Hugging Face's logo

ALBADDAWI
/
ft-wav2vec2-with-minds

Safetensors
wav2vec2
Generated from Trainer
Model card Files Community
ALBADDAWI commited on
Commit
dc58e92
·
verified ·
1 Parent(s): c3ec049

ft-wav2vec2-with-minds

Browse files
Files changed (4) hide show
  1. README.md +113 -0
  2. config.json +133 -0
  3. model.safetensors +3 -0
  4. training_args.bin +3 -0
README.md ADDED
@@ -0,0 +1,113 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ license: cc-by-nc-sa-4.0
3
+ base_model: audeering/wav2vec2-large-robust-12-ft-emotion-msp-dim
4
+ tags:
5
+ - generated_from_trainer
6
+ metrics:
7
+ - accuracy
8
+ model-index:
9
+ - name: ft-wav2vec2-with-minds
10
+ results: []
11
+ ---
12
+
13
+ <!-- This model card has been generated automatically according to the information the Trainer had access to. You
14
+ should probably proofread and complete it, then remove this comment. -->
15
+
16
+ # ft-wav2vec2-with-minds
17
+
18
+ This model is a fine-tuned version of [audeering/wav2vec2-large-robust-12-ft-emotion-msp-dim](https://huggingface.co/audeering/wav2vec2-large-robust-12-ft-emotion-msp-dim) on the None dataset.
19
+ It achieves the following results on the evaluation set:
20
+ - Loss: 0.2564
21
+ - Accuracy: 0.9400
22
+
23
+ ## Model description
24
+
25
+ More information needed
26
+
27
+ ## Intended uses & limitations
28
+
29
+ More information needed
30
+
31
+ ## Training and evaluation data
32
+
33
+ More information needed
34
+
35
+ ## Training procedure
36
+
37
+ ### Training hyperparameters
38
+
39
+ The following hyperparameters were used during training:
40
+ - learning_rate: 4e-05
41
+ - train_batch_size: 120
42
+ - eval_batch_size: 120
43
+ - seed: 42
44
+ - gradient_accumulation_steps: 4
45
+ - total_train_batch_size: 480
46
+ - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
47
+ - lr_scheduler_type: linear
48
+ - lr_scheduler_warmup_ratio: 0.1
49
+ - num_epochs: 50
50
+ - mixed_precision_training: Native AMP
51
+
52
+ ### Training results
53
+
54
+ | Training Loss | Epoch | Step | Validation Loss | Accuracy |
55
+ |:-------------:|:-----:|:----:|:---------------:|:--------:|
56
+ | 2.0796 | 1.0 | 9 | 2.0809 | 0.1209 |
57
+ | 2.0789 | 2.0 | 18 | 2.0779 | 0.1406 |
58
+ | 2.0762 | 3.0 | 27 | 2.0724 | 0.1987 |
59
+ | 2.0712 | 4.0 | 36 | 2.0627 | 0.2315 |
60
+ | 2.0601 | 5.0 | 45 | 2.0423 | 0.3327 |
61
+ | 2.0489 | 6.0 | 54 | 1.9888 | 0.5145 |
62
+ | 2.0094 | 7.0 | 63 | 1.8840 | 0.6214 |
63
+ | 1.9088 | 8.0 | 72 | 1.7428 | 0.6429 |
64
+ | 1.7904 | 9.0 | 81 | 1.5916 | 0.6448 |
65
+ | 1.6668 | 10.0 | 90 | 1.4391 | 0.7029 |
66
+ | 1.5889 | 11.0 | 99 | 1.3026 | 0.7591 |
