Papers with Code · CC BY-SA 4.0

ImageNet

아카이브 평가 135건 · 마지막 적재/변경 2026-10-30

평가 기록

과거 수집 기록이며 최신 순위가 아닙니다. 데이터셋 버전·split·실험 조건과 지표 방향이 검증되지 않은 결과는 직접 비교할 수 없습니다. 날짜는 논문 날짜이며 최신 평가일을 뜻하지 않습니다.

표가 화면보다 넓으면 좌우로 스크롤하여 논문과 구현 링크를 확인하세요.

원본 기록 순서 · 순위 미부여
모델	원본 지표·값	논문	구현
DeepMAD-50M	Top-1 Error Rate 16.1 Accuracy 83.9 FLOPs 8.7G Params 50M	DeepMAD: Mathematical Architecture Design for Deep Convolutional Neural Network 2023-03-05	코드 1
ZenNAS (1.2ms)	Top-1 Error Rate 16.4 Accuracy 83.6 FLOPs 22G Params 180M	Zen-NAS: A Zero-Shot NAS for High-Performance Deep Image Recognition 2021-02-01	코드 1 · 코드 2
GPUNet-D3	Top-1 Error Rate 16.4 FLOPs 15.6G Params 19M	GPUNet: Searching the Deployable Convolution Neural Networks for GPUs 2022-04-26	코드 1
GPUNet-D1	Top-1 Error Rate 17.5 FLOPs 3.66G Params 10.6M	GPUNet: Searching the Deployable Convolution Neural Networks for GPUs 2022-04-26	코드 1
FBNetV3-G	Top-1 Error Rate 17.7 Accuracy 82.3 MACs 2.0G	FBNetV3: Joint Architecture-Recipe Search using Predictor Pretraining 2020-06-03	코드 1 · 코드 2
BossNet-T1+	Top-1 Error Rate 17.8 Accuracy 82.2 MACs 10.5G	BossNAS: Exploring Hybrid CNN-transformers with Block-wisely Self-supervised Neural Architecture Search 2021-03-23	코드 1
NASViT-A5	Top-1 Error Rate 18.2 Accuracy 81.8 FLOPs 757M	NASViT: Neural Architecture Search for Efficient Vision Transformers with Gradient Conflict aware Supernet Training 2021-09-29	코드 1
FBNetV5	Top-1 Error Rate 18.2 FLOPs 726M	FBNetV5: Neural Architecture Search for Multiple Tasks in One Run 2021-11-19	연결 없음
FBNetV5-A-CLS	Top-1 Error Rate 18.3 Accuracy 81.7 FLOPs 685M	FBNetV5: Neural Architecture Search for Multiple Tasks in One Run 2021-11-19	연결 없음
NASViT-A4	Top-1 Error Rate 18.6 Accuracy 81.4 FLOPs 591M	NASViT: Neural Architecture Search for Efficient Vision Transformers with Gradient Conflict aware Supernet Training 2021-09-29	코드 1
NASViT-A3	Top-1 Error Rate 19.0 Accuracy 81.0 FLOPs 528M	NASViT: Neural Architecture Search for Efficient Vision Transformers with Gradient Conflict aware Supernet Training 2021-09-29	코드 1
AlphaNet-A6	Top-1 Error Rate 19.2 Accuracy 80.8 FLOPs 709M	AlphaNet: Improved Training of Supernets with Alpha-Divergence 2021-02-16	코드 1 · 코드 2
UniNet-B0	Top-1 Error Rate 19.2 FLOPs 555M Params 11.5M	UniNet: Unified Architecture Search with Convolution, Transformer, and MLP 2022-07-12	코드 1 · 코드 2
AlphaNet-A5 (base)	Top-1 Error Rate 19.4 Accuracy 80.6 FLOPs 596M	AlphaNet: Improved Training of Supernets with Alpha-Divergence 2021-02-16	코드 1 · 코드 2
NAT-M4	Top-1 Error Rate 19.5 Accuracy 80.5 Params 9.1M MACs 600M	Neural Architecture Transfer 2020-05-12	코드 1 · 코드 2
NASViT-A2	Top-1 Error Rate 19.5 Accuracy 80.5 FLOPs 421M	NASViT: Neural Architecture Search for Efficient Vision Transformers with Gradient Conflict aware Supernet Training 2021-09-29	코드 1
NSGANetV2-xl	Top-1 Error Rate 19.6 Accuracy 80.4 Params 8.7M MACs 593M	NSGANetV2: Evolutionary Multi-Objective Surrogate-Assisted Neural Architecture Search 2020-07-20	코드 1
FBNetV3-E	Top-1 Error Rate 19.6 Accuracy 80.4 MACs 752M	FBNetV3: Joint Architecture-Recipe Search using Predictor Pretraining 2020-06-03	코드 1 · 코드 2
AlphaNet-A5 (small)	Top-1 Error Rate 19.7 Accuracy 80.3 FLOPs 491M	AlphaNet: Improved Training of Supernets with Alpha-Divergence 2021-02-16	코드 1 · 코드 2
ISyNet-N3	Top-1 Error Rate 19.84	ISyNet: Convolutional Neural Networks design for AI accelerator 2021-09-04	코드 1 · 코드 2 · 코드 3 · 코드 4 · 코드 5 · 코드 6 · 코드 7 · 코드 8 · 코드 9
HardcoreNAS_E_KD	Top-1 Error Rate 19.9 Accuracy 80.1 MACs 330M	HardCoRe-NAS: Hard Constrained diffeRentiable Neural Architecture Search 2021-02-23	코드 1 · 코드 2
AttentiveNAS-A5	Top-1 Error Rate 19.9 Accuracy 80.1 MACs 491M	AttentiveNAS: Improving Neural Architecture Search via Attentive Sampling 2020-11-18	코드 1 · 코드 2
AlphaNet-A4	Top-1 Error Rate 20.0 Accuracy 80.0 FLOPs 444M	AlphaNet: Improved Training of Supernets with Alpha-Divergence 2021-02-16	코드 1 · 코드 2
NAT-M3	Top-1 Error Rate 20.1 Accuracy 79.9 Params 9.1M MACs 490M	Neural Architecture Transfer 2020-05-12	코드 1 · 코드 2
AttentiveNAS-A4	Top-1 Error Rate 20.2 Accuracy 79.8 MACs 444M	AttentiveNAS: Improving Neural Architecture Search via Attentive Sampling 2020-11-18	코드 1 · 코드 2
SEDONA (ResNet-152 Aux. Ens., K=4)	Top-1 Error Rate 20.2	SEDONA: Search for Decoupled Neural Networks toward Greedy Block-wise Learning 2021-01-01	코드 1
NASViT-A1	Top-1 Error Rate 20.3 Accuracy 79.7 FLOPs 309M	NASViT: Neural Architecture Search for Efficient Vision Transformers with Gradient Conflict aware Supernet Training 2021-09-29	코드 1
GPUNet-D0	Top-1 Error Rate 20.3 FLOPs 720M Params 6.2M	GPUNet: Searching the Deployable Convolution Neural Networks for GPUs 2022-04-26	코드 1
FBNetV3-C	Top-1 Error Rate 20.4 Accuracy 79.6 MACs 544M	FBNetV3: Joint Architecture-Recipe Search using Predictor Pretraining 2020-06-03	코드 1 · 코드 2
BigNASModel-L	Top-1 Error Rate 20.5 Accuracy 79.5 Params 6.4M MACs 586M	BigNAS: Scaling Up Neural Architecture Search with Big Single-Stage Models 2020-03-24	코드 1