PM2.5 forecasting on Remote Sensing Dataset (test)
0.68Pearson RViT-ICKy
Evaluation Results
| Method | Links | ||||
|---|---|---|---|---|---|
| ViT-ICKyBackbone Architecture=Vision Transformer (ViT), Joint Model Type=ICKy2022.05 | 0.68 | 56.56 | 41.41 | 12,208 | |
| S. MAE-ViT-RFBackbone Architecture=Vision Transformer (ViT), Seasonality Modeling=true, Joint Model Type=Random Forest, Pre-training=Masked Autoencoder (MAE)2022.05 | 0.67 | 53.87 | 40.78 | 31.09 | |
| S. ViT-RFBackbone Architecture=Vision Transformer (ViT), Seasonality Modeling=true, Joint Model Type=Random Forest2022.05 | 0.66 | 56.45 | 41.73 | 14.69 | |
| DeepViT-ICKyBackbone Architecture=DeepViT, Joint Model Type=ICKy2022.05 | 0.66 | 52.41 | 35.93 | 38,220 | |
| S. Deep-ViT-RFBackbone Architecture=DeepViT, Seasonality Modeling=true, Joint Model Type=Random Forest2022.05 | 0.65 | 56.36 | 42.46 | 17.63 | |
| S. CNN-RFBackbone Architecture=CNN, Seasonality Modeling=true, Joint Model Type=Random Forest2022.05 | 0.62 | 53.36 | 39.38 | 96.77 | |
| CNN-ICKyBackbone Architecture=CNN, Joint Model Type=ICKy2022.05 | 0.62 | 53.46 | 39.76 | 10.92 | |
| ViT-RFBackbone Architecture=Vision Transformer (ViT), Seasonality Modeling=false, Joint Model Type=Random Forest2022.05 | 0.07 | 190.82 | 181.63 | — | |
| CNN-RFBackbone Architecture=CNN, Seasonality Modeling=false, Joint Model Type=Random Forest2022.05 | 0 | 194.63 | 185.83 | — |