update model card README.md

README.md

@@ -18,8 +18,8 @@ should probably proofread and complete it, then remove this comment. -->
 
 This model is a fine-tuned version of [bookbot/distil-ast-audioset](https://huggingface.co/bookbot/distil-ast-audioset) on the GTZAN dataset.
 It achieves the following results on the evaluation set:
-- Loss: 0.3803
-- Accuracy: 0.86
+- Loss: 0.5033
+- Accuracy: 0.89
 
 ## Model description
 
@@ -47,17 +47,22 @@ The following hyperparameters were used during training:
 - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
 - lr_scheduler_type: linear
 - lr_scheduler_warmup_ratio: 0.1
-- num_epochs: 5
+- num_epochs: 10
 
 ### Training results
 
 | Training Loss | Epoch | Step | Validation Loss | Accuracy |
 |:-------------:|:-----:|:----:|:---------------:|:--------:|
-| 0.6408        | 1.0   | 112  | 0.8619          | 0.75     |
-| 0.2707        | 2.0   | 225  | 0.5248          | 0.81     |
-| 0.1536        | 3.0   | 337  | 0.6750          | 0.8      |
-| 0.163         | 4.0   | 450  | 0.4019          | 0.89     |
-| 0.0039        | 4.98  | 560  | 0.3803          | 0.86     |
+| 0.7719        | 1.0   | 112  | 1.0881          | 0.65     |
+| 0.3801        | 2.0   | 225  | 0.8942          | 0.7      |
+| 0.3706        | 3.0   | 337  | 0.9499          | 0.75     |
+| 0.3541        | 4.0   | 450  | 0.5243          | 0.87     |
+| 0.0132        | 5.0   | 562  | 0.5716          | 0.81     |
+| 0.0221        | 6.0   | 675  | 0.5164          | 0.87     |
+| 0.0001        | 7.0   | 787  | 0.4789          | 0.91     |
+| 0.0002        | 8.0   | 900  | 0.5062          | 0.87     |
+| 0.0528        | 9.0   | 1012 | 0.5029          | 0.89     |
+| 0.0002        | 9.96  | 1120 | 0.5033          | 0.89     |
 
 
 ### Framework versions