67
+ | 1.4522 | 12.0 | 108 | 1.1715 | 0.7901 |
68
+ | 1.3301 | 13.0 | 117 | 1.0506 | 0.8257 |
69
+ | 1.2325 | 14.0 | 126 | 0.9515 | 0.8472 |
70
+ | 1.1669 | 15.0 | 135 | 0.8527 | 0.8557 |
71
+ | 1.0915 | 16.0 | 144 | 0.7745 | 0.8697 |
72
+ | 1.0157 | 17.0 | 153 | 0.7060 | 0.8772 |
73
+ | 0.9657 | 18.0 | 162 | 0.6602 | 0.8744 |
74
+ | 0.8975 | 19.0 | 171 | 0.6002 | 0.8903 |
75
+ | 0.8403 | 20.0 | 180 | 0.5651 | 0.8932 |
76
+ | 0.8059 | 21.0 | 189 | 0.5243 | 0.8960 |
77
+ | 0.731 | 22.0 | 198 | 0.4860 | 0.9044 |
78
+ | 0.7139 | 23.0 | 207 | 0.4634 | 0.9044 |
79
+ | 0.6903 | 24.0 | 216 | 0.4450 | 0.9082 |
80
+ | 0.6597 | 25.0 | 225 | 0.4221 | 0.9072 |
81
+ | 0.6146 | 26.0 | 234 | 0.4013 | 0.9166 |
82
+ | 0.6162 | 27.0 | 243 | 0.3853 | 0.9119 |
83
+ | 0.6252 | 28.0 | 252 | 0.3886 | 0.9100 |
84
+ | 0.5666 | 29.0 | 261 | 0.3478 | 0.9269 |
85
+ | 0.5698 | 30.0 | 270 | 0.3489 | 0.9250 |
86
+ | 0.5575 | 31.0 | 279 | 0.3354 | 0.9260 |
87
+ | 0.5298 | 32.0 | 288 | 0.3299 | 0.9203 |
88
+ | 0.5267 | 33.0 | 297 | 0.3128 | 0.9297 |
89
+ | 0.5558 | 34.0 | 306 | 0.3070 | 0.9316 |
90
+ | 0.5541 | 35.0 | 315 | 0.3005 | 0.9335 |
91
+ | 0.5328 | 36.0 | 324 | 0.2908 | 0.9363 |
92
+ | 0.5566 | 37.0 | 333 | 0.2923 | 0.9325 |
93
+ | 0.5184 | 38.0 | 342 | 0.2825 | 0.9363 |
94
+ | 0.4649 | 39.0 | 351 | 0.2739 | 0.9391 |
95
+ | 0.431 | 40.0 | 360 | 0.2698 | 0.9335 |
96
+ | 0.4681 | 41.0 | 369 | 0.2643 | 0.9372 |
97
+ | 0.4918 | 42.0 | 378 | 0.2611 | 0.9372 |
98
+ | 0.4688 | 43.0 | 387 | 0.2608 | 0.9381 |
99
+ | 0.4738 | 44.0 | 396 | 0.2621 | 0.9372 |
100
+ | 0.4669 | 45.0 | 405 | 0.2604 | 0.9381 |
101
+ | 0.4556 | 46.0 | 414 | 0.2596 | 0.9344 |
102
+ | 0.4498 | 47.0 | 423 | 0.2564 | 0.9400 |
103
+ | 0.4738 | 48.0 | 432 | 0.2564 | 0.9400 |
104
+ | 0.4494 | 49.0 | 441 | 0.2564 | 0.9391 |
105
+ | 0.447 | 50.0 | 450 | 0.2564 | 0.9391 |
106
+
107
+
108
+ ### Framework versions
109
+
110
+ - Transformers 4.35.2
111
+ - Pytorch 1.12.1+cu116
112
+ - Datasets 2.15.0
113
+ - Tokenizers 0.15.2
config.json ADDED
@@ -0,0 +1,133 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_name_or_path": "audeering/wav2vec2-large-robust-12-ft-emotion-msp-dim",
3
+ "activation_dropout": 0.1,
4
+ "adapter_attn_dim": null,
5
+ "adapter_kernel_size": 3,
6
+ "adapter_stride": 2,
7
+ "add_adapter": false,
8
+ "apply_spec_augment": true,
9
+ "architectures": [
10
+ "Wav2Vec2ForSequenceClassification"
11
+ ],
12
+ "attention_dropout": 0.1,
13
+ "bos_token_id": 1,
14
+ "classifier_proj_size": 256,
15
+ "codevector_dim": 768,
16
+ "contrastive_logits_temperature": 0.1,
17
+ "conv_bias": true,
18
+ "conv_dim": [
19
+ 512,
