Hugging Face's logo

mikhail-panzo
/
zlm_b32_le5_s12000

Text-to-Audio
Transformers
TensorBoard
Safetensors
speecht5
Generated from Trainer
Inference Endpoints
Model card Files Metrics Community
mikhail-panzo commited on
Commit
084abb0
•
1 Parent(s): 9de99a3

End of training

Browse files
Files changed (3) hide show
  1. README.md +29 -27
  2. model.safetensors +1 -1
  3. runs/May04_13-12-07_a378709cee8e/events.out.tfevents.1714828328.a378709cee8e.459.0 +2 -2
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: 0.3561
19
 
20
  ## Model description
21
 
@@ -35,43 +35,45 @@ More information needed
35
 
36
  The following hyperparameters were used during training:
37
  - learning_rate: 1e-05
38
- - train_batch_size: 32
39
  - eval_batch_size: 8
40
  - seed: 42
 
 
41
  - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
42
  - lr_scheduler_type: linear
43
  - lr_scheduler_warmup_steps: 2000
44
- - training_steps: 12050
45
  - mixed_precision_training: Native AMP
46
 
47
  ### Training results
48
 
49
  | Training Loss | Epoch | Step | Validation Loss |
50
  |:-------------:|:------:|:-----:|:---------------:|
51
- | 0.4311 | 0.2094 | 500 | 0.3863 |
52
- | 0.4432 | 0.4188 | 1000 | 0.3867 |
53
- | 0.4185 | 0.6281 | 1500 | 0.3835 |
54
- | 0.4257 | 0.8375 | 2000 | 0.3799 |
55
- | 0.4112 | 1.0469 | 2500 | 0.3801 |
56
- | 0.4178 | 1.2563 | 3000 | 0.3758 |
57
- | 0.4069 | 1.4657 | 3500 | 0.3738 |
58
- | 0.4015 | 1.6750 | 4000 | 0.3724 |
59
- | 0.4155 | 1.8844 | 4500 | 0.3700 |
60
- | 0.4126 | 2.0938 | 5000 | 0.3674 |
61
- | 0.4084 | 2.3032 | 5500 | 0.3662 |
62
- | 0.396 | 2.5126 | 6000 | 0.3621 |
63
- | 0.4084 | 2.7219 | 6500 | 0.3648 |
64
- | 0.3949 | 2.9313 | 7000 | 0.3608 |
65
- | 0.4045 | 3.1407 | 7500 | 0.3619 |
66
- | 0.4078 | 3.3501 | 8000 | 0.3607 |
67
- | 0.3926 | 3.5595 | 8500 | 0.3583 |
68
- | 0.4007 | 3.7688 | 9000 | 0.3579 |
69
- | 0.3899 | 3.9782 | 9500 | 0.3589 |
70
- | 0.3902 | 4.1876 | 10000 | 0.3566 |
71
- | 0.4023 | 4.3970 | 10500 | 0.3594 |
72
- | 0.3971 | 4.6064 | 11000 | 0.3552 |
73
- | 0.3956 | 4.8157 | 11500 | 0.3562 |
74
- | 0.3974 | 5.0251 | 12000 | 0.3561 |
75
 
76
 
77
  ### 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: 0.3707
19
 
20
  ## Model description
21
 
 
35
 
36
  The following hyperparameters were used during training:
37
  - learning_rate: 1e-05
38
+ - train_batch_size: 16
39
  - eval_batch_size: 8
40
  - seed: 42
41
+ - gradient_accumulation_steps: 2
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: 2000
46
+ - training_steps: 12000
47
  - mixed_precision_training: Native AMP
48
 
49
  ### Training results
50
 
51
  | Training Loss | Epoch | Step | Validation Loss |
52
  |:-------------:|:------:|:-----:|:---------------:|
53
+ | 0.7211 | 0.2094 | 500 | 0.6148 |
54
+ | 0.6059 | 0.4188 | 1000 | 0.5140 |
55
+ | 0.5347 | 0.6283 | 1500 | 0.4725 |
56
+ | 0.4888 | 0.8377 | 2000 | 0.4612 |
57
+ | 0.4923 | 1.0471 | 2500 | 0.4283 |
58
+ | 0.466 | 1.2565 | 3000 | 0.4163 |
59
+ | 0.4535 | 1.4660 | 3500 | 0.4090 |
60
+ | 0.4442 | 1.6754 | 4000 | 0.4009 |
61
+ | 0.4423 | 1.8848 | 4500 | 0.3955 |
62
+ | 0.4539 | 2.0942 | 5000 | 0.3916 |
63
+ | 0.4416 | 2.3037 | 5500 | 0.3870 |
64
+ | 0.4306 | 2.5131 | 6000 | 0.3856 |
65
+ | 0.4242 | 2.7225 | 6500 | 0.3819 |
66
+ | 0.426 | 2.9319 | 7000 | 0.3814 |
67
+ | 0.4105 | 3.1414 | 7500 | 0.3787 |
68
+ | 0.4077 | 3.3508 | 8000 | 0.3750 |
69
+ | 0.4106 | 3.5602 | 8500 | 0.3748 |
70
+ | 0.4228 | 3.7696 | 9000 | 0.3728 |
71
+ | 0.4101 | 3.9791 | 9500 | 0.3719 |
72
+ | 0.4209 | 4.1885 | 10000 | 0.3707 |
73
+ | 0.4091 | 4.3979 | 10500 | 0.3712 |
74
+ | 0.4061 | 4.6073 | 11000 | 0.3715 |
75
+ | 0.4169 | 4.8168 | 11500 | 0.3700 |
76
+ | 0.4088 | 5.0262 | 12000 | 0.3707 |
77
 
78
 
79
  ### Framework versions
model.safetensors CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:56a801f7874b64e033e7d65093ff6a670b591b9e352bb55a9efedb82bb7bd081
3
  size 577789320
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:2f4bc91b2bf1d4a600334b1faa106af9988bbfd956f7192f84cc46e78a068110
3
  size 577789320
runs/May04_13-12-07_a378709cee8e/events.out.tfevents.1714828328.a378709cee8e.459.0 CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:eeb9ba541b42285c08a0959024cc54b423e77455ec5e6cbada22bb3ef1e00f56
3
- size 63624
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:9f1f0ab0b9f1686087b01857d0c271842775d3d160d255601e3cbd37223e67d3
3
+ size 63978