gabrielkasmi commited on
Commit
b9a3fb6
·
verified ·
1 Parent(s): a2aee5d

Update README.md

Browse files
Files changed (1) hide show
  1. README.md +20 -2
README.md CHANGED
@@ -58,6 +58,8 @@ Rules:
58
  - The spatial holdout by department must not be modified. Re-splitting invalidates comparability with published results.
59
  - For Track 3, only the Google training split may be used for training.
60
 
 
 
61
  ---
62
  ## Results
63
 
@@ -112,12 +114,28 @@ Models trained on the official BDAPPV splits using:
112
 
113
  - **Optimizer:** AdamW (lr=1e-4, weight_decay=1e-4)
114
  - **Scheduler:** Cosine annealing
115
- - **Effective batch size:** 32 (batch 16 × grad accum 2)
116
  - **Early stopping:** patience=7 epochs on validation metric
117
  - **Input size:** 400×400 px
118
  - **Initialization:** checkpoints from [Mayer et al. (2022)](https://doi.org/10.1016/j.apenergy.2021.118469), who fine-tuned DeepLabV3-ResNet101 and InceptionV3 on 10 cm/px orthoimagery from North Rhine-Westphalia (Germany) for rooftop PV detection. These checkpoints were then further fine-tuned on BDAPPV using the splits above.
119
 
120
- Training scripts available in the [BDAPPV dataset repository](https://huggingface.co/datasets/gabrielkasmi/bdappv).
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
121
 
122
  ---
123
 
 
58
  - The spatial holdout by department must not be modified. Re-splitting invalidates comparability with published results.
59
  - For Track 3, only the Google training split may be used for training.
60
 
61
+ *IoU/F1 above are foreground (PV-class) only — see [`train/evaluate.py`](train/evaluate.py#L31) for the exact computation.*
62
+
63
  ---
64
  ## Results
65
 
 
114
 
115
  - **Optimizer:** AdamW (lr=1e-4, weight_decay=1e-4)
116
  - **Scheduler:** Cosine annealing
117
+ - **Effective batch size:** 32 (batch_size × grad_accum — see each script's defaults)
118
  - **Early stopping:** patience=7 epochs on validation metric
119
  - **Input size:** 400×400 px
120
  - **Initialization:** checkpoints from [Mayer et al. (2022)](https://doi.org/10.1016/j.apenergy.2021.118469), who fine-tuned DeepLabV3-ResNet101 and InceptionV3 on 10 cm/px orthoimagery from North Rhine-Westphalia (Germany) for rooftop PV detection. These checkpoints were then further fine-tuned on BDAPPV using the splits above.
121
 
122
+
123
+ Training scripts are included in this repo, under [`train/`](train):
124
+
125
+ | File | What it does |
126
+ |---|---|
127
+ | `train_seg.py` | Trains DeepLabV3-ResNet101 (segmentation) |
128
+ | `train_clf.py` | Trains InceptionV3 (classification) |
129
+ | `evaluate.py` | Computes the IoU/F1 numbers reported below (foreground-only, on the official test split) |
130
+ | `dataset.py` | Dataset loaders for both tasks |
131
+ | `pretrained/` | Base checkpoints from Mayer et al. (2022) — starting point before BDAPPV fine-tuning |
132
+
133
+ Reproduce (segmentation, IGN):
134
+ \`\`\`bash
135
+ pip install -r train/requirements.txt
136
+ python train/train_seg.py --provider ign --checkpoint train/pretrained/deeplabv3_weights.tar
137
+ python train/evaluate.py seg --provider ign --checkpoint train/weights/deeplab_ign_best.pth
138
+ \`\`\`
139
 
140
  ---
141