Xenova HF staff commited on
Commit
d2f737c
1 Parent(s): 93be538

Upload folder using huggingface_hub

Browse files
config.json CHANGED
@@ -57,7 +57,7 @@
57
  "scale_embedding": true,
58
  "share_encoder_decoder_embeddings": true,
59
  "static_position_embeddings": true,
60
- "transformers_version": "4.32.0.dev0",
61
  "use_cache": true,
62
  "vocab_size": 65001
63
  }
 
57
  "scale_embedding": true,
58
  "share_encoder_decoder_embeddings": true,
59
  "static_position_embeddings": true,
60
+ "transformers_version": "4.34.0.dev0",
61
  "use_cache": true,
62
  "vocab_size": 65001
63
  }
generation_config.json CHANGED
@@ -1,5 +1,4 @@
1
  {
2
- "_from_model_config": true,
3
  "bad_words_ids": [
4
  [
5
  65000
@@ -12,5 +11,6 @@
12
  "max_length": 512,
13
  "num_beams": 4,
14
  "pad_token_id": 65000,
15
- "transformers_version": "4.32.0.dev0"
 
16
  }
 
1
  {
 
2
  "bad_words_ids": [
3
  [
4
  65000
 
11
  "max_length": 512,
12
  "num_beams": 4,
13
  "pad_token_id": 65000,
14
+ "renormalize_logits": true,
15
+ "transformers_version": "4.34.0.dev0"
16
  }
onnx/decoder_model.onnx CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:dd2c57556209a030144100d8763769bad4248e5091d016ad76749e50e2d919e2
3
- size 368725660
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:7d4c10b0a9b5e0e253cd717b1de70ac61d79624f9cd9d35622fd36c9dcbf47c1
3
+ size 235603669
onnx/decoder_model_merged.onnx CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:19568739bc7ccb217e6a517073389a545e9b9b9f812c9a2c74aa31143c80309f
3
- size 368961142
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:2d3fbdb1de708ceb3531b47dcd460e73f0861a7e1e3bb60faff8ca1a0e4f8cf7
3
+ size 235839236
onnx/decoder_model_merged_quantized.onnx CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:83c7044c96aa2113be3278220287c7800afa890c33e0fdd15ba24e938ddad4a8
3
- size 93819733
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:b5f6d442da7fa69f9d500df90373d2fdca4f799c43696b41888353b5c3c8eb13
3
+ size 60212804
onnx/decoder_model_quantized.onnx CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:a408152e2544a17d23acf68782f69b12f34eab83a9f5df38169c4d48b79a5142
3
- size 93448377
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:ec98bfea479f392987bc0eb5696dbf9b042556144c9da249688b54b462a068bc
3
+ size 59842102
onnx/decoder_with_past_model.onnx CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:1c17794858cb965bb28866230d3bdce2c4d5eee08d00fb7e897f6f6a09111134
3
- size 356075151
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:d215cb509ceb083e9402e3bf7bf0f4a246f4232c43606923c4551b584b1d2da3
3
+ size 222953160
onnx/decoder_with_past_model_quantized.onnx CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:ef1594ef4271af8fe9d1fad4a4b1edae5f36304326023a0b6713c09780bad486
3
- size 90176913
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:28b8a5b8d684ae9e1a0b27ddcc1e08bd9798ea8c97a01f7dc366d217ba5b8942
3
+ size 56570638
quantize_config.json CHANGED
@@ -4,120 +4,120 @@
4
  "per_model_config": {
5
  "encoder_model": {
6
  "op_types": [
7
- "Softmax",
8
- "MatMul",
9
  "Gather",
10
- "Where",
11
  "Pow",
12
  "Unsqueeze",
13
- "Div",
14
- "ConstantOfShape",
15
- "Concat",
16
- "Cast",
17
- "Sub",
18
- "Range",
19
- "ReduceMean",
20
  "Transpose",
21
- "Equal",
22
- "Add",
 
 
23
  "Expand",
24
- "Sigmoid",
25
- "Mul",
26
- "Constant",
27
  "Sqrt",
 
 
 
28
  "Reshape",
29
- "Shape"
 
 
 
 
30
  ],
31
  "weight_type": "QInt8"
32
  },
33
  "decoder_with_past_model": {
34
  "op_types": [
35
- "Softmax",
36
- "MatMul",
37
  "Gather",
38
- "Where",
39
  "Pow",
40
  "Unsqueeze",
41
- "Div",
42
- "ConstantOfShape",
43
- "Concat",
44
- "Cast",
45
- "Sub",
46
- "Range",
47
- "ReduceMean",
48
  "Transpose",
49
- "Equal",
50
- "Add",
 
 
51
  "Expand",
52
- "Sigmoid",
53
- "Mul",
54
- "Constant",
55
  "Sqrt",
 
 
 
56
  "Reshape",
57
- "Shape"
 
 
 
