hts98 commited on
Commit
278a5b5
1 Parent(s): 90ab242

Training in progress, epoch 1

Browse files
added_tokens.json ADDED
@@ -0,0 +1,4 @@
 
 
 
 
 
1
+ {
2
+ "</s>": 109,
3
+ "<s>": 108
4
+ }
config.json CHANGED
@@ -9,7 +9,7 @@
9
  "architectures": [
10
  "Wav2Vec2ForCTC"
11
  ],
12
- "attention_dropout": 0.1,
13
  "bos_token_id": 1,
14
  "classifier_proj_size": 256,
15
  "codevector_dim": 768,
@@ -53,14 +53,13 @@
53
  "feat_proj_dropout": 0.0,
54
  "feat_quantizer_dropout": 0.0,
55
  "final_dropout": 0.0,
56
- "gradient_checkpointing": false,
57
  "hidden_act": "gelu",
58
- "hidden_dropout": 0.1,
59
  "hidden_size": 1024,
60
  "initializer_range": 0.02,
61
  "intermediate_size": 4096,
62
  "layer_norm_eps": 1e-05,
63
- "layerdrop": 0.1,
64
  "mask_feature_length": 10,
65
  "mask_feature_min_masks": 0,
66
  "mask_feature_prob": 0.0,
@@ -78,7 +77,7 @@
78
  "num_hidden_layers": 24,
79
  "num_negatives": 100,
80
  "output_hidden_size": 1024,
81
- "pad_token_id": 105,
82
  "proj_codevector_dim": 768,
83
  "tdnn_dilation": [
84
  1,
@@ -102,8 +101,8 @@
102
  1
103
  ],
104
  "torch_dtype": "float32",
105
- "transformers_version": "4.30.2",
106
  "use_weighted_layer_sum": false,
107
- "vocab_size": 106,
108
  "xvector_output_dim": 512
109
  }
 
9
  "architectures": [
10
  "Wav2Vec2ForCTC"
11
  ],
12
+ "attention_dropout": 0.0,
13
  "bos_token_id": 1,
14
  "classifier_proj_size": 256,
15
  "codevector_dim": 768,
 
53
  "feat_proj_dropout": 0.0,
54
  "feat_quantizer_dropout": 0.0,
55
  "final_dropout": 0.0,
 
56
  "hidden_act": "gelu",
57
+ "hidden_dropout": 0.0,
58
  "hidden_size": 1024,
59
  "initializer_range": 0.02,
60
  "intermediate_size": 4096,
61
  "layer_norm_eps": 1e-05,
62
+ "layerdrop": 0.0,
63
  "mask_feature_length": 10,
64
  "mask_feature_min_masks": 0,
65
  "mask_feature_prob": 0.0,
 
77
  "num_hidden_layers": 24,
78
  "num_negatives": 100,
79
  "output_hidden_size": 1024,
80
+ "pad_token_id": 107,
81
  "proj_codevector_dim": 768,
82
  "tdnn_dilation": [
83
  1,
 
101
  1
102
  ],
103
  "torch_dtype": "float32",
104
+ "transformers_version": "4.31.0.dev0",
105
  "use_weighted_layer_sum": false,
106
+ "vocab_size": 110,
107
  "xvector_output_dim": 512
108
  }
preprocessor_config.json CHANGED
@@ -3,7 +3,8 @@
3
  "feature_extractor_type": "Wav2Vec2FeatureExtractor",
4
  "feature_size": 1,
5
  "padding_side": "right",
6
- "padding_value": 0.0,
 
7
  "return_attention_mask": true,
8
  "sampling_rate": 16000
9
  }
 
3
  "feature_extractor_type": "Wav2Vec2FeatureExtractor",
4
  "feature_size": 1,
5
  "padding_side": "right",
6
+ "padding_value": 0,
7
+ "processor_class": "Wav2Vec2Processor",
8
  "return_attention_mask": true,
9
  "sampling_rate": 16000
10
  }
pytorch_model.bin CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:ca5184a28a77745e775de7517698661308ce5090d57c8ab62a923b790ece8ed1
3
- size 1262336429
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:6f659c9a20b5d330d5b73a45c71b0e1becf93e1349578ac3a304019af8afbcfb
3
+ size 1262352813
runs/Jun30_10-04-05_64df48378bb2/events.out.tfevents.1688120414.64df48378bb2.1419.0 ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:018882fbff9f0b0de67d1635e3c4e7cce04fd3814968bdcdb421927f854516ef
3
+ size 5845
special_tokens_map.json CHANGED
@@ -1,4 +1,34 @@
1
  {
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
2
  "bos_token": "<s>",
3
  "eos_token": "</s>",
4
  "pad_token": "[PAD]",
 
