End of training

README.md

@@ -26,7 +26,7 @@ model-index:
     metrics:
     - name: Wer
       type: wer
-      value: 20.773418434390837
+      value: 20.45616669795382
 ---
 
 <!-- This model card has been generated automatically according to the information the Trainer had access to. You
@@ -36,8 +36,8 @@ should probably proofread and complete it, then remove this comment. -->
 
 This model is a fine-tuned version of [openai/whisper-small](https://huggingface.co/openai/whisper-small) on the Common Voice 11.0 dataset.
 It achieves the following results on the evaluation set:
-- Loss: 0.3144
-- Wer: 20.7734
+- Loss: 0.2601
+- Wer: 20.4562
 
 ## Model description
 
@@ -56,46 +56,29 @@ More information needed
 ### Training hyperparameters
 
 The following hyperparameters were used during training:
-- learning_rate: 5e-07
+- learning_rate: 1e-05
 - train_batch_size: 2
 - eval_batch_size: 8
 - seed: 42
-- gradient_accumulation_steps: 8
-- total_train_batch_size: 16
+- gradient_accumulation_steps: 16
+- total_train_batch_size: 32
 - optimizer: Use OptimizerNames.ADAMW_TORCH with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
 - lr_scheduler_type: linear
 - lr_scheduler_warmup_steps: 500
-- training_steps: 10000
+- training_steps: 5000
 - mixed_precision_training: Native AMP
 
 ### Training results
 
-| Training Loss | Epoch  | Step  | Validation Loss | Wer     |
-|:-------------:|:------:|:-----:|:---------------:|:-------:|
-| 0.5507        | 0.2079 | 500   | 0.3695          | 29.2247 |
-| 0.2802        | 0.4158 | 1000  | 0.3148          | 26.7299 |
-| 0.2408        | 0.6236 | 1500  | 0.2970          | 24.2538 |
-| 0.2208        | 0.8315 | 2000  | 0.2728          | 23.3020 |
-| 0.1811        | 1.0394 | 2500  | 0.2665          | 22.3935 |
-| 0.1096        | 1.2473 | 3000  | 0.2641          | 21.8998 |
-| 0.1068        | 1.4552 | 3500  | 0.2568          | 21.6125 |
-| 0.1042        | 1.6630 | 4000  | 0.2516          | 21.0512 |
-| 0.1001        | 1.8709 | 4500  | 0.2472          | 20.4092 |
-| 0.0827        | 2.0788 | 5000  | 0.2469          | 20.3848 |
-| 0.0672        | 2.2869 | 5500  | 0.2665          | 21.1357 |
-| 0.0673        | 2.4948 | 6000  | 0.2674          | 21.5093 |
-| 0.0681        | 2.7026 | 6500  | 0.2635          | 20.6101 |
-| 0.0661        | 2.9105 | 7000  | 0.2602          | 20.5069 |
-| 0.0494        | 3.1184 | 7500  | 0.2708          | 20.5444 |
-| 0.0352        | 3.3263 | 8000  | 0.2688          | 20.5181 |
-| 0.0338        | 3.5341 | 8500  | 0.2717          | 20.2515 |
-| 0.0318        | 3.7420 | 9000  | 0.2723          | 20.2403 |
-| 0.0309        | 3.9499 | 9500  | 0.2711          | 20.1727 |
-| 0.022         | 4.1578 | 10000 | 0.2758          | 20.1577 |
-| 0.0229        | 8.7351 | 10500 | 0.2930          | 21.1019 |
-| 0.0217        | 9.1508 | 11000 | 0.3086          | 20.9874 |
-| 0.0168        | 9.5666 | 11500 | 0.3026          | 20.7847 |
-| 0.0162        | 9.9823 | 12000 | 0.3144          | 20.7734 |
+| Training Loss | Epoch  | Step | Validation Loss | Wer     |
+|:-------------:|:------:|:----:|:---------------:|:-------:|
+| 0.5279        | 0.4158 | 500  | 0.3311          | 27.6591 |
+| 0.2513        | 0.8316 | 1000 | 0.2866          | 24.5504 |
+| 0.1673        | 1.2478 | 1500 | 0.2735          | 22.8928 |
+| 0.1324        | 1.6635 | 2000 | 0.2645          | 21.8153 |
+| 0.1138        | 2.0797 | 2500 | 0.2613          | 21.3816 |
+| 0.064         | 2.4955 | 3000 | 0.2651          | 21.0006 |
+| 0.0615        | 2.9113 | 3500 | 0.2601          | 20.4562 |
 
 
 ### Framework versions