Upload WhisperForConditionalGeneration

#5
by Dmitriy - opened
Files changed (3) hide show
  1. config.json +113 -11
  2. generation_config.json +10 -26
  3. pytorch_model.bin +2 -2
config.json CHANGED
@@ -1,5 +1,5 @@
1
  {
2
- "_name_or_path": "openai/whisper-small",
3
  "activation_dropout": 0.0,
4
  "activation_function": "gelu",
5
  "apply_spec_augment": false,
@@ -13,19 +13,32 @@
13
  ],
14
  "bos_token_id": 50257,
15
  "classifier_proj_size": 256,
16
- "d_model": 768,
17
- "decoder_attention_heads": 12,
18
- "decoder_ffn_dim": 3072,
19
  "decoder_layerdrop": 0.0,
20
- "decoder_layers": 12,
21
  "decoder_start_token_id": 50258,
22
  "dropout": 0.0,
23
- "encoder_attention_heads": 12,
24
- "encoder_ffn_dim": 3072,
25
  "encoder_layerdrop": 0.0,
26
- "encoder_layers": 12,
27
  "eos_token_id": 50257,
28
- "forced_decoder_ids": null,
 
 
 
 
 
 
 
 
 
 
 
 
 
29
  "init_std": 0.02,
30
  "is_encoder_decoder": true,
31
  "mask_feature_length": 10,
@@ -38,11 +51,100 @@
38
  "max_source_positions": 1500,
39
  "max_target_positions": 448,
40
  "model_type": "whisper",
41
- "num_hidden_layers": 12,
42
  "num_mel_bins": 80,
43
  "pad_token_id": 50257,
44
  "scale_embedding": false,
45
- "suppress_tokens": [],
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
46
  "torch_dtype": "float32",
47
  "transformers_version": "4.30.2",
48
  "use_cache": true,
 
1
  {
2
+ "_name_or_path": "openai/whisper-tiny",
3
  "activation_dropout": 0.0,
4
  "activation_function": "gelu",
5
  "apply_spec_augment": false,
 
13
  ],
14
  "bos_token_id": 50257,
15
  "classifier_proj_size": 256,
16
+ "d_model": 384,
17
+ "decoder_attention_heads": 6,
18
+ "decoder_ffn_dim": 1536,
19
  "decoder_layerdrop": 0.0,
20
+ "decoder_layers": 4,
21
  "decoder_start_token_id": 50258,
22
  "dropout": 0.0,
23
+ "encoder_attention_heads": 6,
24
+ "encoder_ffn_dim": 1536,
25
  "encoder_layerdrop": 0.0,
26
+ "encoder_layers": 4,
27
  "eos_token_id": 50257,
28
+ "forced_decoder_ids": [
29
+ [
30
+ 1,
31
+ 50259
32
+ ],
33
+ [
34
+ 2,
35
+ 50359
36
+ ],
37
+ [
38
+ 3,
39
+ 50363
40
+ ]
41
+ ],
42
  "init_std": 0.02,
43
  "is_encoder_decoder": true,
44
  "mask_feature_length": 10,
 
51
  "max_source_positions": 1500,
52
  "max_target_positions": 448,
53
  "model_type": "whisper",
54
+ "num_hidden_layers": 4,
55
  "num_mel_bins": 80,
56
  "pad_token_id": 50257,
57
  "scale_embedding": false,
58
+ "suppress_tokens": [
59
+ 1,
60
+ 2,
61
+ 7,
62
+ 8,
63
+ 9,
64
+ 10,
65
+ 14,
66
+ 25,
67
+ 26,
68
+ 27,
69
+ 28,
70
+ 29,
71
+ 31,
72
+ 58,
73
+ 59,
74
+ 60,
75
+ 61,
76
+ 62,
77
+ 63,
78
+ 90,
79
+ 91,
80
+ 92,
81
+ 93,
82
+ 359,
83
+ 503,
84
+ 522,
85
+ 542,
86
+ 873,
87
+ 893,
88
+ 902,
89
+ 918,
90
+ 922,
91
+ 931,
92
+ 1350,
93
+ 1853,
94
+ 1982,
95
+ 2460,
96
+ 2627,
97
+ 3246,
98
+ 3253,
99
+ 3268,
100
+ 3536,
101
+ 3846,
102
+ 3961,
103
+ 4183,
104
+ 4667,
105
+ 6585,
106
+ 6647,
107
+ 7273,
108
+ 9061,
109
+ 9383,
110
+ 10428,
111
+ 10929,
112
+ 11938,
113
+ 12033,
114
+ 12331,
115
+ 12562,
116
+ 13793,
117
+ 14157,
118
+ 14635,
119
+ 15265,
120
+ 15618,
121
+ 16553,
122
+ 16604,
123
+ 18362,
124
+ 18956,
125
+ 20075,
126
+ 21675,
127
+ 22520,
128
+ 26130,
129
+ 26161,
130
+ 26435,
131
+ 28279,
132
+ 29464,
133
+ 31650,
134
+ 32302,
135
+ 32470,
136
+ 36865,
137
+ 42863,
138
+ 47425,
139
+ 49870,
140
+ 50254,
141
+ 50258,
142
+ 50358,
143
+ 50359,
144
+ 50360,
145
+ 50361,
146
+ 50362
147
+ ],
148
  "torch_dtype": "float32",
149
  "transformers_version": "4.30.2",
150
  "use_cache": true,
generation_config.json CHANGED
@@ -1,43 +1,27 @@
1
  {
2
  "alignment_heads": [
3
  [
4
- 5,
5
- 3
6
- ],
7
- [
8
- 5,
9
- 9
10
  ],
11
  [
12
- 8,
13
  0
14
  ],
15
  [
16
- 8,
17
- 4
18
- ],
19
- [
20
- 8,
21
- 7
22
  ],
23
  [
24
- 8,
25
- 8
26
- ],
27
- [
28
- 9,
29
- 0
30
- ],
31
- [
32
- 9,
33
- 7
34
  ],
35
  [
36
- 9,
37
- 9
38
  ],
39
  [
40
- 10,
41
  5
42
  ]
43
  ],
 
1
  {
2
  "alignment_heads": [
3
  [
4
+ 2,
5
+ 2
 
 
 
 
6
  ],
7
  [
8
+ 3,
9
  0
10
  ],
11
  [
12
+ 3,
13
+ 2
 
 
 
 
14
  ],
15
  [
16
+ 3,
17
+ 3
 
 
 
 
 
 
 
 
18
  ],
19
  [
20
+ 3,
21
+ 4
22
  ],
23
  [
24
+ 3,
25
  5
26
  ]
27
  ],
pytorch_model.bin CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:be79b01412dff06f4330b1b75078ec1137fe5b22e26ad0ac0b990c9e66310307
3
- size 967102601
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:f6b57958615c3b737077bc10003bf3d48f5a2c620d7bd4628d1f415e5ed87c21
3
+ size 151096617