End of training

README.md

@@ -18,8 +18,8 @@ should probably proofread and complete it, then remove this comment. -->
 
 This model is a fine-tuned version of [MCG-NJU/videomae-base](https://huggingface.co/MCG-NJU/videomae-base) on an unknown dataset.
 It achieves the following results on the evaluation set:
-- Loss: 2.2271
-- Accuracy: 0.25
+- Loss: 0.3808
+- Accuracy: 0.8857
 
 ## Model description
 
@@ -45,16 +45,16 @@ 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
-- training_steps: 40
+- training_steps: 1200
 
 ### Training results
 
 | Training Loss | Epoch | Step | Validation Loss | Accuracy |
 |:-------------:|:-----:|:----:|:---------------:|:--------:|
-| 2.1398        | 0.25  | 10   | 2.3374          | 0.0357   |
-| 1.8581        | 1.25  | 20   | 2.2342          | 0.1429   |
-| 1.9823        | 2.25  | 30   | 2.2072          | 0.2857   |
-| 1.5555        | 3.25  | 40   | 2.2271          | 0.25     |
+| 1.5468        | 0.25  | 300  | 1.2242          | 0.4714   |
+| 0.9665        | 1.25  | 600  | 0.7851          | 0.7286   |
+| 0.5486        | 2.25  | 900  | 1.0866          | 0.7857   |
+| 1.2046        | 3.25  | 1200 | 0.3808          | 0.8857   |
 
 
 ### Framework versions