Drop ultra from object detection results

LICENSE

@@ -0,0 +1,21 @@
+MIT License
+
+Copyright (c) 2024 GRAID Project
+
+Permission is hereby granted, free of charge, to any person obtaining a copy
+of this software and associated documentation files (the "Software"), to deal
+in the Software without restriction, including without limitation the rights
+to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
+copies of the Software, and to permit persons to whom the Software is
+furnished to do so, subject to the following conditions:
+
+The above copyright notice and this permission notice shall be included in all
+copies or substantial portions of the Software.
+
+THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
+AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
+LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
+OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
+SOFTWARE.

LICENSE.md

@@ -0,0 +1,70 @@
+# License
+
+## Creative Commons Attribution-NonCommercial 4.0 International (CC BY-NC 4.0)
+
+GRAID-generated datasets are licensed under the **Creative Commons Attribution-NonCommercial 4.0 International License**.
+
+**You are free to:**
+- **Share** — copy and redistribute the material in any medium or format
+- **Adapt** — remix, transform, and build upon the material
+
+**Under the following terms:**
+- **Attribution** — You must give appropriate credit, provide a link to the license, and indicate if changes were made
+- **NonCommercial** — You may not use the material for commercial purposes
+
+**Full License Text:** https://creativecommons.org/licenses/by-nc/4.0/
+
+---
+
+## Commercial Licensing for GRAID Datasets
+
+### For Commercial Use
+
+**Commercial entities** (including companies, startups, and for-profit organizations) that wish to use GRAID-generated datasets for commercial purposes **must obtain a paid license from MESH**.
+
+The CC BY-NC 4.0 license prohibits commercial use. For commercial licensing:
+
+**Contact:** Karim Elmaaroufi  
+**Company:** MESH  
+**Purpose:** Commercial licensing inquiries
+
+Commercial licensing includes:
+- Permission to use datasets for commercial products and services
+- Permission to create commercial derivative works  
+- Support and legal protections for commercial use cases
+
+### Academic and Non-Commercial Use
+
+✅ **Free for:**
+- Academic research and education
+- University coursework and teaching
+- Non-profit research organizations
+- Personal educational projects
+- Open source research (non-commercial)
+
+---
+
+## Source Dataset Compliance
+
+**Important:** GRAID-generated datasets are derived from various source datasets (BDD100K, Waymo, nuImages, etc.).
+
+**You must comply with BOTH:**
+1. The CC BY-NC 4.0 terms for GRAID-generated questions and metadata
+2. The original source dataset license terms for underlying images and annotations
+
+Each dataset's README contains specific source dataset license requirements.
+
+---
+
+## Attribution
+
+When using GRAID datasets, please cite:
+- The GRAID framework (see dataset README for citation format)
+- The original source dataset
+- This licensing information for clarity on commercial use
+
+---
+
+## Questions?
+
+For licensing questions or commercial inquiries, contact the GRAID team through the project repository or reach out to MESH directly for commercial licensing.

LICENSE_MESH_ACADEMIC.md

@@ -0,0 +1,52 @@
+Creative Commons Attribution-NonCommercial 4.0 International Public License
+==============================================================================
+
+GRAID-Generated Datasets Commercial Licensing Policy
+----------------------------------------------------
+
+The GRAID-generated datasets are licensed under **Creative Commons Attribution-NonCommercial 4.0 International (CC BY-NC 4.0)**.
+
+This license permits:
+- ✅ **Academic and University Use**: Free use for educational, research, and scholarly purposes
+- ✅ **Non-Commercial Use**: Free use for personal, educational, and non-profit purposes
+- ❌ **Commercial Use**: Prohibited without explicit commercial license from MESH
+
+For the full CC BY-NC 4.0 license text, see: https://creativecommons.org/licenses/by-nc/4.0/
+
+Commercial Licensing
+-------------------
+
+**Commercial entities** (including companies, startups, and for-profit organizations) that wish to use GRAID-generated datasets for commercial purposes must obtain a paid license from **MESH**.
+
+**To request a commercial license**, contact:
+- **Karim Elmaaroufi**
+- Email: Available upon request through repository contacts
+
+Commercial licensing includes:
+- Permission to use datasets for commercial products and services
+- Permission to create commercial derivative works
+- Support and legal protections for commercial use cases
+
+Original Source Dataset Compliance
+----------------------------------
+
+GRAID-generated datasets are derived from various source datasets (BDD100K, Waymo, nuImages, etc.). 
+
+**Important**: You must comply with both:
+1. The CC BY-NC 4.0 terms for the GRAID-generated questions and metadata
+2. The original source dataset license terms for the underlying images and annotations
+
+Each dataset's README contains specific information about source dataset license requirements.
+
+Attribution
+----------
+
+When using these datasets, please cite:
+- The GRAID framework
+- The original source dataset
+- This license for commercial use clarification
+
+Contact
+-------
+
+For questions about commercial licensing or compliance, contact the GRAID team through the project repository.

