zainulhakim commited on
Commit
f53efc8
·
verified ·
1 Parent(s): 4496a76

End of training

Browse files
README.md CHANGED
@@ -17,7 +17,7 @@ should probably proofread and complete it, then remove this comment. -->
17
 
18
  This model is a fine-tuned version of [facebook/wav2vec2-base-960h](https://huggingface.co/facebook/wav2vec2-base-960h) on the None dataset.
19
  It achieves the following results on the evaluation set:
20
- - Loss: 7247.8286
21
  - Wer: 1.0
22
  - Cer: 0.9981
23
 
@@ -40,7 +40,7 @@ More information needed
40
  The following hyperparameters were used during training:
41
  - learning_rate: 0.001
42
  - train_batch_size: 4
43
- - eval_batch_size: 8
44
  - seed: 42
45
  - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
46
  - lr_scheduler_type: linear
@@ -51,49 +51,49 @@ The following hyperparameters were used during training:
51
 
52
  | Training Loss | Epoch | Step | Validation Loss | Wer | Cer |
53
  |:-------------:|:-----:|:------:|:---------------:|:---:|:------:|
54
- | 3632.3885 | 1.0 | 2900 | 7247.8286 | 1.0 | 0.9981 |
55
- | 3631.472 | 2.0 | 5800 | 7247.4536 | 1.0 | 0.9981 |
56
- | 3634.3392 | 3.0 | 8700 | 7247.3838 | 1.0 | 0.9981 |
57
- | 3624.8195 | 4.0 | 11600 | 7247.3589 | 1.0 | 0.9981 |
58
- | 3628.096 | 5.0 | 14500 | 7247.3486 | 1.0 | 0.9981 |
59
- | 3625.4922 | 6.0 | 17400 | 7247.3848 | 1.0 | 0.9981 |
60
- | 3624.7733 | 7.0 | 20300 | 7247.3320 | 1.0 | 0.9981 |
61
- | 3625.468 | 8.0 | 23200 | 7247.3823 | 1.0 | 0.9981 |
62
- | 3624.3375 | 9.0 | 26100 | 7247.3701 | 1.0 | 0.9981 |
63
- | 3625.2198 | 10.0 | 29000 | 7247.3511 | 1.0 | 0.9981 |
64
- | 3624.0125 | 11.0 | 31900 | 7247.2944 | 1.0 | 0.9981 |
65
- | 3624.4412 | 12.0 | 34800 | 7247.3662 | 1.0 | 0.9981 |
66
- | 3624.0878 | 13.0 | 37700 | 7247.3164 | 1.0 | 0.9981 |
67
- | 3623.9817 | 14.0 | 40600 | 7247.2944 | 1.0 | 0.9981 |
68
- | 3624.541 | 15.0 | 43500 | 7247.5161 | 1.0 | 0.9981 |
69
- | 3624.0802 | 16.0 | 46400 | 7247.5713 | 1.0 | 0.9981 |
70
- | 3624.113 | 17.0 | 49300 | 7247.6650 | 1.0 | 0.9981 |
71
- | 3714.1267 | 18.0 | 52200 | nan | 1.0 | 0.9981 |
72
- | 0.0 | 19.0 | 55100 | nan | 1.0 | 0.9981 |
73
- | 0.0 | 20.0 | 58000 | nan | 1.0 | 0.9981 |
74
- | 0.0 | 21.0 | 60900 | nan | 1.0 | 0.9981 |
75
- | 0.0 | 22.0 | 63800 | nan | 1.0 | 0.9981 |
76
- | 0.0 | 23.0 | 66700 | nan | 1.0 | 0.9981 |
77
- | 0.0 | 24.0 | 69600 | nan | 1.0 | 0.9981 |
78
- | 0.0 | 25.0 | 72500 | nan | 1.0 | 0.9981 |
79
- | 0.0 | 26.0 | 75400 | nan | 1.0 | 0.9981 |
80
- | 0.0 | 27.0 | 78300 | nan | 1.0 | 0.9981 |
81
- | 0.0 | 28.0 | 81200 | nan | 1.0 | 0.9981 |
82
- | 0.0 | 29.0 | 84100 | nan | 1.0 | 0.9981 |
83
- | 0.0 | 30.0 | 87000 | nan | 1.0 | 0.9981 |
84
- | 0.0 | 31.0 | 89900 | nan | 1.0 | 0.9981 |
85
- | 0.0 | 32.0 | 92800 | nan | 1.0 | 0.9981 |
86
- | 0.0 | 33.0 | 95700 | nan | 1.0 | 0.9981 |
87
- | 0.0 | 34.0 | 98600 | nan | 1.0 | 0.9981 |
88
- | 0.0 | 35.0 | 101500 | nan | 1.0 | 0.9981 |
89
- | 0.0 | 36.0 | 104400 | nan | 1.0 | 0.9981 |
90
- | 0.0 | 37.0 | 107300 | nan | 1.0 | 0.9981 |
91
- | 0.0 | 38.0 | 110200 | nan | 1.0 | 0.9981 |
92
- | 0.0 | 39.0 | 113100 | nan | 1.0 | 0.9981 |
93
- | 0.0 | 40.0 | 116000 | nan | 1.0 | 0.9981 |
94
- | 0.0 | 41.0 | 118900 | nan | 1.0 | 0.9981 |
95
- | 0.0 | 42.0 | 121800 | nan | 1.0 | 0.9981 |
96
- | 0.0 | 43.0 | 124700 | nan | 1.0 | 0.9981 |
97
  | 0.0 | 44.0 | 127600 | nan | 1.0 | 0.9981 |
98
  | 0.0 | 45.0 | 130500 | nan | 1.0 | 0.9981 |
99
  | 0.0 | 46.0 | 133400 | nan | 1.0 | 0.9981 |
 
