Ar4ikov commited on
Commit
5cdcd0a
1 Parent(s): 8316970

Upload UniSpeechSatForSequenceClassification

Browse files
Files changed (2) hide show
  1. config.json +127 -0
  2. pytorch_model.bin +3 -0
config.json ADDED
@@ -0,0 +1,127 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_name_or_path": "Aniemore/unispeech-sat-emotion-russian-resd",
3
+ "activation_dropout": 0.05,
4
+ "apply_spec_augment": true,
5
+ "architectures": [
6
+ "UniSpeechSatForSequenceClassification"
7
+ ],
8
+ "attention_dropout": 0.05,
9
+ "bos_token_id": 1,
10
+ "classifier_proj_size": 768,
11
+ "codevector_dim": 768,
12
+ "contrastive_logits_temperature": 0.1,
13
+ "conv_bias": false,
14
+ "conv_dim": [
15
+ 512,
16
+ 512,
17
+ 512,
18
+ 512,
19
+ 512,
20
+ 512,
21
+ 512
22
+ ],
23
+ "conv_kernel": [
24
+ 10,
25
+ 3,
26
+ 3,
27
+ 3,
28
+ 3,
29
+ 2,
30
+ 2
31
+ ],
32
+ "conv_stride": [
33
+ 5,
34
+ 2,
35
+ 2,
36
+ 2,
37
+ 2,
38
+ 2,
39
+ 2
40
+ ],
41
+ "ctc_loss_reduction": "sum",
42
+ "ctc_zero_infinity": false,
43
+ "diversity_loss_weight": 0.1,
44
+ "do_stable_layer_norm": true,
45
+ "eos_token_id": 2,
46
+ "feat_extract_activation": "gelu",
47
+ "feat_extract_dropout": 0.0,
48
+ "feat_extract_norm": "layer",
49
+ "feat_proj_dropout": 0.05,
50
+ "feat_quantizer_dropout": 0.0,
51
+ "final_dropout": 0.05,
52
+ "finetuning_task": [
53
+ "unispeech_sat_classification"
54
+ ],
55
+ "hidden_act": "gelu",
56
+ "hidden_dropout": 0.05,
57
+ "hidden_size": 1024,
58
+ "id2label": {
59
+ "0": "anger",
60
+ "1": "disgust",
61
+ "2": "enthusiasm",
62
+ "3": "fear",
63
+ "4": "happiness",
64
+ "5": "neutral",
65
+ "6": "sadness"
66
+ },
67
+ "initializer_range": 0.02,
68
+ "intermediate_size": 4096,
69
+ "label2id": {
70
+ "anger": 0,
71
+ "disgust": 1,
72
+ "enthusiasm": 2,
73
+ "fear": 3,
74
+ "happiness": 4,
75
+ "neutral": 5,
76
+ "sadness": 6
77
+ },
78
+ "layer_norm_eps": 1e-05,
79
+ "layerdrop": 0.05,
80
+ "mask_feature_length": 10,
81
+ "mask_feature_min_masks": 0,
82
+ "mask_feature_prob": 0.0,
83
+ "mask_time_length": 10,
84
+ "mask_time_min_masks": 2,
85
+ "mask_time_prob": 0.05,
86
+ "model_type": "unispeech-sat",
87
+ "num_attention_heads": 16,
88
+ "num_clusters": 504,
89
+ "num_codevector_groups": 2,
90
+ "num_codevectors_per_group": 320,
91
+ "num_conv_pos_embedding_groups": 16,
92
+ "num_conv_pos_embeddings": 128,
93
+ "num_feat_extract_layers": 7,
94
+ "num_hidden_layers": 24,
95
+ "num_negatives": 100,
96
+ "pad_token_id": 0,
97
+ "pooling_mode": "mean",
98
+ "problem_type": "single_label_classification",
99
+ "proj_codevector_dim": 768,
100
+ "tdnn_dilation": [
101
+ 1,
102
+ 2,
103
+ 3,
104
+ 1,
105
+ 1
106
+ ],
107
+ "tdnn_dim": [
108
+ 512,
109
+ 512,
110
+ 512,
111
+ 512,
112
+ 1500
113
+ ],
114
+ "tdnn_kernel": [
115
+ 5,
116
+ 3,
117
+ 3,
118
+ 1,
119
+ 1
120
+ ],
121
+ "tokenizer_class": "Wav2Vec2CTCTokenizer",
122
+ "torch_dtype": "float32",
123
+ "transformers_version": "4.25.1",
124
+ "use_weighted_layer_sum": false,
125
+ "vocab_size": 40,
126
+ "xvector_output_dim": 512
127
+ }
pytorch_model.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:275190b27abdb699df97212afea59bba1ce52c5e7b5c0d5e59156e7a52101c2e
3
+ size 1265052313