 
58
  ],
59
  "weight_type": "QInt8"
60
  },
61
- "decoder_model": {
62
  "op_types": [
63
- "Softmax",
64
- "MatMul",
65
  "Gather",
66
- "Where",
67
  "Pow",
68
  "Unsqueeze",
 
 
 
69
  "Div",
70
- "ConstantOfShape",
71
- "Concat",
72
- "Slice",
73
- "Cast",
74
- "Squeeze",
75
- "Sub",
76
- "Range",
77
  "ReduceMean",
78
- "Transpose",
 
79
  "Equal",
80
  "Less",
81
- "Add",
82
- "Expand",
83
- "Sigmoid",
84
- "Mul",
85
- "Constant",
86
- "Sqrt",
87
  "Reshape",
88
- "Shape"
 
 
 
 
 
89
  ],
90
  "weight_type": "QInt8"
91
  },
92
- "decoder_model_merged": {
93
  "op_types": [
94
- "Softmax",
95
- "MatMul",
96
  "Gather",
97
- "Where",
98
  "Pow",
99
  "Unsqueeze",
 
 
 
100
  "Div",
101
- "ConstantOfShape",
102
- "Concat",
103
- "Slice",
104
- "Cast",
105
- "Squeeze",
106
- "If",
107
- "Sub",
108
- "Range",
109
  "ReduceMean",
110
- "Transpose",
 
111
  "Equal",
112
  "Less",
113
- "Add",
114
- "Expand",
115
- "Sigmoid",
116
- "Mul",
117
- "Constant",
118
- "Sqrt",
119
  "Reshape",
120
- "Shape"
 
 
 
 
 
121
  ],
122
  "weight_type": "QInt8"
123
  }
 
4
  "per_model_config": {
5
  "encoder_model": {
6
  "op_types": [
7
+ "Add",
8
+ "Constant",
9
  "Gather",
10
+ "Cast",
11
  "Pow",
12
  "Unsqueeze",
 
 
 
 
 
 
 
13
  "Transpose",
14
+ "MatMul",
15
+ "Where",
16
+ "Div",
17
+ "Softmax",
18
  "Expand",
19
+ "ReduceMean",
 
 
20
  "Sqrt",
21
+ "Equal",
22
+ "Shape",
23
+ "Range",
24
  "Reshape",
25
+ "Mul",
26
+ "Sub",
27
+ "ConstantOfShape",
28
+ "Sigmoid",
29
+ "Concat"
30
  ],
31
  "weight_type": "QInt8"
32
  },
33
  "decoder_with_past_model": {
34
  "op_types": [
35
+ "Add",
36
+ "Constant",
37
  "Gather",
38
+ "Cast",
39
  "Pow",
40
  "Unsqueeze",
 
 
 
 
 
 
 
41
  "Transpose",
42
+ "MatMul",
43
+ "Where",
44
+ "Div",
45
+ "Softmax",
46
  "Expand",
47
+ "ReduceMean",
 
 
48
  "Sqrt",
49
+ "Equal",
50
+ "Shape",
51
+ "Range",
52
  "Reshape",
53
+ "Mul",
54
+ "Sub",
55
+ "ConstantOfShape",
56
+ "Sigmoid",
57
+ "Concat"
58
  ],
59
  "weight_type": "QInt8"
60
  },
61
+ "decoder_model_merged": {
62
  "op_types": [
63
+ "Add",
64
+ "Constant",
65
  "Gather",
66
+ "Cast",
67
  "Pow",
68
  "Unsqueeze",
69
+ "Transpose",
70
+ "MatMul",
71
+ "Where",
72
  "Div",
73
+ "Softmax",
74
+ "Expand",
75
+ "If",
 
 
 
 
76
  "ReduceMean",
77
+ "Slice",
78
+ "Sqrt",
79
  "Equal",
80
  "Less",
81
+ "Shape",
82
+ "Range",
 
 
 
 
83
  "Reshape",
84
+ "Mul",
85
+ "Sub",
86
+ "ConstantOfShape",
87
+ "Sigmoid",
88
+ "Squeeze",
89
+ "Concat"
90
  ],
91
  "weight_type": "QInt8"
92
  },
93
+ "decoder_model": {
94
  "op_types": [
95
+ "Add",
96
+ "Constant",
97
  "Gather",
98
+ "Cast",
99
  "Pow",
100
  "Unsqueeze",
101
+ "Transpose",
102
+ "MatMul",
103
+ "Where",
104
  "Div",
105
+ "Softmax",
106
+ "Expand",
 
 
 
 
 
 
107
  "ReduceMean",
108
+ "Slice",
109
+ "Sqrt",
110
  "Equal",
111
  "Less",
112
+ "Shape",
113
+ "Range",
 
 
 
 
114
  "Reshape",
115
+ "Mul",
116
+ "Sub",
117
+ "ConstantOfShape",
118
+ "Sigmoid",
119
+ "Squeeze",
120
+ "Concat"
121
  ],
122
  "weight_type": "QInt8"
123
  }
tokenizer.json CHANGED
The diff for this file is too large to render. See raw diff