17
 
18
  This model is a fine-tuned version of [facebook/wav2vec2-base-960h](https://huggingface.co/facebook/wav2vec2-base-960h) on the None dataset.
19
  It achieves the following results on the evaluation set:
20
+ - Loss: 3624.4978
21
  - Wer: 1.0
22
  - Cer: 0.9981
23
 
 
40
  The following hyperparameters were used during training:
41
  - learning_rate: 0.001
42
  - train_batch_size: 4
43
+ - eval_batch_size: 4
44
  - seed: 42
45
  - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
46
  - lr_scheduler_type: linear
 
51
 
52
  | Training Loss | Epoch | Step | Validation Loss | Wer | Cer |
53
  |:-------------:|:-----:|:------:|:---------------:|:---:|:------:|
54
+ | 3768.9383 | 1.0 | 2900 | 3624.4978 | 1.0 | 0.9981 |
55
+ | 3664.7982 | 2.0 | 5800 | 3623.8293 | 1.0 | 0.9981 |
56
+ | 3628.479 | 3.0 | 8700 | 3623.8401 | 1.0 | 0.9981 |
57
+ | 3629.406 | 4.0 | 11600 | 3623.7019 | 1.0 | 0.9981 |
58
+ | 3625.2683 | 5.0 | 14500 | 3623.6763 | 1.0 | 0.9981 |
59
+ | 3624.6528 | 6.0 | 17400 | 3623.6826 | 1.0 | 0.9981 |
60
+ | 3624.5862 | 7.0 | 20300 | 3623.6912 | 1.0 | 0.9981 |
61
+ | 3624.5 | 8.0 | 23200 | 3623.6826 | 1.0 | 0.9981 |
62
+ | 3623.9722 | 9.0 | 26100 | 3623.6606 | 1.0 | 0.9981 |
63
+ | 3625.5082 | 10.0 | 29000 | 3623.7019 | 1.0 | 0.9981 |
64
+ | 3624.1595 | 11.0 | 31900 | 3623.6472 | 1.0 | 0.9981 |
65
+ | 3625.175 | 12.0 | 34800 | 3623.6831 | 1.0 | 0.9981 |
66
+ | 3625.192 | 13.0 | 37700 | 3623.6731 | 1.0 | 0.9981 |
67
+ | 3623.9662 | 14.0 | 40600 | 3623.6777 | 1.0 | 0.9981 |
68
+ | 3623.9175 | 15.0 | 43500 | 3623.6653 | 1.0 | 0.9981 |
69
+ | 3623.9675 | 16.0 | 46400 | 3623.6497 | 1.0 | 0.9981 |
70
+ | 3624.471 | 17.0 | 49300 | 3623.6682 | 1.0 | 0.9981 |
71
+ | 3623.9983 | 18.0 | 52200 | 3623.6560 | 1.0 | 0.9981 |
72
+ | 3624.317 | 19.0 | 55100 | 3623.6587 | 1.0 | 0.9981 |
73
+ | 3623.9965 | 20.0 | 58000 | 3623.6743 | 1.0 | 0.9981 |
74
+ | 3623.9733 | 21.0 | 60900 | 3623.6799 | 1.0 | 0.9981 |
75
+ | 3623.9163 | 22.0 | 63800 | 3623.6455 | 1.0 | 0.9981 |
76
+ | 3623.817 | 23.0 | 66700 | 3623.6487 | 1.0 | 0.9981 |
77
+ | 3623.8545 | 24.0 | 69600 | 3623.6912 | 1.0 | 0.9981 |
78
+ | 3623.9572 | 25.0 | 72500 | 3623.6763 | 1.0 | 0.9981 |
79
+ | 3624.0855 | 26.0 | 75400 | 3623.6682 | 1.0 | 0.9981 |
80
+ | 3623.9163 | 27.0 | 78300 | 3623.6519 | 1.0 | 0.9981 |
81
+ | 3623.813 | 28.0 | 81200 | 3623.6863 | 1.0 | 0.9981 |
82
+ | 3624.0905 | 29.0 | 84100 | 3623.6797 | 1.0 | 0.9981 |
83
+ | 3623.847 | 30.0 | 87000 | 3623.6743 | 1.0 | 0.9981 |
84
+ | 3624.7505 | 31.0 | 89900 | 3623.6768 | 1.0 | 0.9981 |
85
+ | 3623.794 | 32.0 | 92800 | 3623.6716 | 1.0 | 0.9981 |
86
+ | 3623.775 | 33.0 | 95700 | 3623.6409 | 1.0 | 0.9981 |
87
+ | 3624.021 | 34.0 | 98600 | 3623.6431 | 1.0 | 0.9981 |
88
+ | 3623.7987 | 35.0 | 101500 | 3623.6394 | 1.0 | 0.9981 |
89
+ | 3623.8223 | 36.0 | 104400 | 3623.6831 | 1.0 | 0.9981 |
90
+ | 3623.979 | 37.0 | 107300 | 3623.6785 | 1.0 | 0.9981 |
91
+ | 3623.809 | 38.0 | 110200 | 3623.6826 | 1.0 | 0.9981 |
92
+ | 3623.857 | 39.0 | 113100 | 3623.6643 | 1.0 | 0.9981 |
93
+ | 3623.8252 | 40.0 | 116000 | 3623.6731 | 1.0 | 0.9981 |
94
+ | 3623.916 | 41.0 | 118900 | 3623.6838 | 1.0 | 0.9981 |
95
+ | 3623.9533 | 42.0 | 121800 | 3623.6694 | 1.0 | 0.9981 |
96
+ | 3618.18 | 43.0 | 124700 | nan | 1.0 | 0.9981 |
97
  | 0.0 | 44.0 | 127600 | nan | 1.0 | 0.9981 |
98
  | 0.0 | 45.0 | 130500 | nan | 1.0 | 0.9981 |
99
  | 0.0 | 46.0 | 133400 | nan | 1.0 | 0.9981 |
best_model/added_tokens.json ADDED
@@ -0,0 +1,4 @@
 
 
 
 
 
