update model card README.md

README.md

@@ -14,7 +14,10 @@ should probably proofread and complete it, then remove this comment. -->
 
 # wav2vec2-base-gn-demo
 
-This model is a fine-tuned version of [facebook/wav2vec2-xls-r-300m](https://huggingface.co/facebook/wav2vec2-xls-r-300m) on the common_voice dataset.
+This model is a fine-tuned version of [azuur/wav2vec2-base-gn-demo](https://huggingface.co/azuur/wav2vec2-base-gn-demo) on the common_voice dataset.
+It achieves the following results on the evaluation set:
+- Loss: 0.6992
+- Wer: 0.7439
 
 ## Model description
 
@@ -34,7 +37,7 @@ More information needed
 
 The following hyperparameters were used during training:
 - learning_rate: 0.0002
-- train_batch_size: 32
+- train_batch_size: 16
 - eval_batch_size: 8
 - seed: 42
 - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
@@ -43,6 +46,19 @@ The following hyperparameters were used during training:
 - num_epochs: 30
 - mixed_precision_training: Native AMP
 
+### Training results
+
+| Training Loss | Epoch | Step | Validation Loss | Wer    |
+|:-------------:|:-----:|:----:|:---------------:|:------:|
+| No log        | 4.0   | 100  | 2.4470          | 1.0015 |
+| No log        | 8.0   | 200  | 0.9011          | 0.8720 |
+| No log        | 12.0  | 300  | 0.8527          | 0.9177 |
+| No log        | 16.0  | 400  | 0.7743          | 0.7988 |
+| 1.0068        | 20.0  | 500  | 0.7984          | 0.8049 |
+| 1.0068        | 24.0  | 600  | 0.7133          | 0.7530 |
+| 1.0068        | 28.0  | 700  | 0.6992          | 0.7439 |
+
+
 ### Framework versions
 
 - Transformers 4.11.3