Hugging Face's logo

Bagus
/
wavlm_finetuned_emodb

Transformers
Safetensors
wavlm
Generated from Trainer
Inference Endpoints
Model card Files Community
Bagus commited on
Commit
b5d9b7a
·
verified ·
1 Parent(s): 91cc754

Model save

Browse files
Files changed (6) hide show
  1. README.md +120 -0
  2. model.safetensors +1 -1
  3. torch/config.json +136 -0
  4. torch/model.safetensors +3 -0
  5. torch/preprocessor_config.json +9 -0
  6. torch/training_args.bin +3 -0
README.md ADDED
@@ -0,0 +1,120 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ base_model: microsoft/wavlm-base
3
+ tags:
4
+ - generated_from_trainer
5
+ model-index:
6
+ - name: wavlm_finetuned_emodb
7
+ results: []
8
+ ---
9
+
10
+ <!-- This model card has been generated automatically according to the information the Trainer had access to. You
11
+ should probably proofread and complete it, then remove this comment. -->
12
+
13
+ # wavlm_finetuned_emodb
14
+
15
+ This model is a fine-tuned version of [microsoft/wavlm-base](https://huggingface.co/microsoft/wavlm-base) on the None dataset.
16
+ It achieves the following results on the evaluation set:
17
+ - Loss: 0.9254
18
+ - Uar: 0.8148
19
+ - Acc: 0.8529
20
+
21
+ ## Model description
22
+
23
+ More information needed
24
+
25
+ ## Intended uses & limitations
26
+
27
+ More information needed
28
+
29
+ ## Training and evaluation data
30
+
31
+ More information needed
32
+
33
+ ## Training procedure
34
+
35
+ ### Training hyperparameters
36
+
37
+ The following hyperparameters were used during training:
38
+ - learning_rate: 0.0001
39
+ - train_batch_size: 8
40
+ - eval_batch_size: 8
41
+ - seed: 42
42
+ - gradient_accumulation_steps: 4
43
+ - total_train_batch_size: 32
44
+ - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
45
+ - lr_scheduler_type: linear
46
+ - num_epochs: 10
47
+ - mixed_precision_training: Native AMP
48
+
49
+ ### Training results
50
+
51
+ | Training Loss | Epoch | Step | Validation Loss | Uar | Acc |
52
+ |:-------------:|:------:|:----:|:---------------:|:------:|:------:|
53
+ | 1.3857 | 0.1538 | 1 | 1.3786 | 0.25 | 0.1985 |
54
+ | 1.3322 | 0.3077 | 2 | 1.3549 | 0.2914 | 0.2426 |
55
+ | 1.3112 | 0.4615 | 3 | 1.3165 | 0.5375 | 0.6103 |
56
+ | 1.2981 | 0.6154 | 4 | 1.2905 | 0.5 | 0.6029 |
57
+ | 1.1317 | 0.7692 | 5 | 1.2923 | 0.4907 | 0.5956 |
58
+ | 1.2078 | 0.9231 | 6 | 1.2619 | 0.5556 | 0.6471 |
59
+ | 0.9237 | 1.0769 | 7 | 1.2254 | 0.5741 | 0.6618 |
60
+ | 0.8396 | 1.2308 | 8 | 1.2247 | 0.5556 | 0.6471 |
61
+ | 1.0354 | 1.3846 | 9 | 1.2076 | 0.5556 | 0.6471 |
62
+ | 0.9205 | 1.5385 | 10 | 1.1891 | 0.5833 | 0.6691 |
63
+ | 0.9071 | 1.6923 | 11 | 1.1704 | 0.6481 | 0.7206 |
64
+ | 0.8132 | 1.8462 | 12 | 1.1988 | 0.6939 | 0.5735 |
65
+ | 0.8994 | 2.0 | 13 | 1.1960 | 0.6574 | 0.5221 |
66
+ | 0.7924 | 2.1538 | 14 | 1.1579 | 0.6658 | 0.5662 |
67
+ | 0.7386 | 2.3077 | 15 | 1.1401 | 0.6944 | 0.7574 |
68
+ | 0.6324 | 2.4615 | 16 | 1.1202 | 0.6111 | 0.6912 |
69
+ | 0.7282 | 2.6154 | 17 | 1.1090 | 0.5833 | 0.6691 |
70
+ | 0.673 | 2.7692 | 18 | 1.0907 | 0.6111 | 0.6912 |
71
+ | 0.623 | 2.9231 | 19 | 1.0578 | 0.7872 | 0.8235 |
72
+ | 0.4954 | 3.0769 | 20 | 1.0357 | 0.8475 | 0.8676 |
73
+ | 0.5201 | 3.2308 | 21 | 1.0365 | 0.7778 | 0.8235 |