1
  {
2
+ "additional_special_tokens": [
3
+ {
4
+ "content": "<s>",
5
+ "lstrip": false,
6
+ "normalized": true,
7
+ "rstrip": false,
8
+ "single_word": false
9
+ },
10
+ {
11
+ "content": "</s>",
12
+ "lstrip": false,
13
+ "normalized": true,
14
+ "rstrip": false,
15
+ "single_word": false
16
+ },
17
+ {
18
+ "content": "<s>",
19
+ "lstrip": false,
20
+ "normalized": true,
21
+ "rstrip": false,
22
+ "single_word": false
23
+ },
24
+ {
25
+ "content": "</s>",
26
+ "lstrip": false,
27
+ "normalized": true,
28
+ "rstrip": false,
29
+ "single_word": false
30
+ }
31
+ ],
32
  "bos_token": "<s>",
33
  "eos_token": "</s>",
34
  "pad_token": "[PAD]",
tokenizer_config.json CHANGED
@@ -5,6 +5,7 @@
5
  "eos_token": "</s>",
6
  "model_max_length": 1000000000000000019884624838656,
7
  "pad_token": "[PAD]",
 
8
  "replace_word_delimiter_char": " ",
9
  "target_lang": null,
10
  "tokenizer_class": "Wav2Vec2CTCTokenizer",
 
5
  "eos_token": "</s>",
6
  "model_max_length": 1000000000000000019884624838656,
7
  "pad_token": "[PAD]",
8
+ "processor_class": "Wav2Vec2Processor",
9
  "replace_word_delimiter_char": " ",
10
  "target_lang": null,
11
  "tokenizer_class": "Wav2Vec2CTCTokenizer",
training_args.bin CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:4f63e8248c90f82c8a0a9ea73e839432cfc0c7956368674a06d74ae0ea6ea517
3
- size 3963
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:645a14c785ae828345ed6de887448e019cdae478f8e4270b5aa6a6943e08281f
3
+ size 4027
vocab.json CHANGED
@@ -1,108 +1,110 @@
1
  {
2
- "0": 6,
3
- "1": 82,
4
- "2": 46,
5
- "3": 88,
6
- "4": 5,
7
- "5": 25,
8
- "6": 44,
9
- "7": 18,
10
- "8": 58,
11
- "9": 40,
12
- "[PAD]": 105,
13
- "[UNK]": 104,
14
- "a": 54,
15
- "b": 67,
16
- "c": 70,
17
- "d": 17,
18
- "e": 57,
19
- "f": 102,
20
- "g": 73,
21
- "h": 21,
22
- "i": 11,
23
- "j": 36,
24
- "k": 50,
25
- "l": 22,
26
- "m": 12,
27
- "n": 81,
28
- "o": 10,
29
- "p": 90,
30
- "q": 35,
31
- "r": 3,
32
- "s": 32,
33
- "t": 29,
34
- "u": 76,
35
- "v": 43,
36
- "w": 99,
37
- "x": 16,
38
- "y": 64,
39
- "z": 93,
40
- "|": 84,
41
- "à": 78,
42
- "á": 49,
43
- "â": 38,
44
- "ã": 69,
45
- "è": 86,
46
- "é": 28,
47
- "ê": 94,
48
- "ì": 59,
 
