Update README.md

README.md

@@ -1,22 +1,20 @@
+
 ---
-title: Emotion Detection (PyTorch)
-emoji: 😄
-colorFrom: blue
-colorTo: purple
-sdk: gradio
-sdk_version: 5.49.1
-app_file: app.py
+title: "Emotion Detection (ViT-Stable)"
+emoji: "😄"
+colorFrom: "blue"
+colorTo: "purple"
+sdk: "gradio"
+sdk_version: "5"
+app_file: "app.py"
 pinned: false
 ---
 
-# PyTorch Facial Emotion Detection (Final Optimized Version)
-
-This Space uses the **most stable PyTorch/Transformers dependency versions** and **memory-safe loading techniques** to maximize the chance of a successful build on a standard CPU tier.
-
-If the app still shows a loading error after a **Factory Reboot**, it indicates a definitive **Out-of-Memory (OOM) error** and requires upgrading the Space's hardware tier (e.g., to a 16GB CPU or T4 GPU).
+# PyTorch Facial Emotion Detection (ViT Stable)
 
-## ⚠️ Important Final Step
+This version uses the **abhilash88/face-emotion-detection** ViT model, which is generally more stable and less resource-intensive to load than the previous checkpoint on standard CPU environments.
 
-1.  Replace all four files with the contents above.
-2.  Go to the **"Settings"** tab of your Space.
-3.  Click **"Factory reboot"** to clear the corrupted build cache.
+**Final Deployment Steps:**
+1. **Replace** all four files with the contents above.
+2. Go to the Space **Settings** tab.
+3. Click **Factory reboot**.