artyomboyko commited on
Commit
d994599
1 Parent(s): 52114a8

Training in progress, epoch 0

Browse files
Files changed (5) hide show
  1. README.md +92 -0
  2. config.json +146 -0
  3. preprocessor_config.json +9 -0
  4. pytorch_model.bin +3 -0
  5. training_args.bin +3 -0
README.md ADDED
@@ -0,0 +1,92 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ license: apache-2.0
3
+ base_model: facebook/wav2vec2-base
4
+ tags:
5
+ - generated_from_trainer
6
+ datasets:
7
+ - superb
8
+ metrics:
9
+ - accuracy
10
+ - recall
11
+ - f1
12
+ model-index:
13
+ - name: wav2vec2-base-finetuned-ks
14
+ results:
15
+ - task:
16
+ name: Audio Classification
17
+ type: audio-classification
18
+ dataset:
19
+ name: superb
20
+ type: superb
21
+ config: ks
22
+ split: validation
23
+ args: ks
24
+ metrics:
25
+ - name: Accuracy
26
+ type: accuracy
27
+ value: 0.9832303618711385
28
+ - name: Recall
29
+ type: recall
30
+ value: 0.9664413018718482
31
+ - name: F1
32
+ type: f1
33
+ value: 0.9719648106690262
34
+ ---
35
+
36
+ <!-- This model card has been generated automatically according to the information the Trainer had access to. You
37
+ should probably proofread and complete it, then remove this comment. -->
38
+
39
+ # wav2vec2-base-finetuned-ks
40
+
41
+ This model is a fine-tuned version of [facebook/wav2vec2-base](https://huggingface.co/facebook/wav2vec2-base) on the superb dataset.
42
+ It achieves the following results on the evaluation set:
43
+ - Loss: 0.0711
44
+ - Accuracy: 0.9832
45
+ - Recall: 0.9664
46
+ - F1: 0.9720
47
+
48
+ ## Model description
49
+
50
+ More information needed
51
+
52
+ ## Intended uses & limitations
53
+
54
+ More information needed
55
+
56
+ ## Training and evaluation data
57
+
58
+ More information needed
59
+
60
+ ## Training procedure
61
+
62
+ ### Training hyperparameters
63
+
64
+ The following hyperparameters were used during training:
65
+ - learning_rate: 3e-05
66
+ - train_batch_size: 16
67
+ - eval_batch_size: 16
68
+ - seed: 42
69
+ - gradient_accumulation_steps: 4
70
+ - total_train_batch_size: 64
71
+ - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
72
+ - lr_scheduler_type: linear
73
+ - lr_scheduler_warmup_ratio: 0.1
74
+ - num_epochs: 5
75
+
76
+ ### Training results
77
+
78
+ | Training Loss | Epoch | Step | Validation Loss | Accuracy | Recall | F1 |
79
+ |:-------------:|:-----:|:----:|:---------------:|:--------:|:------:|:------:|
80
+ | 0.4397 | 1.0 | 798 | 0.2810 | 0.9651 | 0.9289 | 0.9361 |
81
+ | 0.2067 | 2.0 | 1597 | 0.1142 | 0.9769 | 0.9536 | 0.9593 |
82
+ | 0.1881 | 3.0 | 2395 | 0.0829 | 0.9821 | 0.9644 | 0.9693 |
83
+ | 0.1167 | 4.0 | 3194 | 0.0752 | 0.9831 | 0.9644 | 0.9726 |
84
+ | 0.13 | 5.0 | 3990 | 0.0711 | 0.9832 | 0.9664 | 0.9720 |
85
+
86
+
87
+ ### Framework versions
88
+
89
+ - Transformers 4.34.1
90
+ - Pytorch 2.1.0+cu121
91
+ - Datasets 2.14.6
92
+ - Tokenizers 0.14.1
config.json ADDED
@@ -0,0 +1,146 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_name_or_path": "facebook/wav2vec2-base",
3
+ "activation_dropout": 0.0,
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": 256,
16
+ "contrastive_logits_temperature": 0.1,
17
+ "conv_bias": false,
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": 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
+ "freeze_feat_extract_train": true,
56
+ "hidden_act": "gelu",
57
+ "hidden_dropout": 0.1,
58
+ "hidden_size": 768,
59
+ "id2label": {
60
+ "0": "yes",
61
+ "1": "no",
62
+ "10": "_silence_",
63
+ "11": "_unknown_",
64
+ "2": "up",
65
+ "3": "down",
66
+ "4": "left",
67
+ "5": "right",
68
+ "6": "on",
69
+ "7": "off",
70
+ "8": "stop",
71
+ "9": "go"
72
+ },
73
+ "initializer_range": 0.02,
74
+ "intermediate_size": 3072,
75
+ "label2id": {
76
+ "_silence_": "10",
77
+ "_unknown_": "11",
78
+ "down": "3",
79
+ "go": "9",
80
+ "left": "4",
81
+ "no": "1",
82
+ "off": "7",
83
+ "on": "6",
84
+ "right": "5",
85
+ "stop": "8",
86
+ "up": "2",
87
+ "yes": "0"
88
+ },
89
+ "layer_norm_eps": 1e-05,
90
+ "layerdrop": 0.0,
91
+ "mask_channel_length": 10,
92
+ "mask_channel_min_space": 1,
93
+ "mask_channel_other": 0.0,
94
+ "mask_channel_prob": 0.0,
95
+ "mask_channel_selection": "static",
96
+ "mask_feature_length": 10,
97
+ "mask_feature_min_masks": 0,
98
+ "mask_feature_prob": 0.0,
99
+ "mask_time_length": 10,
100
+ "mask_time_min_masks": 2,
101
+ "mask_time_min_space": 1,
102
+ "mask_time_other": 0.0,
103
+ "mask_time_prob": 0.05,
104
+ "mask_time_selection": "static",
105
+ "model_type": "wav2vec2",
106
+ "no_mask_channel_overlap": false,
107
+ "no_mask_time_overlap": false,
108
+ "num_adapter_layers": 3,
109
+ "num_attention_heads": 12,
110
+ "num_codevector_groups": 2,
111
+ "num_codevectors_per_group": 320,
112
+ "num_conv_pos_embedding_groups": 16,
113
+ "num_conv_pos_embeddings": 128,
114
+ "num_feat_extract_layers": 7,
115
+ "num_hidden_layers": 12,
116
+ "num_negatives": 100,
117
+ "output_hidden_size": 768,
118
+ "pad_token_id": 0,
119
+ "proj_codevector_dim": 256,
120
+ "tdnn_dilation": [
121
+ 1,
122
+ 2,
123
+ 3,
124
+ 1,
125
+ 1
126
+ ],
127
+ "tdnn_dim": [
128
+ 512,
129
+ 512,
130
+ 512,
131
+ 512,
132
+ 1500
133
+ ],
134
+ "tdnn_kernel": [
135
+ 5,
136
+ 3,
137
+ 3,
138
+ 1,
139
+ 1
140
+ ],
141
+ "torch_dtype": "float32",
142
+ "transformers_version": "4.34.1",
143
+ "use_weighted_layer_sum": false,
144
+ "vocab_size": 32,
145
+ "xvector_output_dim": 512
146
+ }
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": false,
8
+ "sampling_rate": 16000
9
+ }
pytorch_model.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:9e7024887eca1d343329a04d0ecc1d9ecd850b30d5d5de34ab55082b6dcc2183
3
+ size 378361190
training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:aed92bf4faea04afceb24959ee3aa6ca272ac8e971117fc2a61b1eb67b5b7d2f
3
+ size 4536