Explore model architecture

[rubentea16]

Dataset: RESISC45

Options:

• Resnet50 (25M params) : https://arxiv.org/pdf/1512.03385.pdf
• CoAtNet : https://arxiv.org/abs/2106.04803 (model sampe v7)
• Vision Transformer (ViT) : https://arxiv.org/abs/2010.11929v2 ViT-Base 12 layers, 768 hidden size, 3072 MLP size, 12 Heads, 86M Params
• VOLO : https://arxiv.org/pdf/2106.13112v2.pdf https://github.com/sail-sg/volo

https://paperswithcode.com/sota/image-classification-on-imagenet

[adhiiisetiawan]

• CoAtNet https://arxiv.org/abs/2106.04803 https://paperswithcode.com/sota/image-classification-on-imagenet

• Vision Transformer (ViT) https://arxiv.org/abs/2010.11929v2

[rubentea16]

• ResNet50 -> @rubentea16
• ViT -> @adhiiisetiawan

[rubentea16]

@dewabratapandu

[rubentea16]

Regnet ->https://arxiv.org/abs/2003.13678 -> @rubentea16