update model card README.md

README.md

@@ -18,8 +18,8 @@ should probably proofread and complete it, then remove this comment. -->
 
 This model is a fine-tuned version of [facebook/wav2vec2-xls-r-300m](https://huggingface.co/facebook/wav2vec2-xls-r-300m) on the MULTILINGUAL_LIBRISPEECH - GERMAN dataset.
 It achieves the following results on the evaluation set:
-- Loss: 0.3893
-- Wer: 0.2695
+- Loss: 0.2157
+- Wer: 0.1562
 
 ## Model description
 
@@ -38,22 +38,33 @@ More information needed
 ### Training hyperparameters
 
 The following hyperparameters were used during training:
-- learning_rate: 0.0005
+- learning_rate: 0.0001
 - train_batch_size: 32
 - eval_batch_size: 8
 - seed: 42
 - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
 - lr_scheduler_type: linear
-- lr_scheduler_warmup_steps: 500
-- num_epochs: 15.0
+- lr_scheduler_warmup_steps: 1000
+- num_epochs: 100.0
 - mixed_precision_training: Native AMP
 
 ### Training results
 
 | Training Loss | Epoch | Step | Validation Loss | Wer    |
 |:-------------:|:-----:|:----:|:---------------:|:------:|
-| 0.265         | 7.25  | 500  | 0.4818          | 0.4440 |
-| 0.0522        | 14.49 | 1000 | 0.3902          | 0.2725 |
+| 3.0132        | 7.25  | 500  | 2.9393          | 1.0    |
+| 2.9241        | 14.49 | 1000 | 2.8734          | 1.0    |
+| 1.0766        | 21.74 | 1500 | 0.2773          | 0.2488 |
+| 0.8416        | 28.99 | 2000 | 0.2224          | 0.1990 |
+| 0.8048        | 36.23 | 2500 | 0.2063          | 0.1792 |
+| 0.7664        | 43.48 | 3000 | 0.2088          | 0.1748 |
+| 0.6571        | 50.72 | 3500 | 0.2042          | 0.1668 |
+| 0.7014        | 57.97 | 4000 | 0.2136          | 0.1649 |
+| 0.6171        | 65.22 | 4500 | 0.2139          | 0.1641 |
+| 0.6609        | 72.46 | 5000 | 0.2144          | 0.1621 |
+| 0.6318        | 79.71 | 5500 | 0.2129          | 0.1600 |
+| 0.6222        | 86.96 | 6000 | 0.2124          | 0.1582 |
+| 0.588         | 94.2  | 6500 | 0.2143          | 0.1560 |
 
 
 ### Framework versions