1
+ {
2
+ "</s>": 21,
3
+ "<s>": 20
4
+ }
best_model/config.json ADDED
@@ -0,0 +1,110 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_name_or_path": "facebook/wav2vec2-base-960h",
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
+ "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": "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_dropout": 0.0,
52
+ "feat_extract_norm": "group",
53
+ "feat_proj_dropout": 0.1,
54
+ "feat_quantizer_dropout": 0.0,
55
+ "final_dropout": 0.1,
56
+ "gradient_checkpointing": false,
57
+ "hidden_act": "gelu",
58
+ "hidden_dropout": 0.1,
59
+ "hidden_dropout_prob": 0.1,
60
+ "hidden_size": 768,
61
+ "initializer_range": 0.02,
62
+ "intermediate_size": 3072,
63
+ "layer_norm_eps": 1e-05,
64
+ "layerdrop": 0.1,
65
+ "mask_feature_length": 10,
66
+ "mask_feature_min_masks": 0,
67
+ "mask_feature_prob": 0.0,
68
+ "mask_time_length": 10,
69
+ "mask_time_min_masks": 2,
70
+ "mask_time_prob": 0.05,
71
+ "model_type": "wav2vec2",
72
+ "num_adapter_layers": 3,
73
+ "num_attention_heads": 12,
74
+ "num_codevector_groups": 2,
75
+ "num_codevectors_per_group": 320,
76
+ "num_conv_pos_embedding_groups": 16,
77
+ "num_conv_pos_embeddings": 128,
78
+ "num_feat_extract_layers": 7,
79
+ "num_hidden_layers": 12,
80
+ "num_negatives": 100,
81
+ "output_hidden_size": 768,
82
+ "pad_token_id": 0,
83
+ "proj_codevector_dim": 256,
84
+ "tdnn_dilation": [
85
+ 1,
86
+ 2,
87
+ 3,
88
+ 1,
89
+ 1
90
+ ],
91
+ "tdnn_dim": [
92
+ 512,
93
+ 512,
94
+ 512,
95
+ 512,
96
+ 1500
97
+ ],
98
+ "tdnn_kernel": [
99
+ 5,
100
+ 3,
101
+ 3,
102
+ 1,
103
+ 1
104
+ ],
105
+ "torch_dtype": "float32",
106
+ "transformers_version": "4.43.3",
107
+ "use_weighted_layer_sum": false,
108
+ "vocab_size": 32,
109
+ "xvector_output_dim": 512
110
+ }
best_model/model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:312bea3b9f1558787cce01d114cc7a625236be0cb38e65ca8eb1024af2255432
3
+ size 377611120
best_model/preprocessor_config.json ADDED
@@ -0,0 +1,10 @@
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "do_normalize": true,
3
+ "feature_extractor_type": "Wav2Vec2FeatureExtractor",
4
+ "feature_size": 1,
5
+ "padding_side": "right",
6
+ "padding_value": 0.0,
7
+ "processor_class": "Wav2Vec2Processor",
8
+ "return_attention_mask": false,
9
+ "sampling_rate": 16000
