAL MODELS ; COMPUTING ; COUNTING ; IMAGE DATA PROCESSING OR GENERATION, IN GENERAL ; PHYSICS</subject><creationdate>2023</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=20231024&DB=EPODOC&CC=CN&NR=116934571A$$EHTML$$P50$$Gepo$$Hfree_for_read</linktohtml><link.rule.ids>230,309,781,886,25569,76552</link.rule.ids><linktorsrc>$$Uhttps://worldwide.espacenet.com/publicationDetails/biblio?FT=D&date=20231024&DB=EPODOC&CC=CN&NR=116934571A$$EView_record_in_European_Patent_Office$$FView_record_in_$$GEuropean_Patent_Office$$Hfree_for_read</linktorsrc></links><search><creatorcontrib>SHEN LIANG</creatorcontrib><creatorcontrib>GONG WEIBAO</creatorcontrib><creatorcontrib>YU DIANHAI</creatorcontrib><creatorcontrib>ZENG JINLE</creatorcontrib><creatorcontrib>HAO HONGXIANG</creatorcontrib><title>Task processing method and device, electronic equipment and storage medium</title><description>The invention provides a task processing method, and relates to the technical field of artificial intelligence, in particular to the technical fields of distributed training, large model, pre-training and the like. According to the specific implementation scheme, in response to the fact that it is determined that an ith forward calculation task is executed and reverse input data of a second processing layer is received, a target calculation unit executes a reverse calculation task according to the reverse input data and a forward calculation result corresponding to the jth batch of reverse input data to obtain a reverse calculation result of the second processing layer, the target calculation unit is any calculation unit in the plurality of calculation units; and in response to determining that the execution of the K reverse calculation tasks of the second processing layer related to the current batch of training data is completed, the target calculation unit sends K batch reverse calculation results of the s</description><subject>CALCULATING</subject><subject>COMPUTER SYSTEMS BASED ON SPECIFIC COMPUTATIONAL MODELS</subject><subject>COMPUTING</subject><subject>COUNTING</subject><subject>IMAGE DATA PROCESSING OR GENERATION, IN GENERAL</subject><subject>PHYSICS</subject><fulltext>true</fulltext><rsrctype>patent</rsrctype><creationdate>2023</creationdate><recordtype>patent</recordtype><sourceid>EVB</sourceid><recordid>eNrjZPAKSSzOVigoyk9OLS7OzEtXyE0tychPUUjMS1FISS3LTE7VUUjNSU0uKcrPy0xWSC0szSzITc0rASsoLskvSkxPBepJySzN5WFgTUvMKU7lhdLcDIpuriHOHrqpBfnxqcUFicmpeakl8c5-hoZmlsYmpuaGjsbEqAEAJEk1JA</recordid><startdate>20231024</startdate><enddate>20231024</enddate><creator>SHEN LIANG</creator><creator>GONG WEIBAO</creator><creator>YU DIANHAI</creator><creator>ZENG JINLE</creator><creator>HAO HONGXIANG</creator><scope>EVB</scope></search><sort><creationdate>20231024</creationdate><title>Task processing method and device, electronic equipment and storage medium</title><author>SHEN LIANG ; GONG WEIBAO ; YU DIANHAI ; ZENG JINLE ; HAO HONGXIANG</author></sort><facets><frbrtype>5</frbrtype><frbrgroupid>cdi_FETCH-epo_espacenet_CN116934571A3</frbrgroupid><rsrctype>patents</rsrctype><prefilter>patents</prefilter><language>chi ; eng</language><creationdate>2023</creationdate><topic>CALCULATING</topic><topic>COMPUTER SYSTEMS BASED ON SPECIFIC COMPUTATIONAL MODELS</topic><topic>COMPUTING</topic><topic>COUNTING</topic><topic>IMAGE DATA PROCESSING OR GENERATION, IN GENERAL</topic><topic>PHYSICS</topic><toplevel>online_resources</toplevel><creatorcontrib>SHEN LIANG</creatorcontrib><creatorcontrib>GONG WEIBAO</creatorcontrib><creatorcontrib>YU DIANHAI</creatorcontrib><creatorcontrib>ZENG JINLE</creatorcontrib><creatorcontrib>HAO HONGXIANG</creatorcontrib><collection>esp@cenet</collection></facets><delivery><delcategory>Remote Search Resource</delcategory><fulltext>fulltext_linktorsrc</fulltext></delivery><addata><au>SHEN LIANG</au><au>GONG WEIBAO</au><au>YU DIANHAI</au><au>ZENG JINLE</au><au>HAO HONGXIANG</au><format>patent</format><genre>patent</genre><ristype>GEN</ristype><title>Task processing method and device, electronic equipment and storage medium</title><date>2023-10-24</date><risdate>2023</risdate><abstract>The invention provides a task processing method, and relates to the technical field of artificial intelligence, in particular to the technical fields of distributed training, large model, pre-training and the like. According to the specific implementation scheme, in response to the fact that it is determined that an ith forward calculation task is executed and reverse input data of a second processing layer is received, a target calculation unit executes a reverse calculation task according to the reverse input data and a forward calculation result corresponding to the jth batch of reverse input data to obtain a reverse calculation result of the second processing layer, the target calculation unit is any calculation unit in the plurality of calculation units; and in response to determining that the execution of the K reverse calculation tasks of the second processing layer related to the current batch of training data is completed, the target calculation unit sends K batch reverse calculation results of the s</abstract><oa>free_for_read</oa></addata></record> |
fulltext | fulltext_linktorsrc |
identifier | |
ispartof | |
issn | |
language | chi ; eng |
recordid | cdi_epo_espacenet_CN116934571A |
source | esp@cenet |
subjects | CALCULATING COMPUTER SYSTEMS BASED ON SPECIFIC COMPUTATIONAL MODELS COMPUTING COUNTING IMAGE DATA PROCESSING OR GENERATION, IN GENERAL PHYSICS |
title | Task processing method and device, electronic equipment and storage medium |
url | https://sfx.bib-bvb.de/sfx_tum?ctx_ver=Z39.88-2004&ctx_enc=info:ofi/enc:UTF-8&ctx_tim=2024-12-11T20%3A25%3A02IST&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=SHEN%20LIANG&rft.date=2023-10-24&rft_id=info:doi/&rft_dat=%3Cepo_EVB%3ECN116934571A%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 |