End of training

README.md

@@ -17,8 +17,8 @@ should probably proofread and complete it, then remove this comment. -->
 
 This model is a fine-tuned version of [openai/whisper-small](https://huggingface.co/openai/whisper-small) on the None dataset.
 It achieves the following results on the evaluation set:
-- Loss: 0.5825
-- Wer: 86.7596
+- Loss: 0.5452
+- Wer: 84.0883
 
 ## Model description
 
@@ -45,20 +45,21 @@ The following hyperparameters were used during training:
 - lr_scheduler_type: constant_with_warmup
 - lr_scheduler_warmup_steps: 50
 - training_steps: 800
+- mixed_precision_training: Native AMP
 
 ### Training results
 
-| Training Loss | Epoch | Step | Validation Loss | Wer     |
-|:-------------:|:-----:|:----:|:---------------:|:-------:|
-| 0.0871        | 6.25  | 200  | 0.3892          | 91.2892 |
-| 0.0159        | 12.5  | 400  | 0.4824          | 87.3403 |
-| 0.0087        | 18.75 | 600  | 0.5296          | 85.2497 |
-| 0.0063        | 25.0  | 800  | 0.5825          | 86.7596 |
+| Training Loss | Epoch | Step | Validation Loss | Wer      |
+|:-------------:|:-----:|:----:|:---------------:|:--------:|
+| 0.0869        | 6.25  | 200  | 0.3877          | 89.5470  |
+| 0.0138        | 12.5  | 400  | 0.4962          | 87.4564  |
+| 0.0088        | 18.75 | 600  | 0.5118          | 100.3484 |
+| 0.0058        | 25.0  | 800  | 0.5452          | 84.0883  |
 
 
 ### Framework versions
 
-- Transformers 4.34.0
-- Pytorch 2.0.1+cu118
-- Datasets 2.14.5
+- Transformers 4.35.0
+- Pytorch 2.0.1+cu117
+- Datasets 2.12.0
 - Tokenizers 0.14.0

generation_config.json

@@ -265,5 +265,5 @@
     "transcribe": 50359,
     "translate": 50358
   },
-  "transformers_version": "4.34.0"
+  "transformers_version": "4.35.0"
 }