74
+ | 0.5608 | 3.3846 | 22 | 1.0346 | 0.75 | 0.8015 |
75
+ | 0.6334 | 3.5385 | 23 | 1.0047 | 0.7685 | 0.8162 |
76
+ | 0.3737 | 3.6923 | 24 | 0.9585 | 0.8658 | 0.8897 |
77
+ | 0.5369 | 3.8462 | 25 | 0.9527 | 0.9178 | 0.8824 |
78
+ | 0.3599 | 4.0 | 26 | 0.9682 | 0.8906 | 0.8382 |
79
+ | 0.7642 | 4.1538 | 27 | 0.9418 | 0.8951 | 0.8456 |
80
+ | 0.4882 | 4.3077 | 28 | 0.9095 | 0.9310 | 0.9265 |
81
+ | 0.5011 | 4.4615 | 29 | 0.9378 | 0.8426 | 0.875 |
82
+ | 0.3707 | 4.6154 | 30 | 0.9630 | 0.7963 | 0.8382 |
83
+ | 0.381 | 4.7692 | 31 | 0.9721 | 0.7870 | 0.8309 |
84
+ | 0.2307 | 4.9231 | 32 | 0.9522 | 0.7963 | 0.8382 |
85
+ | 0.2829 | 5.0769 | 33 | 0.9598 | 0.7870 | 0.8309 |
86
+ | 0.2581 | 5.2308 | 34 | 0.9458 | 0.8056 | 0.8456 |
87
+ | 0.4658 | 5.3846 | 35 | 0.9442 | 0.8148 | 0.8529 |
88
+ | 0.2133 | 5.5385 | 36 | 0.9524 | 0.7870 | 0.8309 |
89
+ | 0.1107 | 5.6923 | 37 | 0.9601 | 0.7870 | 0.8309 |
90
+ | 0.3599 | 5.8462 | 38 | 0.9605 | 0.7778 | 0.8235 |
91
+ | 0.3085 | 6.0 | 39 | 0.9522 | 0.7918 | 0.8309 |
92
+ | 0.2739 | 6.1538 | 40 | 0.9564 | 0.7870 | 0.8309 |
93
+ | 0.3279 | 6.3077 | 41 | 0.9582 | 0.7870 | 0.8309 |
94
+ | 0.1346 | 6.4615 | 42 | 0.9646 | 0.7685 | 0.8162 |
95
+ | 0.1429 | 6.6154 | 43 | 0.9695 | 0.7685 | 0.8162 |
96
+ | 0.1 | 6.7692 | 44 | 0.9692 | 0.7685 | 0.8162 |
97
+ | 0.1852 | 6.9231 | 45 | 0.9651 | 0.7685 | 0.8162 |
98
+ | 0.1028 | 7.0769 | 46 | 0.9378 | 0.8056 | 0.8456 |
99
+ | 0.2071 | 7.2308 | 47 | 0.9154 | 0.8195 | 0.8529 |
100
+ | 0.1752 | 7.3846 | 48 | 0.8882 | 0.8566 | 0.8824 |
101
+ | 0.0907 | 7.5385 | 49 | 0.8704 | 0.8843 | 0.9044 |
102
+ | 0.1263 | 7.6923 | 50 | 0.8719 | 0.8798 | 0.8971 |
103
+ | 0.068 | 7.8462 | 51 | 0.8738 | 0.8798 | 0.8971 |
104
+ | 0.0589 | 8.0 | 52 | 0.8881 | 0.8566 | 0.8824 |
105
+ | 0.1494 | 8.1538 | 53 | 0.9001 | 0.8473 | 0.875 |
106
+ | 0.1137 | 8.3077 | 54 | 0.9120 | 0.8288 | 0.8603 |
107
+ | 0.0522 | 8.4615 | 55 | 0.9212 | 0.8148 | 0.8529 |
108
+ | 0.0666 | 8.6154 | 56 | 0.9251 | 0.8148 | 0.8529 |
109
+ | 0.0867 | 8.7692 | 57 | 0.9270 | 0.8148 | 0.8529 |
110
+ | 0.0764 | 8.9231 | 58 | 0.9264 | 0.8148 | 0.8529 |
111
+ | 0.0526 | 9.0769 | 59 | 0.9259 | 0.8148 | 0.8529 |
112
+ | 0.2877 | 9.2308 | 60 | 0.9254 | 0.8148 | 0.8529 |
113
+
114
+
115
+ ### Framework versions
116
+
117
+ - Transformers 4.40.1
118
+ - Pytorch 2.3.0+cu121
119
+ - Datasets 2.19.0
120
+ - Tokenizers 0.19.1
model.safetensors CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:43d34aeb6935ba2e5afcf5614494f437a4f1dbbac3796ae25a03e209a682fb1d
3
  size 379887552
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:cf1bc6d5a92c70de8c0cfc584c74abab649b4621c9218734b33a568f02082c27
3
  size 379887552
torch/config.json ADDED
@@ -0,0 +1,136 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_name_or_path": "microsoft/wavlm-base",
3
+ "activation_dropout": 0.0,
4
+ "adapter_kernel_size": 3,
5
+ "adapter_stride": 2,
6
+ "add_adapter": false,
7
+ "apply_spec_augment": true,
8
+ "architectures": [
9
+ "Model"
10
+ ],
11
+ "attention_dropout": 0.1,
12
+ "bos_token_id": 1,
13
+ "classifier_proj_size": 256,
14
+ "codevector_dim": 256,
15
+ "contrastive_logits_temperature": 0.1,
16
+ "conv_bias": false,
17
+ "conv_dim": [
18
+ 512,
19
+ 512,
20
