End of training

README.md

@@ -6,19 +6,19 @@ tags:
 metrics:
 - wer
 model-index:
-- name: small-Cotrsc
+- name: opus-base
   results: []
 ---
 
 <!-- This model card has been generated automatically according to the information the Trainer had access to. You
 should probably proofread and complete it, then remove this comment. -->
 
-# small-Cotrsc
+# opus-base
 
 This model is a fine-tuned version of [openai/whisper-base](https://huggingface.co/openai/whisper-base) on an unknown dataset.
 It achieves the following results on the evaluation set:
-- Loss: 0.0171
-- Wer: 20.5156
+- Loss: 0.0362
+- Wer: 24.9734
 
 ## Model description
 
@@ -37,24 +37,24 @@ More information needed
 ### Training hyperparameters
 
 The following hyperparameters were used during training:
-- learning_rate: 1e-05
-- train_batch_size: 12
+- learning_rate: 2e-05
+- train_batch_size: 24
 - 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: 320
-- training_steps: 3200
+- lr_scheduler_warmup_steps: 500
+- training_steps: 2000
 - mixed_precision_training: Native AMP
 
 ### Training results
 
 | Training Loss | Epoch  | Step | Validation Loss | Wer     |
 |:-------------:|:------:|:----:|:---------------:|:-------:|
-| 0.0204        | 0.3268 | 800  | 0.0194          | 22.4152 |
-| 0.0205        | 0.6536 | 1600 | 0.0185          | 22.0081 |
-| 0.021         | 0.9804 | 2400 | 0.0173          | 20.7870 |
-| 0.0142        | 1.3072 | 3200 | 0.0171          | 20.5156 |
+| 0.2172        | 0.4085 | 500  | 0.0792          | 47.6621 |
+| 0.0533        | 0.8170 | 1000 | 0.0468          | 31.5887 |
+| 0.0372        | 1.2255 | 1500 | 0.0392          | 26.6472 |
+| 0.0343        | 1.6340 | 2000 | 0.0362          | 24.9734 |
 
 
 ### Framework versions