Xenova HF staff commited on
Commit
c4eb0c7
1 Parent(s): 677567b

Upload folder using huggingface_hub

Browse files
config.json CHANGED
@@ -50,6 +50,7 @@
50
  "max_length": 448,
51
  "max_source_positions": 1500,
52
  "max_target_positions": 448,
 
53
  "model_type": "whisper",
54
  "num_hidden_layers": 4,
55
  "num_mel_bins": 80,
@@ -145,7 +146,7 @@
145
  50361,
146
  50362
147
  ],
148
- "transformers_version": "4.30.2",
149
  "use_cache": true,
150
  "use_weighted_layer_sum": false,
151
  "vocab_size": 51865
 
50
  "max_length": 448,
51
  "max_source_positions": 1500,
52
  "max_target_positions": 448,
53
+ "median_filter_width": 7,
54
  "model_type": "whisper",
55
  "num_hidden_layers": 4,
56
  "num_mel_bins": 80,
 
146
  50361,
147
  50362
148
  ],
149
+ "transformers_version": "4.33.0.dev0",
150
  "use_cache": true,
151
  "use_weighted_layer_sum": false,
152
  "vocab_size": 51865
generation_config.json CHANGED
@@ -1,5 +1,30 @@
1
  {
2
- "alignment_heads": [[2, 2], [3, 0], [3, 2], [3, 3], [3, 4], [3, 5]],
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
3
  "begin_suppress_tokens": [
4
  220,
5
  50257
@@ -218,5 +243,5 @@
218
  "transcribe": 50359,
219
  "translate": 50358
220
  },
221
- "transformers_version": "4.30.2"
222
  }
 
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
+ ],
28
  "begin_suppress_tokens": [
29
  220,
30
  50257
 
243
  "transcribe": 50359,
244
  "translate": 50358
245
  },
246
+ "transformers_version": "4.33.0.dev0"
247
  }
onnx/decoder_model.onnx CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:28fc8dfa729d0fba4dfbd09e2c9f95a70eaa1a657f15a9fb0e6d00d1228ca51f
3
- size 198049530
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:d7668e6e5d90c757759d2a29c918f5f24f6eef073fc5a17c5b86c1a68a17555f
3
+ size 118406656
onnx/decoder_model_merged.onnx CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:3a3f5dfc9cf89e4910a1be6574ed519181eefc743fd125ddcd4a97632fd44e94
3
- size 198197535
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:b81258b0e8acdbedce92eb901d18f1c189b2d6f3d8668261d65117d423f9a5b9
3
+ size 118579599
onnx/decoder_model_merged_quantized.onnx CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:3630e9e28169f8664cdaed6d2b0f00abae079bae7e4336c249308022635721b4
3
- size 50584327
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:6c0c125986b007d2e3734bec84c18bda0152071b90b87fadac6d7764499927a0
3
+ size 30727765
onnx/decoder_model_quantized.onnx CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:6ae32f7a6c43d40740199b20c1720539f3d6dcd73d1bc0ff0c9d22f4386bdc37
3
- size 50348480
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:d1ac41063dd0a01d278dca49214ff8ce1b930cc9e34e2881c56a79a428650534
3
+ size 30460101
onnx/decoder_with_past_model.onnx CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:3df7e2682d1b77dc3706a9c444335ccf4e441fab2e64562d2e4934f70ecfaba2
3
- size 193295315
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:25efcc2a9319b897bc0c3b6b2cd1f8ab8a25ebff6da4be828e4766f4a597d5f6
3
+ size 113654551
onnx/decoder_with_past_model_quantized.onnx CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:ac92773054aca908fca90d7e84a8c97a995526e9fe6292cf00c6e77ff7d223dd
3
- size 49114568
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:26d8d7fa7829644b22762c965351893c3fb992501ed1776d54f126530eafabf8
3
+ size 29229119
onnx/encoder_model.onnx CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:4164766243b81a4bba7fc51bb0c37539d15b11e2d39820d15f65bdea39942feb
3
- size 32900723
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:39e81b6c86a5b2b4beda1bb3145486a769d594801f780a66cad1ae72c7ad2c5e
3
+ size 32909539
onnx/encoder_model_quantized.onnx CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:ca9d7bb2836193704b7e2435e3bbadbed985ac3a79ab7406b244b8865ab1a5c0
3
- size 10113248
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:fd9d995b9dcb0520f0dbf6cf68651af639fc385f594d9d876e69ca2802dc438e
3
+ size 10124910
quantize_config.json ADDED
@@ -0,0 +1,115 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "per_channel": false,
3
+ "reduce_range": false,
4
+ "per_model_config": {
5
+ "decoder_model": {
6
+ "op_types": [
7
+ "Cast",
8
+ "ConstantOfShape",
9
+ "Squeeze",
10
+ "Add",
11
+ "Sub",
12
+ "Equal",
13
+ "Softmax",
14
+ "Where",
15
+ "Less",
16
+ "Concat",
17
+ "Gather",
18
+ "Pow",
19
+ "MatMul",
20
+ "Div",
21
+ "Unsqueeze",
22
+ "Shape",
23
+ "ReduceMean",
24
+ "Transpose",
25
+ "Expand",
26
+ "Constant",
27
+ "Reshape",
28
+ "Sqrt",
29
+ "Slice",
30
+ "Erf",
31
+ "Mul",
32
+ "Range"
33
+ ],
34
+ "weight_type": "QInt8"
35
+ },
36
+ "decoder_model_merged": {
37
+ "op_types": [
38
+ "Cast",
39
+ "ConstantOfShape",
40
+ "Squeeze",
41
+ "Add",
42
+ "Sub",
43
+ "Equal",
44
+ "Softmax",
45
+ "Where",
46
+ "Less",
47
+ "Concat",
48
+ "Gather",
49
+ "Pow",
50
+ "MatMul",
51
+ "If",
52
+ "Div",
53
+ "Unsqueeze",
54
+ "Shape",
55
+ "ReduceMean",
56
+ "Transpose",
57
+ "Expand",
58
+ "Constant",
59
+ "Reshape",
60
+ "Sqrt",
61
+ "Slice",
62
+ "Erf",
63
+ "Mul",
64
+ "Range"
65
+ ],
66
+ "weight_type": "QInt8"
67
+ },
68
+ "decoder_with_past_model": {
69
+ "op_types": [
70
+ "Sub",
71
+ "Div",
72
+ "Unsqueeze",
73
+ "Gather",
74
+ "ReduceMean",
75
+ "Pow",
76
+ "Transpose",
77
+ "Softmax",
78
+ "Constant",
79
+ "Reshape",
80
+ "Sqrt",
81
+ "Slice",
82
+ "Erf",
83
+ "MatMul",
84
+ "Concat",
85
+ "Mul",
86
+ "Add",
87
+ "Shape"
88
+ ],
89
+ "weight_type": "QInt8"
90
+ },
91
+ "encoder_model": {
92
+ "op_types": [
93
+ "Div",
94
+ "Sub",
95
+ "Unsqueeze",
96
+ "ReduceMean",
97
+ "Pow",
98
+ "Transpose",
99
+ "Gather",
100
+ "Softmax",
101
+ "Conv",
102
+ "Constant",
103
+ "Reshape",
104
+ "Sqrt",
105
+ "Erf",
106
+ "MatMul",
107
+ "Concat",
108
+ "Mul",
109
+ "Add",
110
+ "Shape"
111
+ ],
112
+ "weight_type": "QUInt8"
113
+ }
114
+ }
115
+ }