arturslogins
commited on
Commit
•
332378e
1
Parent(s):
6f73192
End of training
Browse files
README.md
CHANGED
@@ -25,7 +25,7 @@ model-index:
|
|
25 |
metrics:
|
26 |
- name: Wer
|
27 |
type: wer
|
28 |
-
value:
|
29 |
---
|
30 |
|
31 |
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
|
@@ -35,8 +35,8 @@ should probably proofread and complete it, then remove this comment. -->
|
|
35 |
|
36 |
This model is a fine-tuned version of [arturslogins/whisper-medium-lv](https://huggingface.co/arturslogins/whisper-medium-lv) on the Recorded Voice dataset.
|
37 |
It achieves the following results on the evaluation set:
|
38 |
-
- Loss: 1.
|
39 |
-
- Wer:
|
40 |
|
41 |
## Model description
|
42 |
|
@@ -56,10 +56,10 @@ More information needed
|
|
56 |
|
57 |
The following hyperparameters were used during training:
|
58 |
- learning_rate: 1e-05
|
59 |
-
- train_batch_size:
|
60 |
-
- eval_batch_size:
|
61 |
- seed: 42
|
62 |
-
- gradient_accumulation_steps:
|
63 |
- total_train_batch_size: 32
|
64 |
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
|
65 |
- lr_scheduler_type: linear
|
@@ -69,13 +69,13 @@ The following hyperparameters were used during training:
|
|
69 |
|
70 |
### Training results
|
71 |
|
72 |
-
| Training Loss | Epoch
|
73 |
-
|
74 |
-
| 0.
|
75 |
-
| 0.
|
76 |
-
| 0.
|
77 |
-
| 0.0001 | 25.
|
78 |
-
| 0.0001 |
|
79 |
|
80 |
|
81 |
### Framework versions
|
|
|
25 |
metrics:
|
26 |
- name: Wer
|
27 |
type: wer
|
28 |
+
value: 43.705463182897866
|
29 |
---
|
30 |
|
31 |
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
|
|
|
35 |
|
36 |
This model is a fine-tuned version of [arturslogins/whisper-medium-lv](https://huggingface.co/arturslogins/whisper-medium-lv) on the Recorded Voice dataset.
|
37 |
It achieves the following results on the evaluation set:
|
38 |
+
- Loss: 1.0843
|
39 |
+
- Wer: 43.7055
|
40 |
|
41 |
## Model description
|
42 |
|
|
|
56 |
|
57 |
The following hyperparameters were used during training:
|
58 |
- learning_rate: 1e-05
|
59 |
+
- train_batch_size: 16
|
60 |
+
- eval_batch_size: 12
|
61 |
- seed: 42
|
62 |
+
- gradient_accumulation_steps: 2
|
63 |
- total_train_batch_size: 32
|
64 |
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
|
65 |
- lr_scheduler_type: linear
|
|
|
69 |
|
70 |
### Training results
|
71 |
|
72 |
+
| Training Loss | Epoch | Step | Validation Loss | Wer |
|
73 |
+
|:-------------:|:-------:|:----:|:---------------:|:-------:|
|
74 |
+
| 0.0097 | 6.3898 | 1000 | 0.9780 | 45.8432 |
|
75 |
+
| 0.0023 | 12.7796 | 2000 | 1.0158 | 44.9525 |
|
76 |
+
| 0.0002 | 19.1693 | 3000 | 1.0549 | 46.1758 |
|
77 |
+
| 0.0001 | 25.5591 | 4000 | 1.0765 | 44.9881 |
|
78 |
+
| 0.0001 | 31.9489 | 5000 | 1.0843 | 43.7055 |
|
79 |
|
80 |
|
81 |
### Framework versions
|