Update dataset documentation and metadata

- Modified data/readme.md with improved dataset descriptions and instructions.
- Added dataset.csv mapping images to their corresponding segmentation masks.
- Added dataset_infos.json with detailed dataset metadata and feature descriptions.

data/readme.md

@@ -0,0 +1,32 @@
+# Urban Climate Dataset
+
+## Dataset Description
+
+This dataset is part of the Urban Climate project. It contains satellite images of urban areas and corresponding segmentation masks that are used for building detection and segmentation tasks.
+
+### Files and Structure:
+
+- **train/**: Contains training images and masks.
+  - **train/Image/**: Original satellite images used for training.
+  - **train/Mask/**: Binary masks where buildings are labeled.
+- **validation/**: Contains validation images and masks for model evaluation.
+  - **validation/Image/**: Satellite images for validation.
+  - **validation/Mask/**: Validation masks corresponding to the images.
+- **test/**: Contains test images and masks to evaluate the model.
+  - **test/Image/**: Satellite images for testing.
+  - **test/Mask/**: Test masks corresponding to the images.
+
+---
+
+configs:
+
+- config_name: default
+  data_files:
+  - split: train
+    path: "data/train/\*"
+  - split: validation
+    path: "data/validation/\*"
+  - split: test
+    path: "data/test/\*"
+
+---

dataset_infos.json

@@ -0,0 +1,17 @@
+{
+  "urban-climate-building detection": {
+    "description": "This dataset is designed for the task of building detection in urban environments using image segmentation. It includes high-resolution aerial or satellite images of urban areas, paired with corresponding segmentation masks that delineate building footprints. The dataset is structured into training, validation, and test sets, and is suitable for developing and evaluating models for urban climate analysis and automated building detection.",
+    "features": {
+      "image": {
+        "dtype": "Image",
+        "description": "High-resolution input images of urban environments, containing various building structures. These images serve as the input for segmentation models to detect and outline building footprints."
+      },
+      "mask": {
+        "dtype": "Image",
+        "description": "Binary segmentation masks corresponding to each input image. Each mask highlights the building footprints, where the buildings are marked in one color (e.g., white) and the background is marked in another color (e.g., black). These masks are used as ground truth for training and evaluating segmentation models."
+      }
+    },
+    "license": "CC BY-SA 4.0",
+    "citation": "@article{your_citation}"
+  }
+}