End of training

README.md

@@ -19,8 +19,8 @@ should probably proofread and complete it, then remove this comment. -->
 
 This model is a fine-tuned version of [openai/whisper-base](https://huggingface.co/openai/whisper-base) on the Common Voice 16 dataset.
 It achieves the following results on the evaluation set:
-- Loss: 0.6417
-- Cer: 16.8380
+- Loss: 0.4508
+- Cer: 12.2672
 
 ## Model description
 
@@ -39,25 +39,26 @@ More information needed
 ### Training hyperparameters
 
 The following hyperparameters were used during training:
-- learning_rate: 1e-05
+- learning_rate: 5e-05
 - train_batch_size: 16
 - eval_batch_size: 16
 - seed: 42
 - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
 - lr_scheduler_type: constant_with_warmup
-- lr_scheduler_warmup_steps: 50
-- training_steps: 500
+- lr_scheduler_warmup_steps: 3000
+- training_steps: 3000
 - mixed_precision_training: Native AMP
 
 ### Training results
 
 | Training Loss | Epoch  | Step | Validation Loss | Cer     |
 |:-------------:|:------:|:----:|:---------------:|:-------:|
-| 3.1341        | 0.0991 | 100  | 1.2010          | 26.4369 |
-| 1.3756        | 0.1982 | 200  | 0.8743          | 20.5919 |
-| 1.1401        | 0.2973 | 300  | 0.7722          | 18.2866 |
-| 0.9782        | 0.3964 | 400  | 0.6795          | 16.8536 |
-| 0.8545        | 0.4955 | 500  | 0.6417          | 16.8380 |
+| 2.3843        | 0.4955 | 500  | 0.7760          | 20.1410 |
+| 0.832         | 0.9911 | 1000 | 0.5657          | 16.6787 |
+| 0.4694        | 1.4866 | 1500 | 0.5170          | 14.5543 |
+| 0.3654        | 1.9822 | 2000 | 0.4646          | 12.7373 |
+| 0.2036        | 2.4777 | 2500 | 0.4744          | 12.9181 |
+| 0.1912        | 2.9732 | 3000 | 0.4508          | 12.2672 |
 
 
 ### Framework versions