Dhruv1000/AgriFM-PASTIS

AgriFM × PASTIS

Reimplementation of AgriFM — A Multi-source Temporal Remote Sensing Foundation Model for Agriculture Mapping — adapted for the PASTIS dataset on AMD GPU (ROCm 7.0).

Results (Fold 1, trained from scratch, 32 minutes)

Metric	Value
mFscore	58.37%
mIoU	44.35%
OA	67.10%
Kappa	60.02%
mPrecision	52.78%
mRecall	70.72%

Per-Class IoU

Class	IoU
Corn	76.76%
Winter rapeseed	74.17%
Beet	73.32%
Soft winter wheat	72.22%
Soybeans	66.81%
Meadow	56.56%
Winter barley	56.56%
Sunflower	53.94%
Background	48.39%
Winter durum wheat	43.30%
Potatoes	37.33%
Grapevine	36.16%
Spring barley	30.51%
Winter triticale	25.01%
Leguminous fodder	23.17%
Fruits/veg/flowers	22.27%
Mixed cereal	17.63%
Orchard	15.25%
Sorghum	13.29%

Model

• Backbone: Video Swin Transformer (synchronized spatiotemporal downsampling)
• Neck: MultiFusionNeck (U-Net style skip connections)
• Head: CropFCNHead
• Parameters: 39.6M (small config)
• Input: Sentinel-2, 32 temporal frames, 128×128 pixels, 10 bands
• Framework: Pure PyTorch, AMD ROCm 7.0 compatible

Training

python train.py \
    --data_root /path/to/PASTIS \
    --fold 1 \
    --small_model \
    --epochs 100 \
    --batch_size 16 \
    --lr 5e-5 \
    --weight_decay 0.05

Citation

@article{li2025agrifm,
  title={AgriFM: A Multi-source Temporal Remote Sensing Foundation Model for Agriculture Mapping},
  author={Li, Wenyuan and others},
  journal={Remote Sensing of Environment},
  year={2025}
}