Merge GC notebooks in main

data/generate_pv_masks.py

@@ -0,0 +1,59 @@
+import numpy as np
+from PIL import Image
+from pathlib import Path
+
+def extract_pv_masks(
+    src_dir: Path,
+    dst_dir: Path,
+    pv_class: int = 0,
+    background_value: int = 255
+):
+    """
+    Extract masks that contain PV modules (class 0) and save a cleaned PV-only mask.
+
+    Parameters
+    ----------
+    src_dir : Path
+        Directory containing original RID superstructure masks.
+    dst_dir : Path
+        Output directory for pv-only masks.
+    pv_class : int, optional
+        Class index representing PV modules (default: 0).
+    background_value : int, optional
+        Value used for non-PV pixels in the output mask (default: 255 for easy visualization).
+    """
+    dst_dir.mkdir(parents=True, exist_ok=True)
+
+    mask_files = list(src_dir.glob("*.png"))
+
+    print(f"Found {len(mask_files)} masks to inspect...")
+    count_saved = 0
+
+    for mask_path in mask_files:
+        mask = np.array(Image.open(mask_path))
+
+        # Check whether PV class exists
+        if pv_class not in np.unique(mask):
+            continue  # skip masks without PV
+
+        # Create PV-only mask
+        pv_mask = np.where(mask == pv_class, pv_class, background_value).astype(np.uint8)
+
+        # Save to new directory
+        out_path = dst_dir / mask_path.name
+        Image.fromarray(pv_mask).save(out_path)
+        count_saved += 1
+
+    print(f"Saved {count_saved} PV-only masks to: {dst_dir}")
+
+if __name__ == "__main__":
+    BASE_DIR = Path(__file__).resolve().parent
+    SRC_MASK_DIR = BASE_DIR / "masks_superstructures_reviewed"
+    DST_PV_MASK_DIR = BASE_DIR / "masks_pv_modules_only"
+
+    extract_pv_masks(
+        src_dir=SRC_MASK_DIR,
+        dst_dir=DST_PV_MASK_DIR,
+        pv_class=0,
+        background_value=255,   # white background for visualization
+    )

data/load_rid_data.py

@@ -0,0 +1,85 @@
+#!/usr/bin/env python3
+"""
+Minimal setup script for the RID dataset.
+
+Place this script in your `data/` directory and run:
+
+    python setup_rid_data_minimal.py
+
+It will:
+- rsync ONLY:
+    - images_roof_centered_geotiff
+    - masks_superstructures_reviewed
+  from the TUM server
+- download them directly into THIS folder (data/)
+- skip downloading if the folder already exists and is non-empty
+"""
+
+import subprocess
+from pathlib import Path
+
+SERVER = "rsync://m1655470@dataserv.ub.tum.de/m1655470"
+BASE_DIR = Path(__file__).resolve().parent     # data/
+DATA_DIR = BASE_DIR
+
+FOLDERS = [
+    "images_roof_centered_geotiff",
+    "masks_superstructures_reviewed",
+]
+
+
+def folder_has_data(path: Path) -> bool:
+    """Return True if folder exists and contains at least one file."""
+    return path.exists() and any(path.iterdir())
+
+
+def rsync_folder(remote_name: str):
+    """Rsync a single folder from the server into DATA_DIR."""
+    dst = DATA_DIR / remote_name
+
+    # Skip if already present and non-empty
+    if folder_has_data(dst):
+        print(f"✔ Skipping '{remote_name}' — folder already exists and is not empty.")
+        return
+
+    # Ensure directory exists
+    dst.mkdir(parents=True, exist_ok=True)
+
+    src = f"{SERVER}/{remote_name}/"
+    cmd = [
+        "rsync",
+        "-av",
+        "--progress",
+        src,
+        str(dst) + "/",     # ensure trailing slash
+    ]
+
+    print(f">>> Downloading '{remote_name}' from server...")
+    print("    ", " ".join(cmd))
+    print(">>> You may be asked for the password (m1655470).")
+
+    try:
+        subprocess.run(cmd, check=True)
+    except FileNotFoundError:
+        raise SystemExit("ERROR: rsync not found. Please install rsync and try again.")
+    except subprocess.CalledProcessError as e:
+        raise SystemExit(f"ERROR: rsync for {remote_name} failed with exit code {e.returncode}.")
+
+
+def main():
+    print("=== Minimal RID dataset setup (only 2 folders, with existence check) ===")
+    print(f"Data directory: {DATA_DIR}")
+    print()
+
+    for folder in FOLDERS:
+        rsync_folder(folder)
+        print()
+
+    print("=== Done! ===")
+    print("Folders now present:")
+    for folder in FOLDERS:
+        print(" •", DATA_DIR / folder)
+
+
+if __name__ == "__main__":
+    main()

