Xenova HF staff commited on
Commit
788d976
1 Parent(s): 7e80f6c

Upload folder using huggingface_hub

Browse files
config.json CHANGED
@@ -54,7 +54,7 @@
54
  "scale_embedding": true,
55
  "share_encoder_decoder_embeddings": true,
56
  "static_position_embeddings": true,
57
- "transformers_version": "4.32.0.dev0",
58
  "use_cache": true,
59
  "vocab_size": 7018
60
  }
 
54
  "scale_embedding": true,
55
  "share_encoder_decoder_embeddings": true,
56
  "static_position_embeddings": true,
57
+ "transformers_version": "4.34.0.dev0",
58
  "use_cache": true,
59
  "vocab_size": 7018
60
  }
generation_config.json CHANGED
@@ -1,5 +1,4 @@
1
  {
2
- "_from_model_config": true,
3
  "bad_words_ids": [
4
  [
5
  7017
@@ -12,5 +11,6 @@
12
  "max_length": 512,
13
  "num_beams": 6,
14
  "pad_token_id": 7017,
15
- "transformers_version": "4.32.0.dev0"
 
16
  }
 
1
  {
 
2
  "bad_words_ids": [
3
  [
4
  7017
 
11
  "max_length": 512,
12
  "num_beams": 6,
13
  "pad_token_id": 7017,
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 sha256:628189beacf0105e04154618c4b17d32ff20e78e063aa1538ec75f606a9794ae
3
- size 130995355
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:a037c27fb529b7f16e0447a638d090578697c173df59a89601382ad917d7a0c3
3
+ size 116622550
onnx/decoder_model_merged.onnx CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:0a3522f89ed5ae56e2e2ebdecea5d3d0ebad3b548f76c8d3593fe200f321203d
3
- size 131230835
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:30a89228146d3cf8ec2a223d28c20fba307530f933a8ba11a22b757bbd232ece
3
+ size 116858115
onnx/decoder_model_merged_quantized.onnx CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:748025b07033d85ccdf5811afeef8bbbef2a6a14a289c99c9d1df42681c611d8
3
- size 33923284
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:205ac42a2049111dc2201059a4a810f976e95c4c5582b41aa84e1cfcee287476
3
+ size 30293569
onnx/decoder_model_quantized.onnx CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:40c40cc8b4650cb14164c3486fbbbde5bf6ae670fcb489df0b805566393cc680
3
- size 33551930
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:57332e74b29d75672de88ef08ef7e31c2ff8de9f95542ef321cbb6c86c212a25
3
+ size 29922870
onnx/decoder_with_past_model.onnx CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:20f52f007233530e84ef7f2b1831483a1062dad6742cb0aa39646935d9c4fc96
3
- size 118344846
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:eda9458ac7a08542fa63cf7f3253f54e4adfdffa3fa57ab7aba06e423c182fea
3
+ size 103972041
onnx/decoder_with_past_model_quantized.onnx CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:b1b12d6b3645f3de6ea5290a821afa4b5eaaa464a752d17f795f7294726a5e3b
3
- size 30280466
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:58e4e0dfc4c0a15cba8c67f356f3dc256c27c5a64f9824006dd9adac73df4994
3
+ size 26651406
quantize_config.json CHANGED
@@ -4,120 +4,120 @@
4
  "per_model_config": {
5
  "encoder_model": {
6
  "op_types": [
 
 
7
  "Transpose",
8
- "Sigmoid",
9
  "Sub",
10
- "MatMul",
11
  "Constant",
12
- "Reshape",
 
 
 
13
  "ConstantOfShape",
14
- "Expand",
15
  "ReduceMean",
16
- "Concat",
17
- "Div",
18
- "Softmax",
19
- "Mul",
20
- "Where",
21
- "Sqrt",
22
- "Equal",
23
  "Range",
24
- "Unsqueeze",
25
- "Cast",
26
  "Add",
27
- "Pow",
28
- "Gather",
29
- "Shape"
 
 
 
30
  ],
31
  "weight_type": "QInt8"
32
  },
33
  "decoder_with_past_model": {
34
  "op_types": [
 
 
35
  "Transpose",
36
- "Sigmoid",
37
  "Sub",
38
- "MatMul",
39
  "Constant",
40
- "Reshape",
 
 
 
41
  "ConstantOfShape",
42
- "Expand",
43
  "ReduceMean",
44
- "Concat",
45
- "Div",
46
- "Softmax",
47
- "Mul",
48
- "Where",
49
- "Sqrt",
50
- "Equal",
51
  "Range",
52
- "Unsqueeze",
53
- "Cast",
54
  "Add",
55
- "Pow",
56
- "Gather",
57
- "Shape"
 
 
 
58
  ],
59
  "weight_type": "QInt8"
60
  },
61
- "decoder_model": {
62
  "op_types": [
63
- "Squeeze",
 
