Xenova HF staff commited on
Commit
45798ae
·
1 Parent(s): e0f2563

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": 57027
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": 57027
60
  }
generation_config.json CHANGED
@@ -1,5 +1,4 @@
1
  {
2
- "_from_model_config": true,
3
  "bad_words_ids": [
4
  [
5
  57026
@@ -12,5 +11,6 @@
12
  "max_length": 512,
13
  "num_beams": 4,
14
  "pad_token_id": 57026,
15
- "transformers_version": "4.32.0.dev0"
 
16
  }
 
1
  {
 
2
  "bad_words_ids": [
3
  [
4
  57026
 
11
  "max_length": 512,
12
  "num_beams": 4,
13
  "pad_token_id": 57026,
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 sha256:051e9addc94048593a7aa66f85b43eab61a31143b2b75df26299f339dc5ae46f
3
- size 336032260
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:688a07be73040a687fc4ceaa415495b2893cb2f3d55ef2ac132f42a93818917c
3
+ size 219241021
onnx/decoder_model_merged.onnx CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:db660e4987af32e3002a5edae09a32f9f6723687927578e724ff383c77b148a9
3
- size 336267742
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:ea7d14f0dbe21ee7003b9d80766a1f1e37563af015f82897423aff1196d1fc39
3
+ size 219476588
onnx/decoder_model_merged_quantized.onnx CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:a1c543c3bdb86a69fbfc98433ea4a40138a67b4ea02e5e7781064751c980a4b0
3
- size 85582591
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:4d1da147fbe5627314ef69128bde6e9a2fa88e62cc26d2c8182b292e6667aa57
3
+ size 56098220
onnx/decoder_model_quantized.onnx CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:596abab09b6c950a5f608dd33332c3bcae2c48345e7c0d5071891e3e1f842e2c
3
- size 85211235
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:fcabc77cfdb85daddd518daf3e125a4c5f1d0a578d994d08cd423ff4ccfc14b1
3
+ size 55727518
onnx/decoder_with_past_model.onnx CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:cee4067da2932088a07a6c11b8ef7b5eb092e943f1e073310931c35d796a3792
3
- size 323381751
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:25c444beb1e8c8d487f8db7e17607544c73022543eefb7f28da96a4f4c5fa476
3
+ size 206590512
onnx/decoder_with_past_model_quantized.onnx CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:198f28eafb3c1de5942ad92dadf569e49bad78a49130d51e9fa53c2dae037254
3
- size 81939771
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:019c00295a75e749b6d4291bac9ffe6a236ea8db8b12f7e2682903fa220855f1
3
+ size 52456054
quantize_config.json CHANGED
@@ -4,120 +4,120 @@
4
  "per_model_config": {
5
  "encoder_model": {
6
  "op_types": [
7
- "Unsqueeze",
8
- "Range",
9
  "ConstantOfShape",
10
- "Gather",
11
  "Constant",
12
  "Cast",
13
- "Equal",
14
- "Pow",
15
- "Transpose",
16
- "Mul",
17
- "Reshape",
18
- "MatMul",
19
  "ReduceMean",
20
- "Expand",
21
- "Shape",
22
  "Where",
23
- "Div",
 
 
 
24
  "Softmax",
25
  "Concat",
26
- "Sqrt",
27
- "Sub",
28
- "Add",
29
- "Sigmoid"
 
 
 
 
30
  ],
31
  "weight_type": "QInt8"
32
  },
33
  "decoder_with_past_model": {
34
  "op_types": [
35
- "Unsqueeze",
36
- "Range",
37
  "ConstantOfShape",
38
- "Gather",
39
  "Constant",
40
  "Cast",
41
- "Equal",
42
- "Pow",
43
- "Transpose",
44
- "Mul",
45
- "Reshape",
46
- "MatMul",
47
  "ReduceMean",
48
- "Expand",
49
- "Shape",
50
  "Where",
51
- "Div",
 
 
 
 
52
  "Softmax",
53
  "Concat",
54
- "Sqrt",
55
- "Sub",
56
- "Add",
57
- "Sigmoid"
 
 
 
 
58
  ],
59
  "weight_type": "QInt8"
60
  },
61
- "decoder_model": {
62
  "op_types": [
63
- "Squeeze",
64
- "Unsqueeze",
65
- "Range",
66
  "ConstantOfShape",
67
- "Gather",
68
  "Constant",
69
  "Cast",
70
- "Equal",
71
- "Pow",
72
- "Slice",
73
- "Transpose",
74
- "Mul",
75
- "Reshape",
76
- "MatMul",
77
  "ReduceMean",
78
- "Expand",
79
- "Shape",
80
- "Less",
81
  "Where",
82
- "Div",
 
