Model save

README.md

@@ -15,7 +15,7 @@ should probably proofread and complete it, then remove this comment. -->
 
 This model is a fine-tuned version of [EdBianchi/vit-fire-detection](https://huggingface.co/EdBianchi/vit-fire-detection) on the None dataset.
 It achieves the following results on the evaluation set:
-- Loss: 0.5913
+- Loss: 0.7160
 
 ## Model description
 
@@ -40,22 +40,18 @@ The following hyperparameters were used during training:
 - seed: 42
 - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
 - lr_scheduler_type: linear
-- num_epochs: 10
+- lr_scheduler_warmup_ratio: 0.1
+- num_epochs: 5
 
 ### Training results
 
 | Training Loss | Epoch | Step | Validation Loss |
 |:-------------:|:-----:|:----:|:---------------:|
-| No log        | 1.0   | 15   | 0.6945          |
-| No log        | 2.0   | 30   | 0.4927          |
-| No log        | 3.0   | 45   | 0.5516          |
-| No log        | 4.0   | 60   | 0.4700          |
-| No log        | 5.0   | 75   | 0.5426          |
-| No log        | 6.0   | 90   | 0.5149          |
-| No log        | 7.0   | 105  | 0.5401          |
-| No log        | 8.0   | 120  | 0.5532          |
-| No log        | 9.0   | 135  | 0.5662          |
-| No log        | 10.0  | 150  | 0.5913          |
+| No log        | 1.0   | 15   | 0.8514          |
+| No log        | 2.0   | 30   | 0.5119          |
+| No log        | 3.0   | 45   | 0.6616          |
+| No log        | 4.0   | 60   | 0.6998          |
+| No log        | 5.0   | 75   | 0.7160          |
 
 
 ### Framework versions