feat(dataset): add evaluation annotations and semantic labels

README.md

@@ -14,11 +14,48 @@ tags:
 - benchmark
 - neural-symbolic
 - multimodal-large-language-models
+configs:
+- config_name: task2
+  data_files:
+  - split: easy
+    path: Task2/parquet/easy/*.parquet
+  - split: medium
+    path: Task2/parquet/medium/*.parquet
+  - split: hard
+    path: Task2/parquet/hard/*.parquet
+- config_name: task3
+  data_files:
+  - split: easy
+    path: Task3/parquet/easy/*.parquet
+  - split: medium
+    path: Task3/parquet/medium/*.parquet
+  - split: hard
+    path: Task3/parquet/hard/*.parquet
+- config_name: task2-evaluation
+  data_files:
+  - split: easy
+    path: Task2/evaluation/easy.parquet
+  - split: medium
+    path: Task2/evaluation/medium.parquet
+  - split: hard
+    path: Task2/evaluation/hard.parquet
+- config_name: task3-evaluation
+  data_files:
+  - split: easy
+    path: Task3/evaluation/easy.parquet
+  - split: medium
+    path: Task3/evaluation/medium.parquet
+  - split: hard
+    path: Task3/evaluation/hard.parquet
+- config_name: task3-labels
+  data_files:
+  - split: train
+    path: Task3/labels/labels.parquet
 ---
 
 <div align="center">
 
-<h1>NeSy-Route: A Neural-Symbolic Benchmark for Constrained Route Planning in Remote Sensing</h1>
+<h1>NeSy-Route: A Neuro-Symbolic Benchmark for Constrained Route Planning in Remote Sensing</h1>
 
 <p>
   <a href="https://mingyang1010.github.io/NeSy-Route/">
@@ -46,10 +83,28 @@ NeSy-Route is organized into three benchmark tasks for evaluating perception, sy
 | Task 2 | Zero-shot | 7,659 | 3,712 | 1,604 | 12,975 | Constraint-aware semantic and region reasoning |
 | Task 3 | Zero-shot | 6,492 | 2,705 | 1,624 | 10,821 | Constrained route planning with waypoint or trajectory outputs |
 
-Task 1 and Task 2 provide semantic constraint annotations for vector-based reasoning. Task 3 provides route-planning samples together with traversability maps, cost maps, ground-truth paths, and difficulty-level subsets.
+Task 1 and Task 2 provide semantic constraint annotations for vector-based reasoning. Task 3 provides route-planning samples with endpoint annotations, optimal ground-truth trajectories, trajectory visualizations, and difficulty-level subsets.
 
 Task 2 annotated images are packaged as Parquet shards under `Task2/parquet/{easy,medium,hard}/`. Each row includes the annotated image bytes and the corresponding query, region, and difficulty metadata.
 
+Task 3 samples are packaged as Parquet shards under `Task3/parquet/{easy,medium,hard}/`. Each row includes the original remote-sensing image, query and endpoint metadata, the optimal ground-truth trajectory, and its visualization.
+
+Lightweight scoring annotations are available under `Task2/evaluation/` and `Task3/evaluation/`. Task 3 also provides 3,109 deduplicated semantic masks in a single `Task3/labels/labels.parquet` file. The evaluation code combines each mask with the sample-specific traversability and cost vectors to reconstruct the maps used by AR, CR, and VR.
+
+## Loading the Data
+
+```python
+from datasets import load_dataset
+
+task2 = load_dataset("Ming1010/NeSy-Route", "task2")
+task3 = load_dataset("Ming1010/NeSy-Route", "task3")
+
+# Lightweight ground truth used by the public scorer
+task2_eval = load_dataset("Ming1010/NeSy-Route", "task2-evaluation")
+task3_eval = load_dataset("Ming1010/NeSy-Route", "task3-evaluation")
+task3_labels = load_dataset("Ming1010/NeSy-Route", "task3-labels", split="train")
+```
+
 ## Evaluation Code
 
 Evaluation scripts, prompt templates, and running instructions are available in the GitHub repository:

Task2/evaluation/easy.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:a8106bc220e3109e256211494111fa76df6832804775001913d40a4844f48341
+size 40369

Task2/evaluation/hard.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:8e1df1b1ef93465b43636811abcb1ba39ee910c132256d0d635a05a5744abf1b
+size 17510

Task2/evaluation/manifest.json

@@ -0,0 +1,28 @@
+{
+  "task": "Task2",
+  "purpose": "Lightweight ground-truth annotations for evaluation; no RGB or visualization payloads.",
+  "columns": [
+    "sample_id",
+    "query_id",
+    "traverse_vector",
+    "cost_vector",
+    "region_vector"
+  ],
+  "splits": {
+    "easy": {
+      "file": "easy.parquet",
+      "num_rows": 7659,
+      "size_bytes": 40369
+    },
+    "medium": {
+      "file": "medium.parquet",
+      "num_rows": 3712,
+      "size_bytes": 33618
+    },
+    "hard": {
+      "file": "hard.parquet",
+      "num_rows": 1604,
+      "size_bytes": 17510
+    }
+  }
+}

Task2/evaluation/medium.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:0d3b286130e739ecb39802b4dc6501cc888c5ac636482862f85a772ea5e0fe69
+size 33618

Task3/evaluation/easy.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:09cd730d8a9fc2f7088ef628997b9456dd794170347d79d3f514cc3183706bde
+size 11065606

Task3/evaluation/hard.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:8decfb223ee705339f3f31e5898b469129d80ed92702796831b0da11a7dc904d
+size 3494475

Task3/evaluation/manifest.json

@@ -0,0 +1,31 @@
+{
+  "task": "Task3",
+  "purpose": "Lightweight ground-truth annotations for evaluation; no RGB or visualization payloads.",
+  "columns": [
+    "sample_id",
+    "query_id",
+    "source_image_name",
+    "traverse_vector",
+    "cost_vector",
+    "start_point",
+    "end_point",
+    "gt_trajectory"
+  ],
+  "splits": {
+    "easy": {
+      "file": "easy.parquet",
+      "num_rows": 6492,
+      "size_bytes": 11065606
+    },
+    "medium": {
+      "file": "medium.parquet",
+      "num_rows": 2705,
+      "size_bytes": 5325749
+    },
+    "hard": {
+      "file": "hard.parquet",
+      "num_rows": 1624,
+      "size_bytes": 3494475
+    }
+  }
+}

Task3/evaluation/medium.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:662f97de4e5f049c07d6e554cfbac8425f031d15e19fac6e918edfb3a4047623
+size 5325749

Task3/labels/labels.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:75183ae496cb73e93d049c0fd8f8c3565ac1f46b18c886e1df4014f8da9eeaae
+size 90175607

Task3/labels/manifest.json

@@ -0,0 +1,20 @@
+{
+  "num_labels": 3109,
+  "file": "labels.parquet",
+  "size_bytes": 90175607,
+  "sha256": "75183ae496cb73e93d049c0fd8f8c3565ac1f46b18c886e1df4014f8da9eeaae",
+  "encoding": "lossless 8-bit PNG stored in a Hugging Face Image feature",
+  "join_key": "source_image_name",
+  "coordinate_format": "label[y, x]",
+  "class_names": {
+    "0": "Empty",
+    "1": "Bareland",
+    "2": "Rangeland",
+    "3": "Developed space",
+    "4": "Road",
+    "5": "Tree",
+    "6": "Water",
+    "7": "Agriculture land",
+    "8": "Building"
+  }
+}