http://rdf.ncbi.nlm.nih.gov/pubchem/patent/CN-112001455-A
Outgoing Links
Predicate | Object |
---|---|
assignee | http://rdf.ncbi.nlm.nih.gov/pubchem/patentassignee/MD5_697991755f183a3087ac2d9a2e8d03f9 |
classificationCPCInventive | http://rdf.ncbi.nlm.nih.gov/pubchem/patentcpc/G06F18-214 http://rdf.ncbi.nlm.nih.gov/pubchem/patentcpc/G06N20-20 http://rdf.ncbi.nlm.nih.gov/pubchem/patentcpc/G06N3-084 |
classificationIPCInventive | http://rdf.ncbi.nlm.nih.gov/pubchem/patentipc/G06N20-20 http://rdf.ncbi.nlm.nih.gov/pubchem/patentipc/G06N3-08 http://rdf.ncbi.nlm.nih.gov/pubchem/patentipc/G06K9-62 |
filingDate | 2020-09-29-04:00^^<http://www.w3.org/2001/XMLSchema#date> |
inventor | http://rdf.ncbi.nlm.nih.gov/pubchem/patentinventor/MD5_4678bf69627fa16be2e38e017d0dc75c http://rdf.ncbi.nlm.nih.gov/pubchem/patentinventor/MD5_53401adfc33f56aa9f904337f26a3e37 http://rdf.ncbi.nlm.nih.gov/pubchem/patentinventor/MD5_bfbbda1ea581dad07669203e9cbeb68a http://rdf.ncbi.nlm.nih.gov/pubchem/patentinventor/MD5_aa50282abe167c755f9bb33e50ef96f7 http://rdf.ncbi.nlm.nih.gov/pubchem/patentinventor/MD5_739e2848a3734aae161ec5d4bd410781 http://rdf.ncbi.nlm.nih.gov/pubchem/patentinventor/MD5_be3279b2d8d0c8c28a5f7a883ac57d02 http://rdf.ncbi.nlm.nih.gov/pubchem/patentinventor/MD5_c514c0af73c6a945920202df8ea00aaa |
publicationDate | 2020-11-27-04:00^^<http://www.w3.org/2001/XMLSchema#date> |
publicationNumber | CN-112001455-A |
titleOfInvention | Model training method, device and electronic device |
abstract | The present application discloses a model training method, device and electronic device, and relates to the technical field of deep learning. The specific implementation scheme is: the first electronic device sends the number of gradients to the second electronic device; receives the first fusion gradient sent by the second electronic device based on the number of gradients; according to the first fusion gradient and the second fusion gradient, Obtain a target fusion gradient, where the second fusion gradient is obtained by fusion according to N gradients taken out from the gradient queue corresponding to the first processor, where N is the number of gradients; send the second electronic device to the second electronic device the target fusion gradient; updating the parameters of the learning model of the first electronic device according to the target fusion gradient. Due to the first fusion gradient sent by the second electronic device, gradient fusion is performed on the N gradients, which can reduce the number of communications between the first electronic device and the second electronic device and improve the training efficiency of the learning model. |
priorityDate | 2020-09-29-04:00^^<http://www.w3.org/2001/XMLSchema#date> |
type | http://data.epo.org/linked-data/def/patent/Publication |
Incoming Links
Predicate | Subject |
---|---|
isDiscussedBy | http://rdf.ncbi.nlm.nih.gov/pubchem/compound/CID22978774 http://rdf.ncbi.nlm.nih.gov/pubchem/substance/SID419701332 |
Total number of triples: 23.