Update README.md

README.md

@@ -28,7 +28,7 @@ This model predicts per‑pixel labels among: **clear**, **thick cloud**, **thin
 
 ## ✨ Highlights
 
-- **Input:** Sentinel‑2 L1C RGB tiles/patches (float32, shape `B×3×H×W`, bands **B04‑B03‑B02**).  
+- **Input:** Sentinel‑2 L1C RGB tiles/patches (float32, shape `B×3×512×512`, bands **B04‑B03‑B02**).  
 - **Backbone:** `mit_b4` (MiT encoder via `segmentation_models_pytorch`).  
 - **Output:** Logits `B×4×H×W` (apply softmax + argmax).  
 - **Files:** `model.py`, `config.json`, and weights.