update model card README.md
Browse files
README.md
CHANGED
@@ -14,13 +14,13 @@ should probably proofread and complete it, then remove this comment. -->
|
|
14 |
|
15 |
This model is a fine-tuned version of [Hyuk/wav2vec2-korean-v3](https://huggingface.co/Hyuk/wav2vec2-korean-v3) on the None dataset.
|
16 |
It achieves the following results on the evaluation set:
|
17 |
-
- eval_loss:
|
18 |
-
- eval_wer: 0.
|
19 |
-
- eval_runtime: 56.
|
20 |
-
- eval_samples_per_second: 10.
|
21 |
-
- eval_steps_per_second: 1.
|
22 |
-
- epoch:
|
23 |
-
- step:
|
24 |
|
25 |
## Model description
|
26 |
|
@@ -48,7 +48,7 @@ The following hyperparameters were used during training:
|
|
48 |
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
|
49 |
- lr_scheduler_type: linear
|
50 |
- lr_scheduler_warmup_steps: 400
|
51 |
-
- num_epochs:
|
52 |
- mixed_precision_training: Native AMP
|
53 |
|
54 |
### Framework versions
|
|
|
14 |
|
15 |
This model is a fine-tuned version of [Hyuk/wav2vec2-korean-v3](https://huggingface.co/Hyuk/wav2vec2-korean-v3) on the None dataset.
|
16 |
It achieves the following results on the evaluation set:
|
17 |
+
- eval_loss: 0.9860
|
18 |
+
- eval_wer: 0.3797
|
19 |
+
- eval_runtime: 56.5559
|
20 |
+
- eval_samples_per_second: 10.167
|
21 |
+
- eval_steps_per_second: 1.273
|
22 |
+
- epoch: 9.73
|
23 |
+
- step: 12300
|
24 |
|
25 |
## Model description
|
26 |
|
|
|
48 |
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
|
49 |
- lr_scheduler_type: linear
|
50 |
- lr_scheduler_warmup_steps: 400
|
51 |
+
- num_epochs: 10
|
52 |
- mixed_precision_training: Native AMP
|
53 |
|
54 |
### Framework versions
|