Merge branch 'main' of https://huggingface.co/Plim/xls-r-1b-cv_8-fr into main
Browse files
README.md
CHANGED
@@ -79,10 +79,10 @@ The following hyperparameters were used during training:
|
|
79 |
| 0.8015 | 3.44 | 12000 | inf | 0.1880 |
|
80 |
| 0.774 | 3.73 | 13000 | inf | 0.1834 |
|
81 |
|
82 |
-
It achieves the best result
|
83 |
- Wer: 0.1834
|
84 |
|
85 |
-
Some
|
86 |
|
87 |
### Framework versions
|
88 |
|
@@ -102,3 +102,4 @@ python eval.py --model_id Plim/xls-r-1b-cv_8-fr --dataset mozilla-foundation/com
|
|
102 |
|
103 |
```bash
|
104 |
python eval.py --model_id Plim/xls-r-1b-cv_8-fr --dataset speech-recognition-community-v2/dev_data --config fr --split validation --chunk_length_s 5.0 --stride_length_s 1.0
|
|
|
|
79 |
| 0.8015 | 3.44 | 12000 | inf | 0.1880 |
|
80 |
| 0.774 | 3.73 | 13000 | inf | 0.1834 |
|
81 |
|
82 |
+
It achieves the best result on the validation set on STEP 13000:
|
83 |
- Wer: 0.1834
|
84 |
|
85 |
+
Some problem occurs when calculating the validation loss.
|
86 |
|
87 |
### Framework versions
|
88 |
|
|
|
102 |
|
103 |
```bash
|
104 |
python eval.py --model_id Plim/xls-r-1b-cv_8-fr --dataset speech-recognition-community-v2/dev_data --config fr --split validation --chunk_length_s 5.0 --stride_length_s 1.0
|
105 |
+
```
|