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: 0.6185
-- Accuracy: 0.8516
+- Loss: 1.8301
+- Accuracy: 0.3677
 
 ## Model description
 
@@ -44,18 +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: 450
+- training_steps: 75
 
 ### Training results
 
 | Training Loss | Epoch | Step | Validation Loss | Accuracy |
 |:-------------:|:-----:|:----:|:---------------:|:--------:|
-| 1.9248        | 0.17  | 75   | 1.6910          | 0.3143   |
-| 1.1184        | 1.17  | 150  | 0.8519          | 0.7      |
-| 0.3505        | 2.17  | 225  | 0.5482          | 0.7857   |
-| 0.2939        | 3.17  | 300  | 0.5075          | 0.8286   |
-| 0.0447        | 4.17  | 375  | 0.1741          | 0.9714   |
-| 0.0643        | 5.17  | 450  | 0.1370          | 0.9571   |
+| 1.886         | 1.0   | 75   | 1.6964          | 0.3429   |
 
 
 ### Framework versions

runs/May24_13-42-29_4b27c78cc6c7/events.out.tfevents.1716558625.4b27c78cc6c7.35.1

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:96ab18f982b7f1127ab61a7d1f6dedb0ae1963a5393f6bbe96e8563858bcbd9b
+size 722