Xenova HF staff commited on
Commit
18c5d69
1 Parent(s): 6376534

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": 62802
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": 62802
60
  }
generation_config.json CHANGED
@@ -1,5 +1,4 @@
1
  {
2
- "_from_model_config": true,
3
  "bad_words_ids": [
4
  [
5
  62801
@@ -12,5 +11,6 @@
12
  "max_length": 512,
13
  "num_beams": 4,
14
  "pad_token_id": 62801,
15
- "transformers_version": "4.32.0.dev0"
 
16
  }
 
1
  {
 
2
  "bad_words_ids": [
3
  [
4
  62801
 
11
  "max_length": 512,
12
  "num_beams": 4,
13
  "pad_token_id": 62801,
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 sha256:7eb49039a175bc6466ad5307e2b7040c4b914761e2f7c83138a35f4bdba6ff15
3
- size 359709760
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:07c929174708c515c7b88a990cd8342b49d167f176d528060c73782101fe4d72
3
+ size 231091321
onnx/decoder_model_merged.onnx CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:442cb0cc6ad76aaa284f19d2de626b3b46b6871103fbf93e385305b74633d64a
3
- size 359945242
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:eaaff5593905594e6712d18b8fcdf22da0a6320dea7f0b3696c362b823dd7876
3
+ size 231326888
onnx/decoder_model_merged_quantized.onnx CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:3fa113495604aebe8357d0274c75fd0fa008f5f255c99995869086bf276f67cd
3
- size 91548166
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:c0e553b0c26611780c70417c9b4c1ca7169e9cdb18ba9e649197efb017af9f02
3
+ size 59078120
onnx/decoder_model_quantized.onnx CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:0b6b843f0d4b3438b3341451b079c7fafce89c2fc3d6f29ae50e89b57dc14385
3
- size 91176810
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:94e6bdf90866151787d014b7b77be0adb0f036418ee04d235ec01238aa8426b9
3
+ size 58707418
onnx/decoder_with_past_model.onnx CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:7afb0044dcc5a48d2115fc8efc4e45746beb6853f4422de61991ae2b4cf8fdff
3
- size 347059251
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:e8f30d358f0360a9ee61fbd7b3089f9f29af7edfde787866e024caacf55d9f4a
3
+ size 218440812
onnx/decoder_with_past_model_quantized.onnx CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:1bfc93dd0d70f0d022eec33d25b7140f13dfd426cac8a8c290070ad9ee7b98f3
3
- size 87905346
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:35596df81744a00bd1b591d449ad33d5131b2fef7af38d2a6f795e1a98ba30c6
3
+ size 55435954
quantize_config.json CHANGED
@@ -4,120 +4,120 @@
4
  "per_model_config": {
5
  "encoder_model": {
6
  "op_types": [
7
- "Unsqueeze",
8
- "Gather",
9
- "Sub",
10
- "Expand",
11
- "Equal",
12
- "Constant",
13
  "Add",
14
- "Sigmoid",
15
  "Shape",
 
 
 
 
 
16
  "Where",
 
17
  "ReduceMean",
 
 
 
 
 
 
18
  "Reshape",
19
- "Softmax",
20
- "MatMul",
21
- "Sqrt",
22
- "Pow",
23
  "Mul",
24
- "Div",
25
- "Range",
26
- "ConstantOfShape",
27
- "Concat",
28
- "Cast",
29
- "Transpose"
30
  ],
31
  "weight_type": "QInt8"
32
  },
33
  "decoder_with_past_model": {
34
  "op_types": [
35
- "Unsqueeze",
36
- "Gather",
37
- "Sub",
38
- "Expand",
39
- "Equal",
40
- "Constant",
41
  "Add",
42
- "Sigmoid",
43
  "Shape",
 
 
 
 
 
44
  "Where",
 
45
  "ReduceMean",
 
 
 
 
 
 
46
  "Reshape",
47
- "Softmax",
48
- "MatMul",
49
- "Sqrt",
50
- "Pow",
51
  "Mul",
52
- "Div",
53
- "Range",
54
- "ConstantOfShape",
55
- "Concat",
56
- "Cast",
57
- "Transpose"
58
  ],
59
  "weight_type": "QInt8"
60
  },
61
- "decoder_model": {
62
  "op_types": [
63
- "Less",
64
- "Unsqueeze",
65
- "Gather",
66
- "Sub",
67
- "Expand",
68
- "Equal",
69
- "Constant",
70
  "Add",
71
- "Sigmoid",
72
  "Shape",
 
 
 
 
 
 
 
73
  "Where",
 
74
  "ReduceMean",
 
 
 
 
 
 
 