20
+ 512,
21
+ 512,
22
+ 512,
23
+ 512,
24
+ 512,
25
+ 512
26
+ ],
27
+ "conv_kernel": [
28
+ 10,
29
+ 3,
30
+ 3,
31
+ 3,
32
+ 3,
33
+ 2,
34
+ 2
35
+ ],
36
+ "conv_stride": [
37
+ 5,
38
+ 2,
39
+ 2,
40
+ 2,
41
+ 2,
42
+ 2,
43
+ 2
44
+ ],
45
+ "ctc_loss_reduction": "sum",
46
+ "ctc_zero_infinity": false,
47
+ "diversity_loss_weight": 0.1,
48
+ "do_stable_layer_norm": true,
49
+ "eos_token_id": 2,
50
+ "feat_extract_activation": "gelu",
51
+ "feat_extract_dropout": 0.0,
52
+ "feat_extract_norm": "layer",
53
+ "feat_proj_dropout": 0.1,
54
+ "feat_quantizer_dropout": 0.0,
55
+ "final_dropout": 0.1,
56
+ "finetuning_task": "wav2vec2_reg",
57
+ "gradient_checkpointing": false,
58
+ "hidden_act": "gelu",
59
+ "hidden_dropout": 0.1,
60
+ "hidden_dropout_prob": 0.1,
61
+ "hidden_size": 1024,
62
+ "id2label": {
63
+ "0": "stop",
64
+ "1": "no",
65
+ "2": "left",
66
+ "3": "right",
67
+ "4": "yes",
68
+ "5": "up",
69
+ "6": "down",
70
+ "7": "go"
71
+ },
72
+ "initializer_range": 0.02,
73
+ "intermediate_size": 4096,
74
+ "label2id": {
75
+ "down": "6",
76
+ "go": "7",
77
+ "left": "2",
78
+ "no": "1",
79
+ "right": "3",
80
+ "stop": "0",
81
+ "up": "5",
82
+ "yes": "4"
83
+ },
84
+ "layer_norm_eps": 1e-05,
85
+ "layerdrop": 0.1,
86
+ "mask_feature_length": 10,
87
+ "mask_feature_min_masks": 0,
88
+ "mask_feature_prob": 0.0,
89
+ "mask_time_length": 10,
90
+ "mask_time_min_masks": 2,
91
+ "mask_time_prob": 0.05,
92
+ "model_type": "wav2vec2",
93
+ "num_adapter_layers": 3,
94
+ "num_attention_heads": 16,
95
+ "num_codevector_groups": 2,
96
+ "num_codevectors_per_group": 320,
97
+ "num_conv_pos_embedding_groups": 16,
98
+ "num_conv_pos_embeddings": 128,
99
+ "num_feat_extract_layers": 7,
100
+ "num_hidden_layers": 12,
101
+ "num_negatives": 100,
102
+ "output_hidden_size": 1024,
103
+ "pad_token_id": 0,
104
+ "pooling_mode": "mean",
105
+ "problem_type": "regression",
106
+ "proj_codevector_dim": 768,
107
+ "tdnn_dilation": [
108
+ 1,
109
+ 2,
110
+ 3,
111
+ 1,
112
+ 1
113
+ ],
114
+ "tdnn_dim": [
115
+ 512,
116
+ 512,
117
+ 512,
118
+ 512,
119
+ 1500
120
+ ],
121
+ "tdnn_kernel": [
122
+ 5,
123
+ 3,
124
+ 3,
125
+ 1,
126
+ 1
127
+ ],
128
+ "torch_dtype": "float32",
129
+ "transformers_version": "4.35.2",
130
+ "use_weighted_layer_sum": false,
131
+ "vocab_size": null,
132
+ "xvector_output_dim": 512
133
+ }
model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:144f3ffc9e8cde213bd94555aca38e1ffb5dfb8bc507afc2a1856997f00d139b
3
+ size 658222576
training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:e70e7084200ada453ee3d0ff5ff6e337686bbeaadaa5472f19d09722832ad5e6
3
+ size 4143