presentation/tutorial-new-tutorial-group-1-NARRATIVE.qmd

@@ -0,0 +1,163 @@
+---
+title: "Few-Shot Learning for Rooftop Detection in Satellite Imagery"
+subtitle: "Deep Learning Tutorial"
+author: "Giorgio Coppala, Nadine Daum, Elena Dreyer, Nico Reichardt"
+bibliography: refs.bib
+
+
+resources:
+  - img/**
+
+format:
+  revealjs:
+    theme: dimmery.scss
+    slide-number: true
+    default-image-width: 70%
+    preview-links: auto
+    logo: ""
+    footer: ""
+    transition: slide
+    background-transition: fade
+    self-contained: true
+    html-math-method:
+      method: mathjax
+      url: https://cdn.jsdelivr.net/npm/mathjax@4/tex-mml-chtml.js
+    include-in-header: include.html
+    resources:
+      - img/**
+---
+
+
+## Policy Relevance
+
+- Many public auhorities face the problem of **limited labeled data**
+  (annotation is expensive, slow, or requires domain expertise)
+
+- **Applications:**
+  - medical sector: **rare disease detection**
+  - emergency management: **flood extent mapping**
+  - climate & energy: **solar PV rooftop assessment**
+  - urban planning: **building footprints & infrastructure mapping**
+
+- **Few-shot learning (FSL)** can help:
+  - Learns to **generalize** from *1–5 labeled support examples per class*
+  - (in our case) learns a **feature embedding** and constructs **class prototypes**
+  - Enables segmentation in a **new city** with *minimal additional annotation*
+
+
+
+## Problem Setting
+
+::: {.columns}
+
+::: {.column width="55%"}
+
+- Goal of the tutorial: apply **Prototypical Networks** to
+rooftop segmentation using only a few labeled tiles
+
+- **Few-shot segmentation** allows the model to learn characteristic
+rooftop shapes and textures from a small Geneva subset
+
+- Demonstrates how rooftop maps can be produced for solar potential estimation in a **new geographic setting** with limited labels
+
+:::
+
+::: {.column width="45%"}
+![](figures/picture_use_case.png){width="100%" style="margin-top: 1rem;"}
+
+<div style="font-size: 0.75rem; color:#666; text-align:center; margin-top:0.2rem;">
+Demonstration use case (self-made visualization)
+</div>
+:::
+
+:::
+
+
+## Dataset: [Roofs of Geneva](https://huggingface.co/datasets/raphaelattias/overfitteam-geneva-satellite-images)
+
+- **Size**: 1,050 labeled image-mask pairs
+
+- **Task**: Binary segmentation masks (rooftop vs background)
+
+- **Geographic splits**: 3 grids/ neighborhoods (North, Center, South)
+
+- **Image size**: 250x250 pixels
+
+- **Categories**: Industrial, Residential
+
+
+## Inside the dataset
+
+<div style="text-align:center;">
+![](figures/grids_animation.gif){width="50%"}
+</div>
+
+<div style="font-size:0.75rem; text-align:center; color:#666; margin-top:0.5rem;">
+Geneva Animation: raw image → overlay rooftop → binary mask
+</div>
+
+
+## Discussion
+
+**Room for improvement:**
+
+- Fine-tune / tweak model parameters
+  - Add regularization
+  - Increase number of epochs
+
+- Implement rough approximation of solar potential
+  - e.g. based on IoU over roof area
+
+
+**Open for discussion:**
+
+- Try a different encoder ?
+  - e.g. ResNet-50
+
+- Change train / test split strategy ?
+  - e.g. random shuffle regardless of geographic regions
+
+
+
+<div style="text-align:center; margin-top:3.5em; font-size:1.1em;">
+  <a href="https://github.com/hertie-data-science-lab/tutorial-new-tutorial-group-1/tree/main"
+     target="_blank"
+     style="text-decoration:none;">
+    GitHub Repo
+  </a>
+</div>
+
+
+
+## References
+
+::: {.refs-super-small}
+