10
+ }
best_model/special_tokens_map.json ADDED
@@ -0,0 +1,30 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "bos_token": {
3
+ "content": "<s>",
4
+ "lstrip": false,
5
+ "normalized": false,
6
+ "rstrip": false,
7
+ "single_word": false
8
+ },
9
+ "eos_token": {
10
+ "content": "</s>",
11
+ "lstrip": false,
12
+ "normalized": false,
13
+ "rstrip": false,
14
+ "single_word": false
15
+ },
16
+ "pad_token": {
17
+ "content": "[PAD]",
18
+ "lstrip": true,
19
+ "normalized": false,
20
+ "rstrip": true,
21
+ "single_word": false
22
+ },
23
+ "unk_token": {
24
+ "content": "[UNK]",
25
+ "lstrip": true,
26
+ "normalized": false,
27
+ "rstrip": true,
28
+ "single_word": false
29
+ }
30
+ }
best_model/tokenizer_config.json ADDED
@@ -0,0 +1,48 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "added_tokens_decoder": {
3
+ "18": {
4
+ "content": "[UNK]",
5
+ "lstrip": true,
6
+ "normalized": false,
7
+ "rstrip": true,
8
+ "single_word": false,
9
+ "special": false
10
+ },
11
+ "19": {
12
+ "content": "[PAD]",
13
+ "lstrip": true,
14
+ "normalized": false,
15
+ "rstrip": true,
16
+ "single_word": false,
17
+ "special": false
18
+ },
19
+ "20": {
20
+ "content": "<s>",
21
+ "lstrip": false,
22
+ "normalized": false,
23
+ "rstrip": false,
24
+ "single_word": false,
25
+ "special": true
26
+ },
27
+ "21": {
28
+ "content": "</s>",
29
+ "lstrip": false,
30
+ "normalized": false,
31
+ "rstrip": false,
32
+ "single_word": false,
33
+ "special": true
34
+ }
35
+ },
36
+ "bos_token": "<s>",
37
+ "clean_up_tokenization_spaces": true,
38
+ "do_lower_case": false,
39
+ "eos_token": "</s>",
40
+ "model_max_length": 1000000000000000019884624838656,
41
+ "pad_token": "[PAD]",
42
+ "processor_class": "Wav2Vec2Processor",
43
+ "replace_word_delimiter_char": " ",
44
+ "target_lang": null,
45
+ "tokenizer_class": "Wav2Vec2CTCTokenizer",
46
+ "unk_token": "[UNK]",
47
+ "word_delimiter_token": "|"
48
+ }
best_model/vocab.json ADDED
@@ -0,0 +1,22 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "[PAD]": 19,
3
+ "[UNK]": 18,
4
+ "a": 2,
5
+ "c": 17,
6
+ "d": 11,
7
+ "e": 4,
8
+ "f": 10,
9
+ "h": 7,
10
+ "i": 9,
11
+ "l": 8,
12
+ "m": 14,
13
+ "n": 13,
14
+ "o": 16,
15
+ "p": 3,
16
+ "r": 12,
17
+ "s": 1,
18
+ "t": 15,
19
+ "u": 6,
20
+ "x": 0,
21
+ "|": 5
22
+ }