=== versions ===
python 3.12.0 | torch 2.14.1+cpu | segmentation-models-pytorch 0.5.0 | torchgeo 0.10.1 | rasterio 1.5.2 | numpy 2.5.2
device: CPU (8 threads), CUDA available: False

=== chips ===
loaded 576 chips in 4.8 s: X (576, 12, 128, 128) uint16, Y (576, 128, 128) uint8
bands: B01, B02, B03, B04, B05, B06, B07, B08, B8A, B09, B11, B12
split
train    416
val       80
test      80

=== model ===
loaded SENTINEL2_ALL_MOCO into the encoder: 120 tensors, classifier head dropped
U-Net, ResNet-18 encoder: 14.4 M parameters (11.2 M in the encoder)

=== training: 30 epochs, batch 16, AdamW + one-cycle, weighted cross-entropy + Dice ===
class weights: Building 1.27, Other built-up 0.71, Trees 0.53, Grass and crops 0.32, Water 2.17
pretrained encoder     epoch  1: train loss 2.922 | val loss 2.599 | val mIoU 0.090 | 0.1 min
pretrained encoder     epoch  2: train loss 2.248 | val loss 2.326 | val mIoU 0.243 | 0.3 min
pretrained encoder     epoch  5: train loss 1.128 | val loss 2.197 | val mIoU 0.323 | 0.7 min
pretrained encoder     epoch 10: train loss 0.874 | val loss 0.932 | val mIoU 0.564 | 1.5 min
pretrained encoder     epoch 15: train loss 0.872 | val loss 0.992 | val mIoU 0.547 | 2.2 min
pretrained encoder     epoch 20: train loss 0.803 | val loss 0.839 | val mIoU 0.611 | 3.1 min
pretrained encoder     epoch 25: train loss 0.771 | val loss 0.800 | val mIoU 0.622 | 3.9 min
pretrained encoder     epoch 30: train loss 0.766 | val loss 0.799 | val mIoU 0.622 | 4.8 min
pretrained encoder: 416 training chips, 4.8 min; best val mIoU 0.623 at epoch 29
from scratch           epoch  1: train loss 2.815 | val loss 2.648 | val mIoU 0.022 | 0.2 min
from scratch           epoch  2: train loss 2.246 | val loss 2.170 | val mIoU 0.256 | 0.3 min
from scratch           epoch  5: train loss 1.171 | val loss 3.087 | val mIoU 0.160 | 0.7 min
from scratch           epoch 10: train loss 0.930 | val loss 1.095 | val mIoU 0.543 | 1.4 min
from scratch           epoch 15: train loss 0.882 | val loss 0.891 | val mIoU 0.585 | 2.1 min
from scratch           epoch 20: train loss 0.836 | val loss 1.036 | val mIoU 0.564 | 2.9 min
from scratch           epoch 25: train loss 0.798 | val loss 0.823 | val mIoU 0.612 | 3.9 min
from scratch           epoch 30: train loss 0.794 | val loss 0.809 | val mIoU 0.617 | 4.9 min
from scratch: 416 training chips, 4.9 min; best val mIoU 0.619 at epoch 28
random split (leaky)   epoch  1: train loss 2.938 | val loss 2.573 | val mIoU 0.096 | 0.2 min
random split (leaky)   epoch  2: train loss 2.270 | val loss 2.292 | val mIoU 0.246 | 0.4 min
random split (leaky)   epoch  5: train loss 1.163 | val loss 1.141 | val mIoU 0.551 | 0.9 min
random split (leaky)   epoch 10: train loss 0.943 | val loss 0.981 | val mIoU 0.560 | 1.8 min
random split (leaky)   epoch 15: train loss 0.902 | val loss 0.936 | val mIoU 0.591 | 2.8 min
random split (leaky)   epoch 20: train loss 0.851 | val loss 0.802 | val mIoU 0.610 | 3.7 min
random split (leaky)   epoch 25: train loss 0.808 | val loss 0.786 | val mIoU 0.624 | 4.7 min
random split (leaky)   epoch 30: train loss 0.819 | val loss 0.773 | val mIoU 0.633 | 5.5 min
random split (leaky): 386 training chips, 5.5 min; best val mIoU 0.633 at epoch 30

=== test set: 80 chips in 5 km blocks never seen in training (1.31 M pixels) ===
                   IoU     F1  precision  recall  share
Building         0.210  0.347      0.427   0.293  0.040
Other built-up   0.581  0.735      0.672   0.811  0.123
Trees            0.711  0.831      0.819   0.843  0.260
Grass and crops  0.878  0.935      0.954   0.917  0.562
Water            0.837  0.911      0.874   0.952  0.015

                          overall accuracy  mean IoU  macro F1  Building IoU
pretrained, block split              0.860     0.643     0.752         0.210
scratch, block split                 0.858     0.638     0.745         0.190
pretrained, random split             0.851     0.616     0.730         0.182

confusion matrix (rows: reference label, columns: prediction, % of each reference class):
                 Building  Other built-up  Trees  Grass and crops  Water
Building             29.3            61.1    7.9              1.7    0.0
Other built-up       10.5            81.1    6.4              1.8    0.2
Trees                 0.9             6.0   84.3              8.5    0.4
Grass and crops       0.1             1.5    6.6             91.7    0.1
Water                 0.4             1.5    2.6              0.3   95.2

neighbouring test pixels with the same label: 88.8%

saved unet_resnet18_s2_nottingham.pt (58 MB)
figures written to output/