bdd_depth_config.json

@@ -0,0 +1,45 @@
+{
+  "dataset_name": "bdd",
+  "split": "train+val",
+  "models": [],
+  "use_wbf": false,
+  "confidence_threshold": 0.0,
+  "batch_size": 16,
+  "device": null,
+  "allowable_set": null,
+  "num_workers": 4,
+  "qa_workers": 4,
+  "num_samples": 0,
+  "question_configs": [
+    {
+      "name": "FrontOf",
+      "params": {
+        "margin_ratio": 0.3
+      }
+    },
+    {
+      "name": "BehindOf",
+      "params": {
+        "margin_ratio": 0.3
+      }
+    },
+    {
+      "name": "DepthRanking",
+      "params": {
+        "k": 3,
+        "margin_ratio": 0.3
+      }
+    },
+    {
+      "name": "DepthRanking",
+      "params": {
+        "k": 2,
+        "margin_ratio": 0.3
+      }
+    }
+  ],
+  "save_path": "./bdd_train_gt_dataset_depth",
+  "upload_to_hub": true,
+  "hub_repo_id": "kd7/graid-bdd100k-depth",
+  "hub_private": false
+}

bdd_train_gt_dataset_config.json

@@ -4,11 +4,11 @@
   "models": [],
   "use_wbf": false,
   "confidence_threshold": 0.0,
-  "batch_size": 128,
+  "batch_size": 16,
   "device": null,
   "allowable_set": null,
-  "num_workers": 16,
-  "qa_workers": 32,
+  "num_workers": 8,
+  "qa_workers": 16,
   "num_samples": 0,
   "question_configs": [
     {
@@ -149,10 +149,36 @@
     {
       "name": "MultiChoiceHowMany",
       "params": {}
+    },
+    {
+      "name": "Closer",
+      "params": {
+        "margin_ratio": 0.35
+      }
+    },
+    {
+      "name": "Farther",
+      "params": {
+        "margin_ratio": 0.35
+      }
+    },
+    {
+      "name": "DepthRanking",
+      "params": {
+        "k": 3,
+        "margin_ratio": 0.35
+      }
+    },
+    {
+      "name": "DepthRanking",
+      "params": {
+        "k": 2,
+        "margin_ratio": 0.35
+      }
     }
   ],
   "save_path": "./bdd_train_gt_dataset",
   "upload_to_hub": true,
-  "hub_repo_id": "kd7/graid-bdd100k-ground-truth",
+  "hub_repo_id": "kd7/graid-bdd",
   "hub_private": false
 } 

graid/src/graid/cli/__init__.py

@@ -5,21 +5,27 @@
 """
 
 from .config_manager import ConfigurationManager
-from .validators import ArgumentValidator
 from .error_handler import ErrorHandler
 from .exceptions import (
-    CLIError, ValidationError, DatasetValidationError, COCOValidationError,
-    SplitValidationError, ConfigurationError, ProcessingError, UploadError
+    CLIError,
+    COCOValidationError,
+    ConfigurationError,
+    DatasetValidationError,
+    ProcessingError,
+    SplitValidationError,
+    UploadError,
+    ValidationError,
 )
+from .validators import ArgumentValidator
 
 __all__ = [
     "ConfigurationManager",
-    "ArgumentValidator", 
+    "ArgumentValidator",
     "ErrorHandler",
     "CLIError",
     "ValidationError",
     "DatasetValidationError",
-    "COCOValidationError", 
+    "COCOValidationError",
     "SplitValidationError",
     "ConfigurationError",
     "ProcessingError",