Update README.md

README.md

@@ -11,12 +11,44 @@ tags:
 - Transformers
 - rfdetr
 - supervision
-library_name: transformers
-metrics:
-- mAP@50:95
+model-index:
+- name: BasiliskSeg # Or another unique name you prefer
+  results:
+  - task:
+      type: image-segmentation
+      name: Instance Segmentation
+    metrics:
+    - type: coco
+      value: 0.9668 # Rounded from 0.9667733799769708
+      name: Mask mAP @ IoU=0.50:0.95 | area=all | maxDets=100
+      config: segm
+      args:
+        iouThr: '.50:.05:.95'
+        areaRng: 'all'
+        maxDets: 100
+    # Mask mAP at easy threshold
+    - type: coco
+      value: 0.9783 # Rounded from 0.9782951023893298
+      name: Mask mAP @ IoU=0.50 | area=all | maxDets=100
+      config: segm
+      args:
+        iouThr: '.50'
+        areaRng: 'all'
+        maxDets: 100
+    # Mask Average Recall
+    - type: coco # COCO also reports AR
+      value: 0.9871 # Rounded from 0.9871073298429319
+      name: Mask AR @ IoU=0.50:0.95 | area=all | maxDets=100
+      config: segm # Specify segmentation for AR calculation context
+      args:
+        iouThr: '.50:.05:.95'
+        areaRng: 'all'
+        maxDets: 100
+    source:
+      name: Self-reported via Colab Training
+      url: https://colab.research.google.com/github/roboflow-ai/notebooks/blob/main/notebooks/how-to-finetune-rf-detr-on-detection-dataset.ipynb # Link to the training script/notebook
 ---
 
-
 # Segment-Tulsi(Basil) with Transformers(RF-DETR)
 
 | **Model**              | **Best EMA Mask mAP (@.50:.95)** |