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 [facebook/wav2vec2-base](https://huggingface.co/facebook/wav2vec2-base) on an unknown dataset.
 It achieves the following results on the evaluation set:
-- Loss: 1.9009
-- Accuracy: 0.1648
+- Loss: 1.1607
+- Accuracy: 0.5375
 
 ## Model description
 
@@ -37,7 +37,7 @@ More information needed
 ### Training hyperparameters
 
 The following hyperparameters were used during training:
-- learning_rate: 0.0005
+- learning_rate: 0.0003
 - train_batch_size: 32
 - eval_batch_size: 32
 - seed: 42
@@ -45,15 +45,15 @@ The following hyperparameters were used during training:
 - total_train_batch_size: 128
 - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
 - lr_scheduler_type: linear
-- lr_scheduler_warmup_ratio: 0.01
+- lr_scheduler_warmup_ratio: 0.1
 - num_epochs: 2
 
 ### Training results
 
 | Training Loss | Epoch | Step | Validation Loss | Accuracy |
 |:-------------:|:-----:|:----:|:---------------:|:--------:|
-| 1.8998        | 1.0   | 80   | 1.9038          | 0.1648   |
-| 1.8941        | 2.0   | 160  | 1.9009          | 0.1648   |
+| 1.566         | 1.0   | 80   | 1.2818          | 0.5039   |
+| 1.2082        | 2.0   | 160  | 1.1607          | 0.5375   |
 
 
 ### Framework versions