+ 512,
21
+ 512,
22
+ 512,
23
+ 512,
24
+ 512
25
+ ],
26
+ "conv_kernel": [
27
+ 10,
28
+ 3,
29
+ 3,
30
+ 3,
31
+ 3,
32
+ 2,
33
+ 2
34
+ ],
35
+ "conv_stride": [
36
+ 5,
37
+ 2,
38
+ 2,
39
+ 2,
40
+ 2,
41
+ 2,
42
+ 2
43
+ ],
44
+ "ctc_loss_reduction": "sum",
45
+ "ctc_zero_infinity": false,
46
+ "data": "emodb",
47
+ "diversity_loss_weight": 0.1,
48
+ "do_stable_layer_norm": false,
49
+ "eos_token_id": 2,
50
+ "feat_extract_activation": "gelu",
51
+ "feat_extract_norm": "group",
52
+ "feat_proj_dropout": 0.1,
53
+ "feat_quantizer_dropout": 0.0,
54
+ "final_dropout": 0.0,
55
+ "finetuning_task": "emotion",
56
+ "freeze_feat_extract_train": true,
57
+ "hidden_act": "gelu",
58
+ "hidden_dropout": 0.1,
59
+ "hidden_size": 768,
60
+ "id2label": {
61
+ "0": "anger",
62
+ "1": "happiness",
63
+ "2": "neutral",
64
+ "3": "sadness"
65
+ },
66
+ "initializer_range": 0.02,
67
+ "intermediate_size": 3072,
68
+ "label2id": {
69
+ "anger": 0,
70
+ "happiness": 1,
71
+ "neutral": 2,
72
+ "sadness": 3
73
+ },
74
+ "layer_norm_eps": 1e-05,
75
+ "layerdrop": 0.05,
76
+ "mask_channel_length": 10,
77
+ "mask_channel_min_space": 1,
78
+ "mask_channel_other": 0.0,
79
+ "mask_channel_prob": 0.0,
80
+ "mask_channel_selection": "static",
81
+ "mask_feature_length": 10,
82
+ "mask_feature_min_masks": 0,
83
+ "mask_feature_prob": 0.0,
84
+ "mask_time_length": 10,
85
+ "mask_time_min_masks": 2,
86
+ "mask_time_min_space": 1,
87
+ "mask_time_other": 0.0,
88
+ "mask_time_prob": 0.05,
89
+ "mask_time_selection": "static",
90
+ "max_bucket_distance": 800,
91
+ "model_type": "wavlm",
92
+ "no_mask_channel_overlap": false,
93
+ "no_mask_time_overlap": false,
94
+ "num_adapter_layers": 3,
95
+ "num_attention_heads": 12,
96
+ "num_buckets": 320,
97
+ "num_codevector_groups": 2,
98
+ "num_codevectors_per_group": 320,
99
+ "num_conv_pos_embedding_groups": 16,
100
+ "num_conv_pos_embeddings": 128,
101
+ "num_ctc_classes": 80,
102
+ "num_feat_extract_layers": 7,
103
+ "num_hidden_layers": 12,
104
+ "num_negatives": 100,
105
+ "output_hidden_size": 768,
106
+ "pad_token_id": 0,
107
+ "proj_codevector_dim": 256,
108
+ "sampling_rate": 16000,
109
+ "tdnn_dilation": [
110
+ 1,
111
+ 2,
112
+ 3,
113
+ 1,
114
+ 1
115
+ ],
116
+ "tdnn_dim": [
117
+ 512,
118
+ 512,
119
+ 512,
120
+ 512,
121
+ 1500
122
+ ],
123
+ "tdnn_kernel": [
124
+ 5,
125
+ 3,
126
+ 3,
127
+ 1,
128
+ 1
129
+ ],
130
+ "tokenizer_class": "Wav2Vec2CTCTokenizer",
131
+ "torch_dtype": "float32",
132
+ "transformers_version": "4.40.1",
133
+ "use_weighted_layer_sum": false,
134
+ "vocab_size": 32,
135
+ "xvector_output_dim": 512
136
+ }
torch/model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:cf1bc6d5a92c70de8c0cfc584c74abab649b4621c9218734b33a568f02082c27
3
+ size 379887552
torch/preprocessor_config.json ADDED
@@ -0,0 +1,9 @@
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "do_normalize": true,
3
+ "feature_extractor_type": "Wav2Vec2FeatureExtractor",
4
+ "feature_size": 1,
5
+ "padding_side": "right",
6
+ "padding_value": 0.0,
7
+ "return_attention_mask": true,
8
+ "sampling_rate": 16000
9
+ }
torch/training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:b732ecef99958b88bb71462c8ab8d8e6971a7a86a76c2e133387afcd69fdb120
3
+ size 4984