End of training

README.md

@@ -1,7 +1,7 @@
 ---
 library_name: transformers
 license: apache-2.0
-base_model: openai/whisper-medium.en
+base_model: crossdelenna/medium_cross.en
 tags:
 - generated_from_trainer
 metrics:
@@ -16,10 +16,10 @@ should probably proofread and complete it, then remove this comment. -->
 
 # medium_cross.en
 
-This model is a fine-tuned version of [openai/whisper-medium.en](https://huggingface.co/openai/whisper-medium.en) on an unknown dataset.
+This model is a fine-tuned version of [crossdelenna/medium_cross.en](https://huggingface.co/crossdelenna/medium_cross.en) on an unknown dataset.
 It achieves the following results on the evaluation set:
-- Loss: 0.7616
-- Wer: 34.6503
+- Loss: 0.3034
+- Wer: 15.1384
 
 ## Model description
 
@@ -39,21 +39,22 @@ More information needed
 
 The following hyperparameters were used during training:
 - learning_rate: 1e-05
-- train_batch_size: 16
-- eval_batch_size: 8
+- train_batch_size: 22
+- eval_batch_size: 22
 - 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_warmup_steps: 10
-- training_steps: 401
+- training_steps: 1051
 - mixed_precision_training: Native AMP
 
 ### Training results
 
 | Training Loss | Epoch  | Step | Validation Loss | Wer     |
 |:-------------:|:------:|:----:|:---------------:|:-------:|
-| 2.2838        | 0.5155 | 200  | 1.0021          | 36.0184 |
-| 0.9275        | 1.0309 | 400  | 0.7616          | 34.6503 |
+| 0.664         | 1.2411 | 350  | 0.3998          | 18.2094 |
+| 0.4625        | 2.4823 | 700  | 0.3244          | 16.0633 |
+| 0.3703        | 3.7234 | 1050 | 0.3034          | 15.1384 |
 
 
 ### Framework versions

runs/Feb10_17-08-50_b24e50d8c658/events.out.tfevents.1739207638.b24e50d8c658.405.0

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:73c1659cfd6ae1e78843d542ba5d0a222558ca0c38eb156a9efc0f9ee2518801
-size 7510
+oid sha256:dbbdd3a6db533de96ac1fc8f73287ac223d3cfa065e5a9414dcce2637b47b399
+size 7864