HyperAI초신경
홈
뉴스
최신 연구 논문
튜토리얼
데이터셋
백과사전
SOTA
LLM 모델
GPU 랭킹
컨퍼런스
전체 검색
소개
한국어
HyperAI초신경
Toggle sidebar
전체 사이트 검색...
⌘
K
홈
SOTA
이미지 분류
Image Classification On Emnist Balanced
Image Classification On Emnist Balanced
평가 지표
Accuracy
Trainable Parameters
평가 결과
이 벤치마크에서 각 모델의 성능 결과
Columns
모델 이름
Accuracy
Trainable Parameters
Paper Title
Repository
CNN(Spinal FC)
83.21
16050
SpinalNet: Deep Neural Network with Gradual Input
EMNIST-Tiny
91.20
400000
Efficient Global Neural Architecture Search
OnDev-LCT-4/1
89.39
514960
OnDev-LCT: On-Device Lightweight Convolutional Transformers towards federated learning
-
Convolutional PMM (Parametric Matrix Model)
85.95
349172
Parametric Matrix Models
-
HM2-BP
85.57
665647
Hybrid Macro/Micro Level Backpropagation for Training Deep Spiking Neural Networks
SNNL-5
78.5
-
Improving k-Means Clustering Performance with Disentangled Internal Representations
CNN
79.61
21840
SpinalNet: Deep Neural Network with Gradual Input
TextCaps
90.46
-
Dynamic Routing Between Capsules
WaveMixLite-128/7
91.06
-
WaveMix: A Resource-efficient Neural Network for Image Analysis
Linear Classifier
50.93
-
EMNIST: an extension of MNIST to handwritten letters
VGG-5
91.04
3646000
SpinalNet: Deep Neural Network with Gradual Input
EMNIST-mobile
91.48
2250000
Efficient Global Neural Architecture Search
OnDev-LCT-1/1
89.52
216208
OnDev-LCT: On-Device Lightweight Convolutional Transformers towards federated learning
-
CNN(Spinal FC)
82.77
13820
SpinalNet: Deep Neural Network with Gradual Input
OnDev-LCT-2/1
89.18
315792
OnDev-LCT: On-Device Lightweight Convolutional Transformers towards federated learning
-
OptConv+Log+Perc
87.69
-
Efficient Neural Vision Systems Based on Convolutional Image Acquisition
-
VGG-5(Spinal FC)
91.05
3630000
SpinalNet: Deep Neural Network with Gradual Input
OnDev-LCT-8/1
89.55
913296
OnDev-LCT: On-Device Lightweight Convolutional Transformers towards federated learning
-
PMM (Parametric Matrix Model)
81.57
13792
Parametric Matrix Models
-
OPIUM Classifier
78.94
-
EMNIST: an extension of MNIST to handwritten letters
0 of 20 row(s) selected.
Previous
Next