mikhail-panzo
commited on
Commit
•
a08dd46
1
Parent(s):
060d55f
End of training
Browse files
README.md
CHANGED
@@ -15,7 +15,7 @@ should probably proofread and complete it, then remove this comment. -->
|
|
15 |
|
16 |
This model is a fine-tuned version of [microsoft/speecht5_tts](https://huggingface.co/microsoft/speecht5_tts) on an unknown dataset.
|
17 |
It achieves the following results on the evaluation set:
|
18 |
-
- Loss:
|
19 |
|
20 |
## Model description
|
21 |
|
@@ -36,19 +36,24 @@ More information needed
|
|
36 |
The following hyperparameters were used during training:
|
37 |
- learning_rate: 1e-05
|
38 |
- train_batch_size: 4
|
39 |
-
- eval_batch_size:
|
40 |
- seed: 42
|
|
|
|
|
41 |
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
|
42 |
- lr_scheduler_type: linear
|
43 |
-
-
|
|
|
|
|
44 |
|
45 |
### Training results
|
46 |
|
47 |
| Training Loss | Epoch | Step | Validation Loss |
|
48 |
|:-------------:|:-----:|:----:|:---------------:|
|
49 |
-
|
|
50 |
-
|
|
51 |
-
|
|
|
|
52 |
|
53 |
|
54 |
### Framework versions
|
|
|
15 |
|
16 |
This model is a fine-tuned version of [microsoft/speecht5_tts](https://huggingface.co/microsoft/speecht5_tts) on an unknown dataset.
|
17 |
It achieves the following results on the evaluation set:
|
18 |
+
- Loss: 0.7314
|
19 |
|
20 |
## Model description
|
21 |
|
|
|
36 |
The following hyperparameters were used during training:
|
37 |
- learning_rate: 1e-05
|
38 |
- train_batch_size: 4
|
39 |
+
- eval_batch_size: 2
|
40 |
- seed: 42
|
41 |
+
- gradient_accumulation_steps: 8
|
42 |
+
- total_train_batch_size: 32
|
43 |
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
|
44 |
- lr_scheduler_type: linear
|
45 |
+
- lr_scheduler_warmup_steps: 500
|
46 |
+
- training_steps: 4000
|
47 |
+
- mixed_precision_training: Native AMP
|
48 |
|
49 |
### Training results
|
50 |
|
51 |
| Training Loss | Epoch | Step | Validation Loss |
|
52 |
|:-------------:|:-----:|:----:|:---------------:|
|
53 |
+
| 0.9147 | 13.22 | 1000 | 0.8691 |
|
54 |
+
| 0.812 | 26.45 | 2000 | 0.7710 |
|
55 |
+
| 0.7837 | 39.67 | 3000 | 0.7394 |
|
56 |
+
| 0.7648 | 52.89 | 4000 | 0.7314 |
|
57 |
|
58 |
|
59 |
### Framework versions
|