 
 
83
  "Softmax",
84
  "Concat",
85
- "Sqrt",
86
- "Sub",
87
- "Add",
88
- "Sigmoid"
 
 
 
 
 
 
89
  ],
90
  "weight_type": "QInt8"
91
  },
92
- "decoder_model_merged": {
93
  "op_types": [
94
- "Squeeze",
95
- "Unsqueeze",
96
- "Range",
97
  "ConstantOfShape",
98
- "Gather",
99
- "If",
100
  "Constant",
101
  "Cast",
102
- "Equal",
103
- "Pow",
104
- "Slice",
105
- "Transpose",
106
- "Mul",
107
- "Reshape",
108
- "MatMul",
109
  "ReduceMean",
110
- "Expand",
111
- "Shape",
112
- "Less",
113
  "Where",
114
- "Div",
 
 
 
115
  "Softmax",
116
  "Concat",
117
- "Sqrt",
118
- "Sub",
119
- "Add",
120
- "Sigmoid"
 
 
 
 
 
121
  ],
122
  "weight_type": "QInt8"
123
  }
 
4
  "per_model_config": {
5
  "encoder_model": {
6
  "op_types": [
7
+ "Sub",
 
8
  "ConstantOfShape",
 
9
  "Constant",
10
  "Cast",
11
+ "Gather",
 
 
 
 
 
12
  "ReduceMean",
13
+ "Transpose",
14
+ "Add",
15
  "Where",
16
+ "Sigmoid",
17
+ "Sqrt",
18
+ "Expand",
19
+ "Range",
20
  "Softmax",
21
  "Concat",
22
+ "Pow",
23
+ "Reshape",
24
+ "Mul",
25
+ "Shape",
26
+ "Div",
27
+ "Unsqueeze",
28
+ "MatMul",
29
+ "Equal"
30
  ],
31
  "weight_type": "QInt8"
32
  },
33
  "decoder_with_past_model": {
34
  "op_types": [
35
+ "Sub",
 
36
  "ConstantOfShape",
 
37
  "Constant",
38
  "Cast",
39
+ "Gather",
 
 
 
 
 
40
  "ReduceMean",
41
+ "Transpose",
 
42
  "Where",
43
+ "Add",
44
+ "Sigmoid",
45
+ "Sqrt",
46
+ "Expand",
47
+ "Range",
48
  "Softmax",
49
  "Concat",
50
+ "Pow",
51
+ "Reshape",
52
+ "Mul",
53
+ "Shape",
54
+ "Div",
55
+ "Unsqueeze",
56
+ "MatMul",
57
+ "Equal"
58
  ],
59
  "weight_type": "QInt8"
60
  },
61
+ "decoder_model_merged": {
62
  "op_types": [
63
+ "Slice",
64
+ "Less",
65
+ "Sub",
66
  "ConstantOfShape",
 
67
  "Constant",
68
  "Cast",
69
+ "Gather",
 
 
 
 
 
 
70
  "ReduceMean",
71
+ "Transpose",
72
+ "Add",
 
73
  "Where",
74
+ "Sigmoid",
75
+ "Sqrt",
76
+ "Expand",
77
+ "Range",
78
  "Softmax",
79
  "Concat",
80
+ "Pow",
81
+ "Reshape",
82
+ "Mul",
83
+ "Squeeze",
84
+ "Shape",
85
+ "If",
86
+ "Div",
87
+ "Unsqueeze",
88
+ "MatMul",
89
+ "Equal"
90
  ],
91
  "weight_type": "QInt8"
92
  },
93
+ "decoder_model": {
94
  "op_types": [
95
+ "Slice",
96
+ "Less",
97
+ "Sub",
98
  "ConstantOfShape",
 
 
99
  "Constant",
100
  "Cast",
101
+ "Gather",
 
 
 
 
 
 
102
  "ReduceMean",
103
+ "Transpose",
104
+ "Add",
 
105
  "Where",
106
+ "Sigmoid",
107
+ "Sqrt",
108
+ "Expand",
109
+ "Range",
110
  "Softmax",
111
  "Concat",
112
+ "Pow",
113
+ "Reshape",
114
+ "Mul",
115
+ "Squeeze",
116
+ "Shape",
117
+ "Div",
118
+ "Unsqueeze",
119
+ "MatMul",
120
+ "Equal"
121
  ],
122
  "weight_type": "QInt8"
123
  }
tokenizer.json CHANGED
The diff for this file is too large to render. See raw diff