Update README.md

README.md

@@ -144,7 +144,7 @@ blah blah do this later
 ![Confusion Matrix](confusion_matrix_normalized.png)
 
 #### F1 Confidence Curve
-![F1 Curve](F1_curve.png)
+![BoxF1 Curve](BoxF1_curve.png)
 
 #### Training & Validation Loss Curves
 ![Results](results.png)
@@ -162,7 +162,7 @@ is partially limited by sample size.
 
 ## Limitations and Biases
 
-When tested on the **D2S dataset** (wild images), 
+When tested on the external [**D2S Dataset**](https://www.mvtec.com/company/research/datasets/mvtec-d2s) (wild images), 
 performance dropped significantly. The model missed entire 
 objects, produced low-confidence detections, and misclassified items.
 For example, it labeled a water bottle as `instant_noodles`. This