Xenova HF staff commited on
Commit
f3febda
1 Parent(s): a94e4b7

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": 64172
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": 64172
60
  }
generation_config.json CHANGED
@@ -1,5 +1,4 @@
1
  {
2
- "_from_model_config": true,
3
  "bad_words_ids": [
4
  [
5
  64171
@@ -12,5 +11,6 @@
12
  "max_length": 512,
13
  "num_beams": 6,
14
  "pad_token_id": 64171,
15
- "transformers_version": "4.32.0.dev0"
 
16
  }
 
1
  {
 
2
  "bad_words_ids": [
3
  [
4
  64171
 
11
  "max_length": 512,
12
  "num_beams": 6,
13
  "pad_token_id": 64171,
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:cd3de45f8fdde535bf41f3dccc59178b8bce1261ee612fd7624b721648cb0d9f
3
- size 365326760
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:7c2c73c173eb5ddda454045a18a8b65e4a4d23e66b4807750dcb49da4e4b30dc
3
+ size 233902561
onnx/decoder_model_merged.onnx CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:85c39cfc7b6d8961aa41abce3c3bffb36410e4ffc076e8eefe0c6bd6e55bb673
3
- size 365562242
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:37b5faee83ca9fd550f15dee02bb728985e761ff5c7c419883d825f6169ec5f8
3
+ size 234138128
onnx/decoder_model_merged_quantized.onnx CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:37f1da6c7d89948bff44e4e40ca1b35375b512571f2f025f54428bdc4cb91c1a
3
- size 92963376
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:6add167a0cd3f78aa298b8f927d2e8645e33cb17df11e92967f0c5b5703c8c4d
3
+ size 59785040
onnx/decoder_model_quantized.onnx CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:f264e657940ee7dc16020d45ec2f05f54674e1eb329fb5d14b6c24ec3d201e87
3
- size 92592020
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:8ac053f1ef3db2dd9013fe66bfc9e30c0a762457ec11610f1e6b0a5cab9ac8d7
3
+ size 59414338
onnx/decoder_with_past_model.onnx CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:a599332dad1b9ee934588db22759740d8b37e1041e19cafb3335e9b1992bf308
3
- size 352676251
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:665a8687082d0ac0d1c0d1b0dd7f35f08f9ecadd9a7187137c142a383867a0c2
3
+ size 221252052
onnx/decoder_with_past_model_quantized.onnx CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:81fde1f1afcf8635b296c9b32131d94fcaac42f725603bb1e0d26d284c07d7a9
3
- size 89320556
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:ba684d13ac8b91a499b5a7886a5d9988e46d17bbb15d9487e03906c0d72fabde
3
+ size 56142874
quantize_config.json CHANGED
@@ -2,122 +2,122 @@
2
  "per_channel": true,
3
  "reduce_range": true,
4
  "per_model_config": {
5
- "encoder_model": {
6
  "op_types": [
7
- "Sigmoid",
8
- "Sqrt",
9
  "Pow",
10
- "Sub",
11
  "Softmax",
12
- "Reshape",
13
- "Where",
14
- "Shape",
15
- "Div",
16
- "Transpose",
17
  "MatMul",
18
- "Unsqueeze",
19
- "ConstantOfShape",
20
- "Equal",
21
- "ReduceMean",
22
- "Expand",
23
- "Cast",
24
  "Mul",
25
- "Constant",
26
- "Gather",
27
- "Add",
28
  "Concat",
29
- "Range"
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
30
  ],
31
  "weight_type": "QInt8"
32
  },
33
- "decoder_with_past_model": {
34
  "op_types": [
35
- "Sigmoid",
36
- "Sqrt",
37
  "Pow",
38
- "Sub",
39
  "Softmax",
40
- "Reshape",
41
- "Where",
42
- "Shape",
43
- "Div",
44
- "Transpose",
45
  "MatMul",
46
- "Unsqueeze",
47
- "ConstantOfShape",
48
- "Equal",
49
- "ReduceMean",
50
- "Expand",
51
- "Cast",
52
  "Mul",
53
- "Constant",
54
- "Gather",
55
- "Add",
56
  "Concat",
57
- "Range"
 
 
 
 
 
 
 
 
 
 
 
 
 
