Model save

README.md

@@ -17,8 +17,8 @@ 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: 0.5781
-- Accuracy: 0.7742
+- Loss: 1.9178
+- Accuracy: 0.3548
 
 ## Model description
 
@@ -44,16 +44,13 @@ 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: 300
+- training_steps: 75
 
 ### Training results
 
 | Training Loss | Epoch | Step | Validation Loss | Accuracy |
 |:-------------:|:-----:|:----:|:---------------:|:--------:|
-| 1.8832        | 0.25  | 75   | 1.5556          | 0.6429   |
-| 0.9517        | 1.25  | 150  | 0.6312          | 0.8      |
-| 0.3181        | 2.25  | 225  | 0.4654          | 0.8143   |
-| 0.1388        | 3.25  | 300  | 0.4881          | 0.8143   |
+| 1.8633        | 1.0   | 75   | 1.8419          | 0.3857   |
 
 
 ### Framework versions

runs/Apr14_12-21-41_5447b4b0ab00/events.out.tfevents.1713097662.5447b4b0ab00.34.1

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:63c4ae6e1f4acb2b168774c0dfd203df1bb0e6142018b8c925bee5b31c94cc9f
+size 405