Add Model documentation

README.md

@@ -21,7 +21,7 @@ model-index:
       type: mnist
     metrics:
     - type: accuracy
-      value: 99.16
+      value: 98.96
       name: Test Accuracy
 ---
 
@@ -57,8 +57,8 @@ Flatten -> FC(512) -> Dropout -> FC(256) -> Dropout -> FC(10)
 
 ## Performance
 
-- **Test Accuracy**: 99.16%
-- **Test Loss**: 0.0320
+- **Test Accuracy**: 98.96%
+- **Test Loss**: 0.0341
 - **Parameters**: 817,354
 
 ## Training Configuration