End of training

README.md

@@ -22,7 +22,7 @@ model-index:
     metrics:
     - name: Wer
       type: wer
-      value: 38.846825859056494
+      value: 28.91705069124424
 ---
 
 <!-- This model card has been generated automatically according to the information the Trainer had access to. You
@@ -32,9 +32,9 @@ should probably proofread and complete it, then remove this comment. -->
 
 This model is a fine-tuned version of [openai/whisper-tiny](https://huggingface.co/openai/whisper-tiny) on the PolyAI/minds14 dataset.
 It achieves the following results on the evaluation set:
-- Loss: 1.0683
-- Wer Ortho: 40.0731
-- Wer: 38.8468
+- Loss: 0.8974
+- Wer Ortho: 30.4322
+- Wer: 28.9171
 
 ## Model description
 
@@ -60,21 +60,14 @@ The following hyperparameters were used during training:
 - 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: 4000
+- training_steps: 500
 - mixed_precision_training: Native AMP
 
 ### Training results
 
-| Training Loss | Epoch  | Step | Validation Loss | Wer Ortho | Wer     |
-|:-------------:|:------:|:----:|:---------------:|:---------:|:-------:|
-| 0.0007        | 17.86  | 500  | 0.8423          | 37.6370   | 35.5853 |
-| 0.0002        | 35.71  | 1000 | 0.9065          | 38.1242   | 36.3425 |
-| 0.0001        | 53.57  | 1500 | 0.9445          | 39.6468   | 38.0897 |
-| 0.0001        | 71.43  | 2000 | 0.9743          | 39.4641   | 38.0897 |
-| 0.0           | 89.29  | 2500 | 1.0029          | 39.8295   | 38.4391 |
-| 0.0           | 107.14 | 3000 | 1.0259          | 39.8904   | 38.5556 |
-| 0.0           | 125.0  | 3500 | 1.0475          | 39.8904   | 38.6721 |
-| 0.0           | 142.86 | 4000 | 1.0683          | 40.0731   | 38.8468 |
+| Training Loss | Epoch | Step | Validation Loss | Wer Ortho | Wer     |
+|:-------------:|:-----:|:----:|:---------------:|:---------:|:-------:|
+| 0.0007        | 17.86 | 500  | 0.8974          | 30.4322   | 28.9171 |
 
 
 ### Framework versions