 
75
  "Reshape",
76
- "Softmax",
77
- "MatMul",
78
- "Sqrt",
79
- "Pow",
80
  "Mul",
81
- "Div",
82
- "Squeeze",
83
- "Range",
84
- "ConstantOfShape",
85
- "Concat",
86
- "Cast",
87
- "Slice",
88
- "Transpose"
89
  ],
90
  "weight_type": "QInt8"
91
  },
92
- "decoder_model_merged": {
93
  "op_types": [
94
- "Less",
95
- "Unsqueeze",
96
- "Gather",
97
- "Sub",
98
- "Expand",
99
- "Equal",
100
- "Constant",
101
- "If",
102
  "Add",
103
- "Sigmoid",
104
  "Shape",
 
 
 
 
 
 
 
105
  "Where",
 
106
  "ReduceMean",
 
 
 
 
 
 
 
107
  "Reshape",
108
- "Softmax",
109
- "MatMul",
110
- "Sqrt",
111
- "Pow",
112
  "Mul",
113
- "Div",
114
- "Squeeze",
115
- "Range",
116
- "ConstantOfShape",
117
- "Concat",
118
- "Cast",
119
- "Slice",
120
- "Transpose"
121
  ],
122
  "weight_type": "QInt8"
123
  }
 
4
  "per_model_config": {
5
  "encoder_model": {
6
  "op_types": [
7
+ "Concat",
 
 
 
 
 
8
  "Add",
9
+ "Cast",
10
  "Shape",
11
+ "ConstantOfShape",
12
+ "Equal",
13
+ "MatMul",
14
+ "Sqrt",
15
+ "Transpose",
16
  "Where",
17
+ "Gather",
18
  "ReduceMean",
19
+ "Sub",
20
+ "Unsqueeze",
21
+ "Range",
22
+ "Sigmoid",
23
+ "Constant",
24
+ "Div",
25
  "Reshape",
 
 
 
 
26
  "Mul",
27
+ "Expand",
28
+ "Softmax",
29
+ "Pow"
 
 
 
30
  ],
31
  "weight_type": "QInt8"
32
  },
33
  "decoder_with_past_model": {
34
  "op_types": [
35
+ "Concat",
 
 
 
 
 
36
  "Add",
37
+ "Cast",
38
  "Shape",
39
+ "ConstantOfShape",
40
+ "Equal",
41
+ "MatMul",
42
+ "Sqrt",
43
+ "Transpose",
44
  "Where",
45
+ "Gather",
46
  "ReduceMean",
47
+ "Sub",
48
+ "Unsqueeze",
49
+ "Range",
50
+ "Sigmoid",
51
+ "Constant",
52
+ "Div",
53
  "Reshape",
 
 
 
 
54
  "Mul",
55
+ "Expand",
56
+ "Softmax",
57
+ "Pow"
 
 
 
58
  ],
59
  "weight_type": "QInt8"
60
  },
61
+ "decoder_model_merged": {
62
  "op_types": [
63
+ "Concat",
 
 
 
 
 
 
64
  "Add",
65
+ "Cast",
66
  "Shape",
67
+ "ConstantOfShape",
68
+ "Equal",
69
+ "MatMul",
70
+ "Sqrt",
71
+ "Slice",
72
+ "Less",
73
+ "Transpose",
74
  "Where",
75
+ "Gather",
76
  "ReduceMean",
77
+ "Sub",
78
+ "Unsqueeze",
79
+ "Range",
80
+ "Sigmoid",
81
+ "Squeeze",
82
+ "If",
83
+ "Constant",
84
+ "Div",
85
  "Reshape",
 
 
 
 
86
  "Mul",
87
+ "Expand",
88
+ "Softmax",
89
+ "Pow"
 
 
 
 
 
90
  ],
91
  "weight_type": "QInt8"
92
  },
93
+ "decoder_model": {
94
  "op_types": [
95
+ "Concat",
 
 
 
 
 
 
 
96
  "Add",
97
+ "Cast",
98
  "Shape",
99
+ "ConstantOfShape",
100
+ "Equal",
101
+ "MatMul",
102
+ "Sqrt",
103
+ "Slice",
104
+ "Less",
105
+ "Transpose",
106
  "Where",
107
+ "Gather",
108
  "ReduceMean",
109
+ "Sub",
110
+ "Unsqueeze",
111
+ "Range",
112
+ "Sigmoid",
113
+ "Squeeze",
114
+ "Constant",
115
+ "Div",
116
  "Reshape",
 
 
 
 
117
  "Mul",
118
+ "Expand",
119
+ "Softmax",
120
+ "Pow"
 
 
 
 
 
121
  ],
122
  "weight_type": "QInt8"
123
  }
tokenizer.json CHANGED
The diff for this file is too large to render. See raw diff