ImanNalia commited on
Commit
e009359
1 Parent(s): c6d2932

exp1_wav2vec2_base

Browse files
README.md ADDED
@@ -0,0 +1,65 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ license: apache-2.0
3
+ base_model: facebook/wav2vec2-base
4
+ tags:
5
+ - generated_from_trainer
6
+ metrics:
7
+ - wer
8
+ model-index:
9
+ - name: wav2vec2
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
+ # wav2vec2
17
+
18
+ This model is a fine-tuned version of [facebook/wav2vec2-base](https://huggingface.co/facebook/wav2vec2-base) on the None dataset.
19
+ It achieves the following results on the evaluation set:
20
+ - Loss: 1.4145
21
+ - Wer: 0.4598
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: 0.0001
41
+ - train_batch_size: 32
42
+ - eval_batch_size: 16
43
+ - seed: 42
44
+ - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
45
+ - lr_scheduler_type: linear
46
+ - lr_scheduler_warmup_steps: 2000
47
+ - num_epochs: 50
48
+
49
+ ### Training results
50
+
51
+ | Training Loss | Epoch | Step | Validation Loss | Wer |
52
+ |:-------------:|:-----:|:----:|:---------------:|:------:|
53
+ | 4.245 | 9.26 | 500 | 2.9303 | 0.9871 |
54
+ | 2.5133 | 18.52 | 1000 | 1.3210 | 0.6705 |
55
+ | 0.6779 | 27.78 | 1500 | 1.2663 | 0.4949 |
56
+ | 0.3209 | 37.04 | 2000 | 1.3584 | 0.4872 |
57
+ | 0.19 | 46.3 | 2500 | 1.4145 | 0.4598 |
58
+
59
+
60
+ ### Framework versions
61
+
62
+ - Transformers 4.39.2
63
+ - Pytorch 1.12.1+cu116
64
+ - Datasets 2.18.0
65
+ - Tokenizers 0.15.2
config.json ADDED
@@ -0,0 +1,119 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
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
+ "Wav2Vec2ForCTC"
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": "mean",
46
+ "ctc_zero_infinity": false,
47
+ "diversity_loss_weight": 0.1,
48
+ "do_stable_layer_norm": false,
49
+ "dropout": 0.2,
50
+ "eos_token_id": 2,
51
+ "feat_extract_activation": "gelu",
52
+ "feat_extract_norm": "group",
53
+ "feat_proj_dropout": 0.1,
54
+ "feat_quantizer_dropout": 0.0,
55
+ "final_dropout": 0.0,
56
+ "freeze_feat_extract_train": true,
57
+ "hidden_act": "gelu",
58
+ "hidden_dropout": 0.1,
59
+ "hidden_size": 768,
60
+ "initializer_range": 0.02,
61
+ "intermediate_size": 3072,
62
+ "layer_norm_eps": 1e-05,
63
+ "layerdrop": 0.0,
64
+ "mask_channel_length": 10,
65
+ "mask_channel_min_space": 1,
66
+ "mask_channel_other": 0.0,
67
+ "mask_channel_prob": 0.0,
68
+ "mask_channel_selection": "static",
69
+ "mask_feature_length": 10,
70
+ "mask_feature_min_masks": 0,
71
+ "mask_feature_prob": 0.0,
72
+ "mask_time_length": 10,
73
+ "mask_time_min_masks": 2,
74
+ "mask_time_min_space": 1,
75
+ "mask_time_other": 0.0,
76
+ "mask_time_prob": 0.05,
77
+ "mask_time_selection": "static",
78
+ "model_type": "wav2vec2",
79
+ "no_mask_channel_overlap": false,
80
+ "no_mask_time_overlap": false,
81
+ "num_adapter_layers": 3,
82
+ "num_attention_heads": 12,
83
+ "num_codevector_groups": 2,
84
+ "num_codevectors_per_group": 320,
85
+ "num_conv_pos_embedding_groups": 16,
86
+ "num_conv_pos_embeddings": 128,
87
+ "num_feat_extract_layers": 7,
88
+ "num_hidden_layers": 12,
89
+ "num_negatives": 100,
90
+ "output_hidden_size": 768,
91
+ "pad_token_id": 29,
92
+ "proj_codevector_dim": 256,
93
+ "tdnn_dilation": [
94
+ 1,
95
+ 2,
96
+ 3,
97
+ 1,
98
+ 1
99
+ ],
100
+ "tdnn_dim": [
101
+ 512,
102
+ 512,
103
+ 512,
104
+ 512,
105
+ 1500
106
+ ],
107
+ "tdnn_kernel": [
108
+ 5,
109
+ 3,
110
+ 3,
111
+ 1,
112
+ 1
113
+ ],
114
+ "torch_dtype": "float32",
115
+ "transformers_version": "4.39.2",
116
+ "use_weighted_layer_sum": false,
117
+ "vocab_size": 32,
118
+ "xvector_output_dim": 512
119
+ }
model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:f678e1fb283b31bc3146941ed703af7fc6a084ceef32e6e03bfad04252ddc883
3
+ size 377611072
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
+ }
runs/Apr02_01-17-56_nr4cl7nhqz/events.out.tfevents.1712021341.nr4cl7nhqz.190.0 ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:70c8aabe5ee6e9c91eef79fec2f7943bf4dbf1b218e54c45954a610a048c304c
3
+ size 6275
runs/Apr02_02-22-59_n9gmn6uc4r/events.out.tfevents.1712024623.n9gmn6uc4r.161.0 ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:802a966ae5939cf00dc3b62f2259e83476ebbd70a7695eabaa467d3c2569bf5e
3
+ size 9274
training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:ad05f4493cd740286d33952fab4c7e8d3c5adbe309a2c4a1c753e6c2e9da01b4
3
+ size 4463