HyperAIHyperAI
الرئيسيةالأخبارأحدث الأوراق البحثيةالدروسمجموعات البياناتالموسوعةSOTAنماذج LLMلوحة الأداء GPUالفعاليات
البحث
حول
HyperAIHyperAI

Network Pruning On Imagenet Resnet 50 90

المقاييس

Top-1 Accuracy

النتائج

نتائج أداء النماذج المختلفة على هذا المعيار القياسي

اسم النموذج
Top-1 Accuracy
Paper TitleRepository
Feather76.93Feather: An Elegant Solution to Effective DNN Sparsification
AC/DC75.64AC/DC: Alternating Compressed/DeCompressed Training of Deep Neural Networks
GMP73.91Soft Threshold Weight Reparameterization for Learnable Sparsity
CS75.5Winning the Lottery with Continuous Sparsification
ST-376.03Are Straight-Through gradients and Soft-Thresholding all you need for Sparse Training?
ProbMask74.68Effective Sparsification of Neural Networks with Global Sparsity Constraint
STR74.31Soft Threshold Weight Reparameterization for Learnable Sparsity
DNW74Discovering Neural Wirings
Spartan76.17Spartan: Differentiable Sparsity via Regularized Transportation
0 of 9 row(s) selected.
Network Pruning On Imagenet Resnet 50 90 | SOTA | HyperAI