v1.0

The following files contains the Swin2-MoSE pretrained model on the Sen2venus (4x and 2x) and OLI2MSI (3x) datasets.

Swin2-MoSE pretrained after experiment 4:

• sen2venus_exp4_4x_v5.zip - training Swin2-MoSE on 4x task of Sen2Venus dataset
• oli2msi_exp4_3x_v5.zip - training Swin2-MoSE on 3x task of OLI2MSI dataset
• sen2venus_exp4_2x_v5.zip - training Swin2-MoSE on 2x task of Sen2Venus dataset

Train / Test csv files for Sen2Venus dataset:

• sen2venus_train_test_csv.zip - It contains the csv files we created to split sen2venus dataset. It can be used as input of the script scripts/sen2venus/rebuild.py (see README.md)

To use pretrained models (loading the model-epoch.pt file) and/or test fast evaluation of the models themself (loading the result-epoch.pt file), unpack files as show here:

output/
├── oli2msi_exp4_3x_v5
│   ├── cfg.yml
│   ├── checkpoints
│   │   └── model-70.pt
│   └── eval
│       └── results-70.pt
├── sen2venus_exp4_2x_v5
│   ├── cfg.yml
│   ├── checkpoints
│   │   └── model-70.pt
│   └── eval
│       └── results-70.pt
└── sen2venus_exp4_4x_v5
    ├── cfg.yml
    ├── checkpoints
    │   └── model-70.pt
    └── eval
        └── results-70.pt

As written inside the README, we recap how you can run the training/eval:

Train

python src/main.py --phase train --config $CONFIG_FILE --output $OUT_DIR --epochs ${EPOCH} --epoch -1

Validate

python src/main.py --phase test --config $CONFIG_FILE --output $OUT_DIR --epoch ${EPOCH}
python src/main.py --phase test --config $CONFIG_FILE --batch_size 32 --eval_method bicubic

Show results

python src/main.py --phase vis --config $CONFIG_FILE --output $OUT_DIR --num_images 3 --epoch ${EPOCH}
python src/main.py --phase vis --config $CONFIG_FILE --output output/sen2venus_4x_bicubic --num_images 3 --eval_method bicubic
python src/main.py --phase vis --config $CONFIG_FILE --output output/sen2venus_4x_bicubic --num_images 3 --eval_method bicubic --dpi 1200

Compute mean/std

python src/main.py --phase mean_std --config $CONFIG_FILE

Measure execution average time

python src/main.py --phase avg_time --config $CONFIG_FILE --repeat_times 1000 --warm_times 20 --batch_size 8