End of training
Browse files
README.md
CHANGED
@@ -20,9 +20,9 @@ should probably proofread and complete it, then remove this comment. -->
|
|
20 |
|
21 |
This model is a fine-tuned version of [CKSINGH/whisper-small-hi-iiib](https://huggingface.co/CKSINGH/whisper-small-hi-iiib) on the Common Voice 13.0 dataset.
|
22 |
It achieves the following results on the evaluation set:
|
23 |
-
- Loss: 0.
|
24 |
-
- Wer:
|
25 |
-
- Cer:
|
26 |
|
27 |
## Model description
|
28 |
|
@@ -47,16 +47,17 @@ The following hyperparameters were used during training:
|
|
47 |
- seed: 42
|
48 |
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
|
49 |
- lr_scheduler_type: linear
|
50 |
-
- lr_scheduler_warmup_steps:
|
51 |
-
- training_steps:
|
52 |
- mixed_precision_training: Native AMP
|
53 |
|
54 |
### Training results
|
55 |
|
56 |
-
| Training Loss | Epoch | Step | Validation Loss | Wer | Cer
|
57 |
-
|
58 |
-
| 0.
|
59 |
-
| 0.
|
|
|
60 |
|
61 |
|
62 |
### Framework versions
|
|
|
20 |
|
21 |
This model is a fine-tuned version of [CKSINGH/whisper-small-hi-iiib](https://huggingface.co/CKSINGH/whisper-small-hi-iiib) on the Common Voice 13.0 dataset.
|
22 |
It achieves the following results on the evaluation set:
|
23 |
+
- Loss: 0.2473
|
24 |
+
- Wer: 21.4253
|
25 |
+
- Cer: 7.5700
|
26 |
|
27 |
## Model description
|
28 |
|
|
|
47 |
- seed: 42
|
48 |
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
|
49 |
- lr_scheduler_type: linear
|
50 |
+
- lr_scheduler_warmup_steps: 500
|
51 |
+
- training_steps: 3000
|
52 |
- mixed_precision_training: Native AMP
|
53 |
|
54 |
### Training results
|
55 |
|
56 |
+
| Training Loss | Epoch | Step | Validation Loss | Wer | Cer |
|
57 |
+
|:-------------:|:-----:|:----:|:---------------:|:-------:|:------:|
|
58 |
+
| 0.0382 | 3.57 | 1000 | 0.2021 | 22.8043 | 8.1173 |
|
59 |
+
| 0.0027 | 7.14 | 2000 | 0.2321 | 21.6614 | 7.6820 |
|
60 |
+
| 0.0001 | 10.71 | 3000 | 0.2473 | 21.4253 | 7.5700 |
|
61 |
|
62 |
|
63 |
### Framework versions
|