Hugging Face's logo

spudnik
/
wavlm-emotion

Audio Classification
Transformers
Safetensors
wavlm
Generated from Trainer
Model card Files
xet
 Community
spudnik commited on
Commit
9db0795
·
verified ·
1 Parent(s): 42ca6ea

spudnik/group55_wavlm_audio_emotion_recognition

Browse files
Files changed (3) hide show
  1. README.md +55 -35
  2. model.safetensors +1 -1
  3. training_args.bin +1 -1
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/wavlm-base](https://huggingface.co/microsoft/wavlm-base) on the None dataset.
17
  It achieves the following results on the evaluation set:
18
- - Loss: 1.9462
19
 
20
  ## Model description
21
 
@@ -34,49 +34,69 @@ More information needed
34
  ### Training hyperparameters
35
 
36
  The following hyperparameters were used during training:
37
- - learning_rate: 5e-06
38
- - train_batch_size: 16
39
- - eval_batch_size: 16
40
  - seed: 42
41
  - optimizer: Use OptimizerNames.ADAMW_TORCH with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
42
  - lr_scheduler_type: linear
43
- - num_epochs: 30
44
  - mixed_precision_training: Native AMP
45
 
46
  ### Training results
47
 
48
  | Training Loss | Epoch | Step | Validation Loss |
49
  |:-------------:|:-----:|:----:|:---------------:|
50
- | 1.9476 | 1.0 | 191 | 1.9472 |
51
- | 1.9471 | 2.0 | 382 | 1.9478 |
52
- | 1.9462 | 3.0 | 573 | 1.9462 |
53
- | 1.9485 | 4.0 | 764 | 1.9473 |
54
- | 1.9465 | 5.0 | 955 | 1.9447 |
55
- | 1.9481 | 6.0 | 1146 | 1.9501 |
56
- | 1.9472 | 7.0 | 1337 | 1.9492 |
57
- | 1.9466 | 8.0 | 1528 | 1.9474 |
58
- | 1.9468 | 9.0 | 1719 | 1.9467 |
59
- | 1.9467 | 10.0 | 1910 | 1.9439 |
60
- | 1.9478 | 11.0 | 2101 | 1.9480 |
61
- | 1.9462 | 12.0 | 2292 | 1.9495 |
62
- | 1.9471 | 13.0 | 2483 | 1.9496 |
63
- | 1.947 | 14.0 | 2674 | 1.9495 |
64
- | 1.9473 | 15.0 | 2865 | 1.9488 |
65
- | 1.9485 | 16.0 | 3056 | 1.9487 |
66
- | 1.9465 | 17.0 | 3247 | 1.9465 |
67
- | 1.9475 | 18.0 | 3438 | 1.9465 |
68
- | 1.9476 | 19.0 | 3629 | 1.9466 |
69
- | 1.9474 | 20.0 | 3820 | 1.9466 |
70
- | 1.9463 | 21.0 | 4011 | 1.9460 |
71
- | 1.9473 | 22.0 | 4202 | 1.9448 |
72
- | 1.9465 | 23.0 | 4393 | 1.9453 |
73
- | 1.9477 | 24.0 | 4584 | 1.9450 |
74
- | 1.9456 | 25.0 | 4775 | 1.9455 |
75
- | 1.9463 | 26.0 | 4966 | 1.9463 |
76
- | 1.9473 | 27.0 | 5157 | 1.9460 |
77
- | 1.948 | 28.0 | 5348 | 1.9462 |
78
- | 1.9462 | 29.0 | 5539 | 1.9462 |
79
- | 1.9456 | 30.0 | 5730 | 1.9462 |
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
80
 
81
 
82
  ### Framework versions
 
15
 
16
  This model is a fine-tuned version of [microsoft/wavlm-base](https://huggingface.co/microsoft/wavlm-base) on the None dataset.
17
  It achieves the following results on the evaluation set:
18
+ - Loss: 1.9485
19
 
20
  ## Model description
21
 
 
34
  ### Training hyperparameters
35
 
36
  The following hyperparameters were used during training:
37
+ - learning_rate: 1e-05
38
+ - train_batch_size: 32
39
+ - eval_batch_size: 32
40
  - seed: 42
41
  - optimizer: Use OptimizerNames.ADAMW_TORCH with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
42
  - lr_scheduler_type: linear
43
+ - num_epochs: 50
44
  - mixed_precision_training: Native AMP
45
 
