Upload README.md with huggingface_hub

README.md

@@ -19,14 +19,14 @@ model-index:
       type: cifar100
     metrics:
     - type: accuracy
-      value: 0.7843
+      value: 0.7926
 ---
 
 # Model Card for Model ID
 
 This model is a small resnet18 trained on cifar100.
 
-- **Test Accuracy:** 0.7843
+- **Test Accuracy:** 0.7926
 - **License:** MIT
 
 ## How to Get Started with the Model
@@ -53,9 +53,9 @@ Training data is cifar100.
 
 - **dataset**: `cifar100`
 
-- **batch_size**: `64`
+- **batch_size**: `128`
 
-- **epochs**: `200`
+- **epochs**: `300`
 
 - **validation_frequency**: `5`
 
@@ -73,7 +73,7 @@ Training data is cifar100.
 
 - **scheduler**: `CosineAnnealingLR`
 
-- **scheduler_kwargs**: `{'T_max': 190}`
+- **scheduler_kwargs**: `{'T_max': 280}`
 
 - **debug**: `False`