Model save
Browse files
README.md
CHANGED
|
@@ -14,6 +14,14 @@ should probably proofread and complete it, then remove this comment. -->
|
|
| 14 |
# wav2vec2-large-xlsr-georgian_v1
|
| 15 |
|
| 16 |
This model is a fine-tuned version of [facebook/wav2vec2-large-xlsr-53](https://huggingface.co/facebook/wav2vec2-large-xlsr-53) on the None dataset.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 17 |
|
| 18 |
## Model description
|
| 19 |
|
|
@@ -41,7 +49,7 @@ The following hyperparameters were used during training:
|
|
| 41 |
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
|
| 42 |
- lr_scheduler_type: linear
|
| 43 |
- lr_scheduler_warmup_steps: 200
|
| 44 |
-
- num_epochs:
|
| 45 |
- mixed_precision_training: Native AMP
|
| 46 |
|
| 47 |
### Framework versions
|
|
|
|
| 14 |
# wav2vec2-large-xlsr-georgian_v1
|
| 15 |
|
| 16 |
This model is a fine-tuned version of [facebook/wav2vec2-large-xlsr-53](https://huggingface.co/facebook/wav2vec2-large-xlsr-53) on the None dataset.
|
| 17 |
+
It achieves the following results on the evaluation set:
|
| 18 |
+
- eval_loss: 0.1133
|
| 19 |
+
- eval_wer: 0.3244
|
| 20 |
+
- eval_runtime: 1068.9642
|
| 21 |
+
- eval_samples_per_second: 10.86
|
| 22 |
+
- eval_steps_per_second: 0.679
|
| 23 |
+
- epoch: 6.51
|
| 24 |
+
- step: 4000
|
| 25 |
|
| 26 |
## Model description
|
| 27 |
|
|
|
|
| 49 |
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
|
| 50 |
- lr_scheduler_type: linear
|
| 51 |
- lr_scheduler_warmup_steps: 200
|
| 52 |
+
- num_epochs: 15
|
| 53 |
- mixed_precision_training: Native AMP
|
| 54 |
|
| 55 |
### Framework versions
|
runs/Nov21_18-08-54_ae179e9c0ff1/events.out.tfevents.1700591687.ae179e9c0ff1.3443.0
CHANGED
|
Binary files a/runs/Nov21_18-08-54_ae179e9c0ff1/events.out.tfevents.1700591687.ae179e9c0ff1.3443.0 and b/runs/Nov21_18-08-54_ae179e9c0ff1/events.out.tfevents.1700591687.ae179e9c0ff1.3443.0 differ
|
|
|