FRINK Linked Data Fragments server

Matches in SemOpenAlex for { <https://semopenalex.org/work/W3201822691> ?p ?o ?g. }

• W3201822691 abstract "The influential Residual Networks designed by He et al. remain the gold-standard architecture in numerous scientific publications. They typically serve as the default architecture in studies, or as baselines when new architectures are proposed. Yet there has been significant progress on best practices for training neural networks since the inception of the ResNet architecture in 2015. Novel optimization & data-augmentation have increased the effectiveness of the training recipes. In this paper, we re-evaluate the performance of the vanilla ResNet-50 when trained with a procedure that integrates such advances. We share competitive training settings and pre-trained models in the timm open-source library, with the hope that they will serve as better baselines for future work. For instance, with our more demanding training setting, a vanilla ResNet-50 reaches 80.4% top-1 accuracy at resolution 224x224 on ImageNet-val without extra data or distillation. We also report the performance achieved with popular models with our training procedure." @default.
• W3201822691 created "2021-10-11" @default.
• W3201822691 creator A5032644449 @default.
• W3201822691 creator A5033867133 @default.
• W3201822691 creator A5048883432 @default.
• W3201822691 date "2021-10-01" @default.
• W3201822691 modified "2023-09-27" @default.
• W3201822691 title "ResNet strikes back: An improved training procedure in timm" @default.
• W3201822691 cites W104184427 @default.
• W3201822691 cites W2086161653 @default.
• W3201822691 cites W2095705004 @default.
• W3201822691 cites W2097117768 @default.
• W3201822691 cites W2108598243 @default.
• W3201822691 cites W2117539524 @default.
• W3201822691 cites W2138011018 @default.
• W3201822691 cites W2163605009 @default.
• W3201822691 cites W2183341477 @default.
• W3201822691 cites W2194775991 @default.
• W3201822691 cites W2331143823 @default.
• W3201822691 cites W2533598788 @default.
• W3201822691 cites W2549139847 @default.
• W3201822691 cites W2612445135 @default.
• W3201822691 cites W2752782242 @default.
• W3201822691 cites W2765407302 @default.
• W3201822691 cites W2768282280 @default.
• W3201822691 cites W2785430118 @default.
• W3201822691 cites W2792287754 @default.
• W3201822691 cites W2804047946 @default.
• W3201822691 cites W2911925209 @default.
• W3201822691 cites W2943152387 @default.
• W3201822691 cites W2944223741 @default.
• W3201822691 cites W2946948417 @default.
• W3201822691 cites W2950181225 @default.
• W3201822691 cites W2950541952 @default.
• W3201822691 cites W2962900737 @default.
• W3201822691 cites W2963855133 @default.
• W3201822691 cites W2971315489 @default.
• W3201822691 cites W2994088087 @default.
• W3201822691 cites W2995435108 @default.
• W3201822691 cites W2998508940 @default.
• W3201822691 cites W2999512283 @default.
• W3201822691 cites W3015045227 @default.
• W3201822691 cites W3016719260 @default.
• W3201822691 cites W3018378048 @default.
• W3201822691 cites W3034363135 @default.
• W3201822691 cites W3034429256 @default.
• W3201822691 cites W3035452548 @default.
• W3201822691 cites W3091401866 @default.
• W3201822691 cites W3092043977 @default.
• W3201822691 cites W3102631365 @default.
• W3201822691 cites W3118608800 @default.
• W3201822691 cites W3119786062 @default.
• W3201822691 cites W3121523901 @default.
• W3201822691 cites W3128633047 @default.
• W3201822691 cites W3135921327 @default.
• W3201822691 cites W3138994021 @default.
• W3201822691 cites W3157506437 @default.
• W3201822691 cites W3159481202 @default.
• W3201822691 cites W3163465952 @default.
• W3201822691 cites W3170874841 @default.
• W3201822691 cites W3172064272 @default.
• W3201822691 cites W3174068320 @default.
• W3201822691 cites W3199892937 @default.
• W3201822691 cites W3202088435 @default.
• W3201822691 cites W3202947737 @default.
• W3201822691 hasPublicationYear "2021" @default.
• W3201822691 type Work @default.
• W3201822691 sameAs 3201822691 @default.
• W3201822691 citedByCount "7" @default.
• W3201822691 countsByYear W32018226912021 @default.
• W3201822691 crossrefType "posted-content" @default.
• W3201822691 hasAuthorship W3201822691A5032644449 @default.
• W3201822691 hasAuthorship W3201822691A5033867133 @default.
• W3201822691 hasAuthorship W3201822691A5048883432 @default.
• W3201822691 hasConcept C11413529 @default.
• W3201822691 hasConcept C119857082 @default.
• W3201822691 hasConcept C123657996 @default.
• W3201822691 hasConcept C153294291 @default.
• W3201822691 hasConcept C154945302 @default.
• W3201822691 hasConcept C155512373 @default.
• W3201822691 hasConcept C166957645 @default.
• W3201822691 hasConcept C193415008 @default.
• W3201822691 hasConcept C205649164 @default.
• W3201822691 hasConcept C2777211547 @default.
• W3201822691 hasConcept C2944601119 @default.
• W3201822691 hasConcept C38652104 @default.
• W3201822691 hasConcept C41008148 @default.
• W3201822691 hasConcept C50644808 @default.
• W3201822691 hasConcept C51632099 @default.
• W3201822691 hasConceptScore W3201822691C11413529 @default.
• W3201822691 hasConceptScore W3201822691C119857082 @default.
• W3201822691 hasConceptScore W3201822691C123657996 @default.
• W3201822691 hasConceptScore W3201822691C153294291 @default.
• W3201822691 hasConceptScore W3201822691C154945302 @default.
• W3201822691 hasConceptScore W3201822691C155512373 @default.
• W3201822691 hasConceptScore W3201822691C166957645 @default.
• W3201822691 hasConceptScore W3201822691C193415008 @default.
• W3201822691 hasConceptScore W3201822691C205649164 @default.
• W3201822691 hasConceptScore W3201822691C2777211547 @default.
• W3201822691 hasConceptScore W3201822691C2944601119 @default.

• next