End of training

README.md

@@ -24,7 +24,7 @@ model-index:
     metrics:
     - name: Wer
       type: wer
-      value: 14.757744241461475
+      value: 25.210484511517077
 ---
 
 <!-- This model card has been generated automatically according to the information the Trainer had access to. You
@@ -34,8 +34,8 @@ 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 Chinese English dataset.
 It achieves the following results on the evaluation set:
-- Loss: 0.3703
-- Wer: 14.7577
+- Loss: 0.8111
+- Wer: 25.2105
 
 ## Model description
 
@@ -59,22 +59,21 @@ The following hyperparameters were used during training:
 - eval_batch_size: 1
 - seed: 42
 - 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_type: cosine
 - lr_scheduler_warmup_steps: 500
-- training_steps: 3000
+- training_steps: 2000
 
 ### Training results
 
 | Training Loss | Epoch  | Step | Validation Loss | Wer     |
 |:-------------:|:------:|:----:|:---------------:|:-------:|
-| 0.0154        | 1.2837 | 1000 | 0.3491          | 14.8054 |
-| 0.0152        | 2.5674 | 2000 | 0.3584          | 14.4083 |
-| 0.0017        | 3.8511 | 3000 | 0.3703          | 14.7577 |
+| 0.0008        | 1.2837 | 1000 | 0.8132          | 25.7347 |
+| 0.0005        | 2.5674 | 2000 | 0.8111          | 25.2105 |
 
 
 ### Framework versions
 
 - Transformers 4.48.3
-- Pytorch 2.5.1+cu124
+- Pytorch 2.6.0+cu124
 - Datasets 3.4.0
 - Tokenizers 0.21.0