asmaa1
/

videomae-base-finetuned-SLT-subset

@@ -17,7 +17,7 @@ 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: 0.2692
 - Accuracy: 1.0
 ## Model description
@@ -44,28 +44,28 @@ 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: 928
 ### Training results
 | Training Loss | Epoch | Step | Validation Loss | Accuracy |
 |:-------------:|:-----:|:----:|:---------------:|:--------:|
-| 3.9733        | 0.06  | 59   | 3.6798          | 0.025    |
-| 3.9747        | 1.06  | 118  | 3.6314          | 0.1      |
-| 3.6534        | 2.06  | 177  | 3.3504          | 0.05     |
-| 3.6011        | 3.06  | 236  | 3.2174          | 0.1      |
-| 3.5557        | 4.06  | 295  | 3.0762          | 0.125    |
-| 3.4256        | 5.06  | 354  | 3.0103          | 0.125    |
-| 3.2026        | 6.06  | 413  | 2.9119          | 0.3      |
-| 2.9769        | 7.06  | 472  | 2.6809          | 0.4      |
-| 2.7325        | 8.06  | 531  | 2.1823          | 0.65     |
-| 2.271         | 9.06  | 590  | 1.6918          | 0.775    |
-| 1.9317        | 10.06 | 649  | 1.0511          | 0.925    |
-| 1.6471        | 11.06 | 708  | 0.6689          | 0.975    |
-| 0.8839        | 12.06 | 767  | 0.4407          | 1.0      |
-| 0.772         | 13.06 | 826  | 0.3356          | 1.0      |
-| 0.6227        | 14.06 | 885  | 0.2782          | 1.0      |
-| 0.5521        | 15.05 | 928  | 0.2692          | 1.0      |
 ### Framework versions

 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: 0.3299
 - Accuracy: 1.0
 ## Model description
 - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
 - lr_scheduler_type: linear
 - lr_scheduler_warmup_ratio: 0.1
+- training_steps: 960
 ### Training results
 | Training Loss | Epoch | Step | Validation Loss | Accuracy |
 |:-------------:|:-----:|:----:|:---------------:|:--------:|
+| 3.9613        | 0.06  | 60   | 3.7190          | 0.05     |
+| 4.0386        | 1.06  | 120  | 3.6877          | 0.05     |
+| 3.7601        | 2.06  | 180  | 3.6471          | 0.05     |
+| 3.7248        | 3.06  | 240  | 3.5456          | 0.05     |
+| 3.8879        | 4.06  | 300  | 3.5123          | 0.05     |
+| 3.7959        | 5.06  | 360  | 3.4910          | 0.075    |
+| 3.5278        | 6.06  | 420  | 3.4499          | 0.175    |
+| 3.4031        | 7.06  | 480  | 3.3234          | 0.35     |
+| 3.3702        | 8.06  | 540  | 2.9582          | 0.475    |
+| 2.9523        | 9.06  | 600  | 2.1683          | 0.8      |
+| 2.0989        | 10.06 | 660  | 1.5218          | 0.85     |
+| 1.5045        | 11.06 | 720  | 0.9658          | 0.975    |
+| 0.9524        | 12.06 | 780  | 0.6727          | 1.0      |
+| 0.6987        | 13.06 | 840  | 0.4638          | 1.0      |
+| 0.4222        | 14.06 | 900  | 0.3591          | 1.0      |
+| 0.366         | 15.06 | 960  | 0.3299          | 1.0      |
 ### Framework versions

pytorch_model.bin CHANGED Viewed

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:93345d869566ed0c2155e3b040f93865992a2850bf7722f2a465afa2ebd9a067
 size 345095797

 version https://git-lfs.github.com/spec/v1
+oid sha256:01f4c4cc142e8707c024a4b63b4f7df250cccbff78c5137960258debd27ef6f8
 size 345095797