End of training

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-960h](https://huggingface.co/facebook/wav2vec2-base-960h) on the None dataset.
 It achieves the following results on the evaluation set:
-- Loss: 6.3326
-- Wer: 1.7738
+- Loss: 4.7884
+- Wer: 1.0
 
 ## Model description
 
@@ -38,7 +38,7 @@ More information needed
 
 The following hyperparameters were used during training:
 - learning_rate: 0.0001
-- train_batch_size: 4
+- train_batch_size: 8
 - eval_batch_size: 8
 - seed: 42
 - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
@@ -51,10 +51,20 @@ The following hyperparameters were used during training:
 
 | Training Loss | Epoch | Step | Validation Loss | Wer    |
 |:-------------:|:-----:|:----:|:---------------:|:------:|
-| 7.8979        | 5.0   | 500  | 9.4008          | 3.8311 |
-| 6.9854        | 10.0  | 1000 | 7.7231          | 3.0309 |
-| 6.0643        | 15.0  | 1500 | 6.6362          | 2.0578 |
-| 5.6488        | 20.0  | 2000 | 6.3326          | 1.7738 |
+| 8.6131        | 1.43  | 500  | 10.3380         | 1.0    |
+| 7.3718        | 2.86  | 1000 | 8.0143          | 1.0    |
+| 6.0492        | 4.29  | 1500 | 6.5868          | 1.0198 |
+| 5.3856        | 5.71  | 2000 | 5.9299          | 1.0209 |
+| 4.9947        | 7.14  | 2500 | 5.4516          | 1.0083 |
+| 4.6977        | 8.57  | 3000 | 5.1176          | 1.0015 |
+| 4.5413        | 10.0  | 3500 | 4.9814          | 1.0002 |
+| 4.491         | 11.43 | 4000 | 4.9081          | 0.9999 |
+| 4.4496        | 12.86 | 4500 | 4.8644          | 0.9999 |
+| 4.4016        | 14.29 | 5000 | 4.8368          | 0.9999 |
+| 4.3977        | 15.71 | 5500 | 4.8137          | 1.0    |
+| 4.4179        | 17.14 | 6000 | 4.8038          | 1.0    |
+| 4.3831        | 18.57 | 6500 | 4.7890          | 1.0    |
+| 4.3699        | 20.0  | 7000 | 4.7884          | 1.0    |
 
 
 ### Framework versions