64
  "Transpose",
65
- "Sigmoid",
66
- "Sub",
67
- "MatMul",
68
- "Constant",
69
- "Reshape",
70
- "ConstantOfShape",
71
- "Expand",
72
- "ReduceMean",
73
- "Concat",
74
- "Div",
75
  "Softmax",
76
- "Mul",
77
- "Where",
 
78
  "Less",
 
 
79
  "Equal",
80
  "Sqrt",
 
 
 
 
 
81
  "Range",
82
- "Unsqueeze",
83
- "Cast",
84
  "Add",
85
- "Pow",
86
- "Gather",
87
- "Shape",
88
- "Slice"
 
 
89
  ],
90
  "weight_type": "QInt8"
91
  },
92
- "decoder_model_merged": {
93
  "op_types": [
94
- "If",
95
- "Squeeze",
96
  "Transpose",
97
- "Sigmoid",
98
- "Sub",
99
- "MatMul",
100
- "Constant",
101
- "Reshape",
102
- "ConstantOfShape",
103
- "Expand",
104
- "ReduceMean",
105
- "Concat",
106
- "Div",
107
  "Softmax",
108
- "Mul",
109
- "Where",
110
  "Less",
 
 
111
  "Equal",
112
  "Sqrt",
 
 
 
 
 
113
  "Range",
114
- "Unsqueeze",
115
- "Cast",
116
  "Add",
117
- "Pow",
118
- "Gather",
119
- "Shape",
120
- "Slice"
 
 
121
  ],
122
  "weight_type": "QInt8"
123
  }
 
4
  "per_model_config": {
5
  "encoder_model": {
6
  "op_types": [
7
+ "Gather",
8
+ "Where",
9
  "Transpose",
10
+ "Softmax",
11
  "Sub",
12
+ "Unsqueeze",
13
  "Constant",
14
+ "Equal",
15
+ "Sqrt",
16
+ "Shape",
17
+ "Pow",
18
  "ConstantOfShape",
 
19
  "ReduceMean",
 
 
 
 
 
 
 
20
  "Range",
21
+ "MatMul",
22
+ "Reshape",
23
  "Add",
24
+ "Div",
25
+ "Cast",
26
+ "Concat",
27
+ "Mul",
28
+ "Sigmoid",
29
+ "Expand"
30
  ],
31
  "weight_type": "QInt8"
32
  },
33
  "decoder_with_past_model": {
34
  "op_types": [
35
+ "Gather",
36
+ "Where",
37
  "Transpose",
38
+ "Softmax",
39
  "Sub",
40
+ "Unsqueeze",
41
  "Constant",
42
+ "Equal",
43
+ "Sqrt",
44
+ "Shape",
45
+ "Pow",
46
  "ConstantOfShape",
 
47
  "ReduceMean",
 
 
 
 
 
 
 
48
  "Range",
49
+ "MatMul",
50
+ "Reshape",
51
  "Add",
52
+ "Div",
53
+ "Cast",
54
+ "Concat",
55
+ "Mul",
56
+ "Sigmoid",
57
+ "Expand"
58
  ],
59
  "weight_type": "QInt8"
60
  },
61
+ "decoder_model_merged": {
62
  "op_types": [
63
+ "Gather",
64
+ "Where",
65
  "Transpose",
 
 
 
 
 
 
 
 
 
 
66
  "Softmax",
67
+ "Sub",
68
+ "Unsqueeze",
69
+ "If",
70
  "Less",
71
+ "Constant",
72
+ "Squeeze",
73
  "Equal",
74
  "Sqrt",
75
+ "Shape",
76
+ "Pow",
77
+ "ConstantOfShape",
78
+ "ReduceMean",
79
+ "Slice",
80
  "Range",
81
+ "MatMul",
82
+ "Reshape",
83
  "Add",
84
+ "Div",
85
+ "Cast",
86
+ "Concat",
87
+ "Mul",
88
+ "Sigmoid",
89
+ "Expand"
90
  ],
91
  "weight_type": "QInt8"
92
  },
93
+ "decoder_model": {
94
  "op_types": [
95
+ "Gather",
96
+ "Where",
97
  "Transpose",
 
 
 
 
 
 
 
 
 
 
98
  "Softmax",
99
+ "Sub",
100
+ "Unsqueeze",
101
  "Less",
102
+ "Constant",
103
+ "Squeeze",
104
  "Equal",
105
  "Sqrt",
106
+ "Shape",
107
+ "Pow",
108
+ "ConstantOfShape",
109
+ "ReduceMean",
110
+ "Slice",
111
  "Range",
112
+ "MatMul",
113
+ "Reshape",
114
  "Add",
115
+ "Div",
116
+ "Cast",
117
+ "Concat",
118
+ "Mul",
119
+ "Sigmoid",
120
+ "Expand"
121
  ],
122
  "weight_type": "QInt8"
123
  }
tokenizer.json CHANGED
The diff for this file is too large to render. See raw diff