End of training

README.md

@@ -1,8 +1,8 @@
 ---
 license: apache-2.0
+base_model: facebook/wav2vec2-xls-r-300m
 tags:
 - generated_from_trainer
-base_model: facebook/wav2vec2-xls-r-300m
 datasets:
 - make_dataset
 model-index:
@@ -17,8 +17,8 @@ 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 make_dataset dataset.
 It achieves the following results on the evaluation set:
-- Loss: 4.6886
-- Cer: 1.0
+- Loss: 6.4903
+- Cer: 0.8807
 
 ## Model description
 
@@ -38,20 +38,21 @@ More information needed
 
 The following hyperparameters were used during training:
 - learning_rate: 0.0001
-- train_batch_size: 16
+- train_batch_size: 64
 - eval_batch_size: 16
 - seed: 42
 - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
 - lr_scheduler_type: linear
 - lr_scheduler_warmup_steps: 1000
-- num_epochs: 500
+- num_epochs: 1000
 - mixed_precision_training: Native AMP
 
 ### Training results
 
-| Training Loss | Epoch | Step | Validation Loss | Cer |
-|:-------------:|:-----:|:----:|:---------------:|:---:|
-| 17.7096       | 500.0 | 500  | 4.6886          | 1.0 |
+| Training Loss | Epoch  | Step | Validation Loss | Cer    |
+|:-------------:|:------:|:----:|:---------------:|:------:|
+| 18.8342       | 500.0  | 500  | 4.7408          | 1.0    |
+| 0.6325        | 1000.0 | 1000 | 6.4903          | 0.8807 |
 
 
 ### Framework versions