|
gptkbp:instanceOf
|
gptkb:convolutional_neural_network
deep learning model architecture
|
|
gptkbp:activatedBy
|
gptkb:GELU
|
|
gptkbp:architecture
|
pure convolutional
modernized ResNet
|
|
gptkbp:author
|
gptkb:Zhuang_Liu
gptkb:Trevor_Darrell
gptkb:Saining_Xie
gptkb:Chao-Yuan_Wu
gptkb:Christoph_Feichtenhofer
gptkb:Hanzi_Mao
|
|
gptkbp:bench
|
ImageNet top-1 accuracy
|
|
gptkbp:developedBy
|
gptkb:Facebook_AI_Research
|
|
gptkbp:hasVariant
|
ConvNeXT-B
ConvNeXT-L
ConvNeXT-S
ConvNeXT-T
ConvNeXT-XL
|
|
https://www.w3.org/2000/01/rdf-schema#label
|
ConvNeXT
|
|
gptkbp:improves
|
gptkb:ResNet
gptkb:EfficientNet
|
|
gptkbp:inspiredBy
|
gptkb:Vision_Transformer
|
|
gptkbp:introducedIn
|
2022
|
|
gptkbp:normalization
|
gptkb:LayerNorm
|
|
gptkbp:notableFeature
|
depthwise convolution
inspired by transformer design choices
large kernel sizes
|
|
gptkbp:notablePublication
|
A ConvNet for the 2020s
https://arxiv.org/abs/2201.03545
|
|
gptkbp:openSource
|
yes
|
|
gptkbp:parameter
|
28M (ConvNeXT-T)
88M (ConvNeXT-L)
|
|
gptkbp:publishedIn
|
gptkb:arXiv
|
|
gptkbp:repository
|
https://github.com/facebookresearch/ConvNeXT
|
|
gptkbp:trainer
|
ImageNet-1K
ImageNet-22K
|
|
gptkbp:usedFor
|
computer vision
image classification
|
|
gptkbp:bfsParent
|
gptkb:Hugging_Face_models
|
|
gptkbp:bfsLayer
|
7
|