46
  ### Training results
47
 
48
  | Training Loss | Epoch | Step | Validation Loss |
49
  |:-------------:|:-----:|:----:|:---------------:|
50
+ | No log | 1.0 | 96 | 1.9509 |
51
+ | 1.9473 | 2.0 | 192 | 1.9486 |
52
+ | 1.9469 | 3.0 | 288 | 1.9512 |
53
+ | 1.9469 | 4.0 | 384 | 1.9495 |
54
+ | 1.9471 | 5.0 | 480 | 1.9488 |
55
+ | 1.9468 | 6.0 | 576 | 1.9474 |
56
+ | 1.9478 | 7.0 | 672 | 1.9510 |
57
+ | 1.9487 | 8.0 | 768 | 1.9510 |
58
+ | 1.9487 | 9.0 | 864 | 1.9466 |
59
+ | 1.9471 | 10.0 | 960 | 1.9465 |
60
+ | 1.9471 | 11.0 | 1056 | 1.9505 |
61
+ | 1.9477 | 12.0 | 1152 | 1.9467 |
62
+ | 1.9479 | 13.0 | 1248 | 1.9465 |
63
+ | 1.948 | 14.0 | 1344 | 1.9439 |
64
+ | 1.9473 | 15.0 | 1440 | 1.9488 |
65
+ | 1.9488 | 16.0 | 1536 | 1.9432 |
66
+ | 1.9472 | 17.0 | 1632 | 1.9491 |
67
+ | 1.9465 | 18.0 | 1728 | 1.9461 |
68
+ | 1.9468 | 19.0 | 1824 | 1.9507 |
69
+ | 1.9481 | 20.0 | 1920 | 1.9480 |
70
+ | 1.9471 | 21.0 | 2016 | 1.9442 |
71
+ | 1.9477 | 22.0 | 2112 | 1.9466 |
72
+ | 1.9474 | 23.0 | 2208 | 1.9504 |
73
+ | 1.9468 | 24.0 | 2304 | 1.9496 |
74
+ | 1.9476 | 25.0 | 2400 | 1.9464 |
75
+ | 1.9476 | 26.0 | 2496 | 1.9466 |
76
+ | 1.9471 | 27.0 | 2592 | 1.9473 |
77
+ | 1.9467 | 28.0 | 2688 | 1.9485 |
78
+ | 1.9468 | 29.0 | 2784 | 1.9477 |
79
+ | 1.9471 | 30.0 | 2880 | 1.9484 |
80
+ | 1.9458 | 31.0 | 2976 | 1.9472 |
81
+ | 1.9476 | 32.0 | 3072 | 1.9475 |
82
+ | 1.9468 | 33.0 | 3168 | 1.9482 |
83
+ | 1.9468 | 34.0 | 3264 | 1.9495 |
84
+ | 1.9463 | 35.0 | 3360 | 1.9497 |
85
+ | 1.9474 | 36.0 | 3456 | 1.9490 |
86
+ | 1.9462 | 37.0 | 3552 | 1.9481 |
87
+ | 1.9458 | 38.0 | 3648 | 1.9490 |
88
+ | 1.9461 | 39.0 | 3744 | 1.9486 |
89
+ | 1.9446 | 40.0 | 3840 | 1.9488 |
90
+ | 1.946 | 41.0 | 3936 | 1.9490 |
91
+ | 1.9467 | 42.0 | 4032 | 1.9487 |
92
+ | 1.9466 | 43.0 | 4128 | 1.9485 |
93
+ | 1.9463 | 44.0 | 4224 | 1.9486 |
94
+ | 1.9459 | 45.0 | 4320 | 1.9486 |
95
+ | 1.9458 | 46.0 | 4416 | 1.9487 |
96
+ | 1.9464 | 47.0 | 4512 | 1.9485 |
97
+ | 1.946 | 48.0 | 4608 | 1.9485 |
98
+ | 1.9459 | 49.0 | 4704 | 1.9485 |
99
+ | 1.9459 | 50.0 | 4800 | 1.9485 |
100
 
101
 
102
  ### Framework versions
model.safetensors CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:6888077516daaa659582a615d36f0d39ed9fdb520ab5aad454235dbed81247d1
3
  size 378352396
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:99000f3e1a9057e29f9515df0bfc450d77194fb2ed4dc2ede8d6f025c6748637
3
  size 378352396
training_args.bin CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:cbc6efb5f4537290c8c3aad70a7478ac0afc81c160474819d6956f7d24a3ab22
3
  size 5240
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:8b2ef79bc71f399a7c1698c8c4a5746d22f85645f604a1a2d15ada70aa6a0831
3
  size 5240