End of training
Browse files
README.md
CHANGED
|
@@ -19,6 +19,9 @@ should probably proofread and complete it, then remove this comment. -->
|
|
| 19 |
# Whisper large-v2 Korean - ML_project_custom_data_1epoch_with500
|
| 20 |
|
| 21 |
This model is a fine-tuned version of [openai/whisper-large-v2](https://huggingface.co/openai/whisper-large-v2) on the customd_ataset dataset.
|
|
|
|
|
|
|
|
|
|
| 22 |
|
| 23 |
## Model description
|
| 24 |
|
|
@@ -51,7 +54,7 @@ The following hyperparameters were used during training:
|
|
| 51 |
|
| 52 |
| Training Loss | Epoch | Step | Validation Loss | Cer |
|
| 53 |
|:-------------:|:-----:|:----:|:---------------:|:-------:|
|
| 54 |
-
|
|
| 55 |
|
| 56 |
|
| 57 |
### Framework versions
|
|
|
|
| 19 |
# Whisper large-v2 Korean - ML_project_custom_data_1epoch_with500
|
| 20 |
|
| 21 |
This model is a fine-tuned version of [openai/whisper-large-v2](https://huggingface.co/openai/whisper-large-v2) on the customd_ataset dataset.
|
| 22 |
+
It achieves the following results on the evaluation set:
|
| 23 |
+
- Loss: 0.6012
|
| 24 |
+
- Cer: 81.9576
|
| 25 |
|
| 26 |
## Model description
|
| 27 |
|
|
|
|
| 54 |
|
| 55 |
| Training Loss | Epoch | Step | Validation Loss | Cer |
|
| 56 |
|:-------------:|:-----:|:----:|:---------------:|:-------:|
|
| 57 |
+
| 0.3222 | 1.0 | 113 | 0.6012 | 81.9576 |
|
| 58 |
|
| 59 |
|
| 60 |
### Framework versions
|