+- **Alsentzer, E., Li, M. M., Kobren, S. N., Noori, A., Undiagnosed Diseases Network, Kohane, I. S., & Zitnik, M.** (2025). Few shot learning for phenotype-driven diagnosis of patients with rare genetic diseases. *npj Digital Medicine, 8*(1), 380. https://doi.org/10.1038/s41746-025-01749-1
+
+- **Castello, R., Walch, A., Attias, R., Cadei, R., Jiang, S., & Scartezzini, J.-L.** (2021). Quantification of the suitable rooftop area for solar panel installation from overhead imagery using convolutional neural networks. *Journal of Physics: Conference Series, 2042*(1), 012002. https://doi.org/10.1088/1742-6596/2042/1/012002
+
+- **Chen, Y., Wei, C., Wang, D., Ji, C., & Li, B.** (2022). Semi-supervised contrastive learning for few-shot segmentation of remote sensing images. *Remote Sensing, 14*(17), 4254. https://doi.org/10.3390/rs14174254
+
+- **Ding, H., Zhang, H., & Jiang, X.** (2022). Self-regularized prototypical network for few-shot semantic segmentation. *Pattern Recognition, 132*, 109018. https://doi.org/10.1016/j.patcog.2022.109018
+
+- **Finn, C., Abbeel, P., & Levine, S.** (2017). Model-agnostic meta-learning for fast adaptation of deep networks. In *International Conference on Machine Learning* (pp. 1126–1135). https://doi.org/10.48550/arXiv.1703.03400
+
+- **Ge, Z., Fan, X., Zhang, J., & Jin, S.** (2025). SegPPD-FS: Segmenting plant pests and diseases in the wild using few-shot learning. *Plant Phenomics*, 100121. https://doi.org/10.1016/j.plaphe.2025.100121
+
+- **Hu, Y., Liu, C., Li, Z., Xu, J., Han, Z., & Guo, J.** (2022). Few-shot building footprint shape classification with relation network. *ISPRS International Journal of Geo-Information, 11*(5), 311. https://doi.org/10.3390/ijgi11050311
+
+- **Jadon, S.** (2021). COVID-19 detection from scarce chest X-ray image data using few-shot deep learning. In *Medical Imaging 2021* (pp. 161–170). https://doi.org/10.1117/12.2581496
+
+- **Lee, G. Y., Dam, T., Ferdaus, M. M., Poenar, D. P., & Duong, V.** (2025). Enhancing Few-Shot Classification of Benchmark and Disaster Imagery with ATTBHFA-Net. *arXiv preprint* arXiv:2510.18326. https://doi.org/10.48550/arXiv.2510.18326
+
+- **Li, X., He, Z., Zhang, L., Guo, S., Hu, B., & Guo, K.** (2025). CDCNet: Cross-domain few-shot learning with adaptive representation enhancement. *Pattern Recognition, 162*, 111382. https://doi.org/10.1016/j.patcog.2025.111382
+
+- **Puthumanaillam, G., & Verma, U.** (2023). Texture based prototypical network for few-shot semantic segmentation of forest cover: Generalizing for different geographical regions. *Neurocomputing, 538*, 126201. https://doi.org/10.1016/j.neucom.2023.03.062
+
+- **Snell, J., Swersky, K., & Zemel, R.** (2017). Prototypical networks for few-shot learning. *Advances in Neural Information Processing Systems, 30*. https://doi.org/10.48550/arXiv.1703.05175
+
+- **Sung, F., Yang, Y., Zhang, L., Xiang, T., Torr, P. H., & Hospedales, T. M.** (2018). Learning to compare: Relation network for few-shot learning. In *CVPR* (pp. 1199–1208). https://doi.org/10.1109/CVPR.2018.00131
+:::
+
+

presentation/tutorial-new-tutorial-group-1.qmd

@@ -4,9 +4,9 @@ subtitle: "Deep Learning Tutorial"
 author: "Giorgio Coppola, Nadine Daum, Elena Dreyer, Nico Reichardt"
 bibliography: refs.bib
 
-
 resources:
   - img/**
+  - figures/**
 
 format:
   revealjs:

tutorial-new-tutorial-group-1.Rproj

@@ -0,0 +1,15 @@
+Version: 1.0
+
+RestoreWorkspace: Default
+SaveWorkspace: Default
+AlwaysSaveHistory: Default
+
+EnableCodeIndexing: Yes
+UseSpacesForTab: Yes
+NumSpacesForTab: 2
+Encoding: UTF-8
+
+RnwWeave: Sweave
+LaTeX: pdfLaTeX
+
+BuildType: Makefile