End of training

README.md

@@ -25,7 +25,7 @@ model-index:
     metrics:
     - name: Wer
       type: wer
-      value: 100.0
+      value: 134.83959527973963
 ---
 
 <!-- This model card has been generated automatically according to the information the Trainer had access to. You
@@ -35,8 +35,8 @@ 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 Common Voice dataset.
 It achieves the following results on the evaluation set:
-- Loss: 7.6994
-- Wer: 100.0
+- Loss: 5.0988
+- Wer: 134.8396
 
 ## Model description
 
@@ -56,22 +56,26 @@ More information needed
 
 The following hyperparameters were used during training:
 - learning_rate: 0.003
-- train_batch_size: 8
+- train_batch_size: 16
 - eval_batch_size: 8
 - seed: 42
-- gradient_accumulation_steps: 8
+- gradient_accumulation_steps: 4
 - total_train_batch_size: 64
 - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
 - lr_scheduler_type: linear
 - lr_scheduler_warmup_steps: 1000
-- training_steps: 30
+- training_steps: 5000
 - mixed_precision_training: Native AMP
 
 ### Training results
 
-| Training Loss | Epoch | Step | Validation Loss | Wer   |
-|:-------------:|:-----:|:----:|:---------------:|:-----:|
-| 8.8508        | 15.38 | 25   | 7.6994          | 100.0 |
+| Training Loss | Epoch | Step | Validation Loss | Wer      |
+|:-------------:|:-----:|:----:|:---------------:|:--------:|
+| 5.0576        | 1.45  | 1000 | 5.1841          | 158.1475 |
+| 4.6405        | 2.9   | 2000 | 4.8881          | 131.9237 |
+| 4.0627        | 4.35  | 3000 | 4.9336          | 143.5772 |
+| 3.781         | 5.8   | 4000 | 4.9113          | 129.0976 |
+| 3.0831        | 7.25  | 5000 | 5.0988          | 134.8396 |
 
 
 ### Framework versions

runs/Mar04_03-55-52_razer-14/events.out.tfevents.1709510155.razer-14.376.0

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:c9f7cc72f078f27d22c772c40d808895ea5c772aab34816d8e9ff01c1f808dd1
-size 40476
+oid sha256:22f3ca28464a89374e492bbd7a0ee0bd4fd135dacc69561fa2429eaae3f879e4
+size 49588