Model save

README.md

@@ -23,7 +23,7 @@ model-index:
     metrics:
     - name: Accuracy
       type: accuracy
-      value: 0.996
+      value: 1.0
 ---
 
 <!-- This model card has been generated automatically according to the information the Trainer had access to. You
@@ -33,8 +33,8 @@ should probably proofread and complete it, then remove this comment. -->
 
 This model is a fine-tuned version of [Marqo/nsfw-image-detection-384](https://huggingface.co/Marqo/nsfw-image-detection-384) on the imagefolder dataset.
 It achieves the following results on the evaluation set:
-- Loss: 0.0254
-- Accuracy: 0.996
+- Loss: 0.0003
+- Accuracy: 1.0
 
 ## Model description
 
@@ -53,23 +53,21 @@ More information needed
 ### Training hyperparameters
 
 The following hyperparameters were used during training:
-- learning_rate: 5e-05
+- learning_rate: 2e-05
 - train_batch_size: 16
-- eval_batch_size: 8
+- eval_batch_size: 16
 - seed: 42
 - optimizer: Use OptimizerNames.ADAMW_TORCH with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
-- lr_scheduler_type: linear
-- num_epochs: 3
+- lr_scheduler_type: cosine
+- lr_scheduler_warmup_ratio: 0.1
+- num_epochs: 1.5
 
 ### Training results
 
 | Training Loss | Epoch  | Step | Validation Loss | Accuracy |
 |:-------------:|:------:|:----:|:---------------:|:--------:|
-| 0.0375        | 0.5066 | 500  | 0.0067          | 0.998    |
-| 0.0138        | 1.0132 | 1000 | 0.0001          | 1.0      |
-| 0.0028        | 1.5198 | 1500 | 0.0001          | 1.0      |
-| 0.002         | 2.0263 | 2000 | 0.0000          | 1.0      |
-| 0.0013        | 2.5329 | 2500 | 0.0000          | 1.0      |
+| 0.0041        | 0.5066 | 500  | 0.0147          | 0.996    |
+| 0.0023        | 1.0132 | 1000 | 0.0003          | 1.0      |
 
 
 ### Framework versions