 
49
  "í": 47,
50
- "ò": 103,
51
- "ó": 100,
52
- "ô": 8,
53
- "õ": 62,
54
- "ù": 74,
55
- "ú": 52,
56
- "ý": 65,
57
- "ă": 60,
58
- "đ": 75,
59
- "ĩ": 91,
60
- "ũ": 37,
61
- "ơ": 97,
62
- "ư": 71,
63
- "ạ": 0,
64
- "ả": 56,
65
- "ấ": 39,
66
- "ầ": 98,
67
- "ẩ": 42,
68
- "ẫ": 53,
69
- "ậ": 95,
70
- "ắ": 4,
71
- "ằ": 1,
72
- "ẳ": 31,
73
- "ẵ": 63,
74
- "ặ": 15,
75
- "ẹ": 72,
76
- "ẻ": 80,
77
- "ẽ": 33,
78
- "ế": 27,
79
- "ề": 48,
80
- "ể": 68,
81
- "ễ": 51,
82
- "ệ": 96,
83
- "ỉ": 14,
84
- "ị": 2,
85
- "ọ": 23,
86
- "ỏ": 83,
87
- "ố": 66,
88
- "ồ": 34,
89
- "ổ": 77,
90
- "ỗ": 41,
91
- "ộ": 20,
92
- "ớ": 55,
93
- "ờ": 26,
94
- "ở": 85,
95
- "ỡ": 30,
96
- "ợ": 87,
97
- "ụ": 79,
98
- "ủ": 45,
99
- "ứ": 19,
100
- "ừ": 24,
101
- "ử": 61,
102
- "ữ": 92,
103
- "ự": 7,
104
- "ỳ": 89,
105
- "ỷ": 101,
106
- "ỹ": 13,
107
- "₫": 9
108
  }
 
1
  {
2
+ ",": 1,
3
+ ".": 2,
4
+ "0": 3,
5
+ "1": 4,
6
+ "2": 5,
7
+ "3": 6,
8
+ "4": 7,
9
+ "5": 8,
10
+ "6": 9,
11
+ "7": 10,
12
+ "8": 11,
13
+ "9": 12,
14
+ "[PAD]": 107,
15
+ "[UNK]": 106,
16
+ "a": 13,
17
+ "b": 14,
18
+ "c": 15,
19
+ "d": 16,
20
+ "e": 17,
21
+ "f": 18,
22
+ "g": 19,
23
+ "h": 20,
24
+ "i": 21,
25
+ "j": 22,
26
+ "k": 23,
27
+ "l": 24,
28
+ "m": 25,
29
+ "n": 26,
30
+ "o": 27,
31
+ "p": 28,
32
+ "q": 29,
33
+ "r": 30,
34
+ "s": 31,
35
+ "t": 32,
36
+ "u": 33,
37
+ "v": 34,
38
+ "w": 35,
39
+ "x": 36,
40
+ "y": 37,
41
+ "z": 38,
42
+ "|": 0,
43
+ "à": 39,
44
+ "á": 40,
45
+ "â": 41,
46
+ "ã": 42,
47
+ "è": 43,
48
+ "é": 44,
49
+ "ê": 45,
50
+ "ì": 46,
51
  "í": 47,
52
+ "ò": 48,
53
+ "ó": 49,
54
+ "ô": 50,
55
+ "õ": 51,
56
+ "ù": 52,
57
+ "ú": 53,
58
+ "ý": 54,
59
+ "ă": 55,
60
+ "đ": 56,
61
+ "ĩ": 57,
62
+ "ũ": 58,
63
+ "ơ": 59,
64
+ "ư": 60,
65
+ "ạ": 61,
66
+ "ả": 62,
67
+ "ấ": 63,
68
+ "ầ": 64,
69
+ "ẩ": 65,
70
+ "ẫ": 66,
71
+ "ậ": 67,
72
+ "ắ": 68,
73
+ "ằ": 69,
74
+ "ẳ": 70,
75
+ "ẵ": 71,
76
+ "ặ": 72,
77
+ "ẹ": 73,
78
+ "ẻ": 74,
79
+ "ẽ": 75,
80
+ "ế": 76,
81
+ "ề": 77,
82
+ "ể": 78,
83
+ "ễ": 79,
84
+ "ệ": 80,
85
+ "ỉ": 81,
86
+ "ị": 82,
87
+ "ọ": 83,
88
+ "ỏ": 84,
89
+ "ố": 85,
90
+ "ồ": 86,
91
+ "ổ": 87,
92
+ "ỗ": 88,
93
+ "ộ": 89,
94
+ "ớ": 90,
95
+ "ờ": 91,
96
+ "ở": 92,
97
+ "ỡ": 93,
98
+ "ợ": 94,
99
+ "ụ": 95,
100
+ "ủ": 96,
101
+ "ứ": 97,
102
+ "ừ": 98,
103
+ "ử": 99,
104
+ "ữ": 100,
105
+ "ự": 101,
106
+ "ỳ": 102,
107
+ "ỷ": 103,
108
+ "ỹ": 104,
109
+ "₫": 105
110
  }