基准	方法	指标
breast-tumour-classification-on-pcam	ResNet-34 (e)	AUC: 0.942
breast-tumour-classification-on-pcam	ResNet-50 (e)	AUC: 0.948
classification-on-ximagenet-12	ResNet 50	Robustness Score: 0.8985
crowd-counting-on-ucf-qnrf	Resnet101	MAE: 190
domain-adaptation-on-office-31	ResNet-50	Average Accuracy: 76.1
domain-generalization-on-imagenet-a	ResNet-50 (300 Epochs)	Top-1 accuracy %: 4.2
domain-generalization-on-imagenet-r	ResNet-50	Top-1 Error Rate: 63.9
domain-generalization-on-vizwiz	ResNet-101	Accuracy - All Images: 46.3 Accuracy - Clean Images: 50.1 Accuracy - Corrupted Images: 40.5
domain-generalization-on-vizwiz	ResNet-50	Accuracy - All Images: 42.9 Accuracy - Clean Images: 47.7 Accuracy - Corrupted Images: 37.1
domain-generalization-on-vizwiz	ResNet-152	Accuracy - All Images: 47.5 Accuracy - Clean Images: 51.3 Accuracy - Corrupted Images: 43.3
image-classification-on-cifar100	shreynet	1:1 Accuracy: 45.98
image-classification-on-gashissdb	ResNet-18	Accuracy: 98.47 F1-Score: 99.19 Precision: 99.94
image-classification-on-gashissdb	ResNet-50	Accuracy: 98.56 F1-Score: 99.24 Precision: 99.94
image-classification-on-imagenet	ResNet-101	GFLOPs: 7.6 Number of params: 40M Top 1 Accuracy: 78.25%
image-classification-on-imagenet	ResNet-50	GFLOPs: 3.8 Number of params: 25M Top 1 Accuracy: 75.3%
image-classification-on-imagenet	ResNet-152	GFLOPs: 11.3 Top 1 Accuracy: 78.57%
image-classification-on-omnibenchmark	ResNet-101	Average Top-1 Accuracy: 37.4
image-classification-on-omnibenchmark	ResNet-50	Average Top-1 Accuracy: 34.3
image-to-image-translation-on-gtav-to	ResNet101 65.1	mIoU: 41.7
medical-image-classification-on-nct-crc-he	ResNet-18	Accuracy (%): 92.66 F1-Score: 95.23 Precision: 99.90 Specificity: 99.08
medical-image-classification-on-nct-crc-he	ResNet-50	Accuracy (%): 94.72 F1-Score: 97.09 Precision: 100.00 Specificity: 99.34
multi-label-image-classification-on-vizwiz	ResNet151	Accuracy: 47.5
object-detection-on-coco	Faster R-CNN (box refinement, context, multi-scale testing)	box mAP: 34.9
object-detection-on-coco-minival	ATSS (ResNet-50)	AP50: 61.9 AP75: 47.0 box AP: 43.5
object-detection-on-coco-minival	GFL (ResNet-50)	AP50: 63.0 AP75: 48.3 box AP: 44.5
object-detection-on-coco-minival	Cascade Mask R-CNN (ResNet-50)	AP50: 64.3 AP75: 50.5 box AP: 46.3
pedestrian-attribute-recognition-on-uav-human	ResNet	Backpack: 63.5 Gender: 74.7 Hat: 65.2 LCC: 49.7 LCS: 69.3 UCC: 44.4 UCS: 68.9
person-re-identification-on-sysu-30k	ResNet-50 (generalization)	Rank-1: 20.1
retinal-oct-disease-classification-on	ResNet50-v1	Acc: 94.92
retinal-oct-disease-classification-on-oct2017	ResNet50-v1	Acc: 99.3 Sensitivity: 99.3
semantic-segmentation-on-cityscapes-val	Dilated-ResNet (Dilated-ResNet-101)	mIoU: 75.7
semantic-segmentation-on-dada-seg	ResNet-50	mIoU: 18.96
semantic-segmentation-on-dada-seg	ResNet-101	mIoU: 23.60
synthetic-to-real-translation-on-syn2real-c	No Adaptation	Accuracy: 52.4
unsupervised-domain-adaptation-on-office-home	ResNet-50 [cite:CVPR16DRL]	Accuracy: 59.9

用 AI 构建 AI

从想法到上线——通过免费 AI 协同编程、开箱即用的环境和市场最优价格的 GPU 加速您的 AI 开发

AI 协同编程

即用型 GPU

最优价格

立即开始

Hyper Newsletters

订阅我们的最新资讯

我们会在北京时间 每周一的上午九点 向您的邮箱投递本周内的最新更新

邮件发送服务由 MailChimp 提供