Commit
•
2e17100
1
Parent(s):
9623a85
update model card README.md
Browse files
README.md
CHANGED
|
@@ -12,9 +12,6 @@ should probably proofread and complete it, then remove this comment. -->
|
|
| 12 |
# wav2vec2-base-en-in
|
| 13 |
|
| 14 |
This model is a fine-tuned version of [beinghorizontal/wav2vec2-base-en-in](https://huggingface.co/beinghorizontal/wav2vec2-base-en-in) on an unknown dataset.
|
| 15 |
-
It achieves the following results on the evaluation set:
|
| 16 |
-
- Loss: 3.0056
|
| 17 |
-
- Wer: 0.9997
|
| 18 |
|
| 19 |
## Model description
|
| 20 |
|
|
@@ -40,18 +37,9 @@ The following hyperparameters were used during training:
|
|
| 40 |
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
|
| 41 |
- lr_scheduler_type: linear
|
| 42 |
- lr_scheduler_warmup_steps: 1000
|
| 43 |
-
- num_epochs:
|
| 44 |
- mixed_precision_training: Native AMP
|
| 45 |
|
| 46 |
-
### Training results
|
| 47 |
-
|
| 48 |
-
| Training Loss | Epoch | Step | Validation Loss | Wer |
|
| 49 |
-
|:-------------:|:-----:|:----:|:---------------:|:------:|
|
| 50 |
-
| 29.5423 | 0.63 | 200 | 3.1266 | 0.9997 |
|
| 51 |
-
| 3.1063 | 1.26 | 400 | 3.0490 | 0.9997 |
|
| 52 |
-
| 3.0009 | 1.89 | 600 | 3.0056 | 0.9997 |
|
| 53 |
-
|
| 54 |
-
|
| 55 |
### Framework versions
|
| 56 |
|
| 57 |
- Transformers 4.17.0
|
|
|
|
| 12 |
# wav2vec2-base-en-in
|
| 13 |
|
| 14 |
This model is a fine-tuned version of [beinghorizontal/wav2vec2-base-en-in](https://huggingface.co/beinghorizontal/wav2vec2-base-en-in) on an unknown dataset.
|
|
|
|
|
|
|
|
|
|
| 15 |
|
| 16 |
## Model description
|
| 17 |
|
|
|
|
| 37 |
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
|
| 38 |
- lr_scheduler_type: linear
|
| 39 |
- lr_scheduler_warmup_steps: 1000
|
| 40 |
+
- num_epochs: 18
|
| 41 |
- mixed_precision_training: Native AMP
|
| 42 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 43 |
### Framework versions
|
| 44 |
|
| 45 |
- Transformers 4.17.0
|