 
58
  ],
59
  "weight_type": "QInt8"
60
  },
61
- "decoder_model": {
62
  "op_types": [
63
- "Sigmoid",
64
- "Sqrt",
65
  "Pow",
66
- "Sub",
67
  "Softmax",
68
- "Reshape",
69
- "Where",
70
- "Shape",
71
- "Div",
72
- "Transpose",
73
- "Squeeze",
74
  "MatMul",
75
- "Unsqueeze",
76
- "ConstantOfShape",
77
- "Equal",
78
- "ReduceMean",
79
- "Expand",
80
- "Cast",
81
  "Mul",
82
- "Constant",
 
 
 
83
  "Slice",
 
 
 
 
84
  "Gather",
85
- "Add",
86
- "Concat",
87
- "Less",
88
- "Range"
 
 
 
 
 
89
  ],
90
  "weight_type": "QInt8"
91
  },
92
- "decoder_model_merged": {
93
  "op_types": [
94
- "Sigmoid",
95
- "Sqrt",
96
  "Pow",
97
- "Sub",
98
  "Softmax",
99
- "Reshape",
100
- "Where",
101
- "Shape",
102
- "Div",
103
- "Transpose",
104
- "Squeeze",
105
  "MatMul",
106
- "Unsqueeze",
107
- "ConstantOfShape",
108
- "Equal",
109
- "ReduceMean",
110
- "Expand",
111
- "Cast",
112
- "If",
113
  "Mul",
114
- "Constant",
115
- "Slice",
116
- "Gather",
117
- "Add",
118
  "Concat",
119
- "Less",
120
- "Range"
 
 
 
 
 
 
 
 
 
 
 
 
 
121
  ],
122
  "weight_type": "QInt8"
123
  }
 
2
  "per_channel": true,
3
  "reduce_range": true,
4
  "per_model_config": {
5
+ "decoder_model": {
6
  "op_types": [
7
+ "Transpose",
8
+ "Expand",
9
  "Pow",
 
10
  "Softmax",
 
 
 
 
 
11
  "MatMul",
12
+ "Sub",
13
+ "Less",
 
 
 
 
14
  "Mul",
 
 
 
15
  "Concat",
16
+ "Sigmoid",
17
+ "Add",
18
+ "Slice",
19
+ "Div",
20
+ "Cast",
21
+ "ReduceMean",
22
+ "Equal",
23
+ "Gather",
24
+ "Where",
25
+ "ConstantOfShape",
26
+ "Range",
27
+ "Squeeze",
28
+ "Sqrt",
29
+ "Reshape",
30
+ "Shape",
31
+ "Constant",
32
+ "Unsqueeze"
33
  ],
34
  "weight_type": "QInt8"
35
  },
36
+ "encoder_model": {
37
  "op_types": [
38
+ "Transpose",
39
+ "Expand",
40
  "Pow",
 
41
  "Softmax",
 
 
 
 
 
42
  "MatMul",
43
+ "Sub",
 
 
 
 
 
44
  "Mul",
 
 
 
45
  "Concat",
46
+ "Sigmoid",
47
+ "Add",
48
+ "Div",
49
+ "Cast",
50
+ "ReduceMean",
51
+ "Equal",
52
+ "Gather",
53
+ "Where",
54
+ "ConstantOfShape",
55
+ "Range",
56
+ "Sqrt",
57
+ "Reshape",
58
+ "Shape",
59
+ "Constant",
60
+ "Unsqueeze"
61
  ],
62
  "weight_type": "QInt8"
63
  },
64
+ "decoder_model_merged": {
65
  "op_types": [
66
+ "Transpose",
67
+ "Expand",
68
  "Pow",
 
69
  "Softmax",
 
 
 
 
 
 
70
  "MatMul",
71
+ "Sub",
72
+ "Less",
 
 
 
 
73
  "Mul",
74
+ "If",
75
+ "Concat",
76
+ "Sigmoid",
77
+ "Add",
78
  "Slice",
79
+ "Div",
80
+ "Cast",
81
+ "ReduceMean",
82
+ "Equal",
83
  "Gather",
84
+ "Where",
85
+ "ConstantOfShape",
86
+ "Range",
87
+ "Squeeze",
88
+ "Sqrt",
89
+ "Reshape",
90
+ "Shape",
91
+ "Constant",
92
+ "Unsqueeze"
93
  ],
94
  "weight_type": "QInt8"
95
  },
96
+ "decoder_with_past_model": {
97
  "op_types": [
98
+ "Transpose",
99
+ "Expand",
100
  "Pow",
 
101
  "Softmax",
 
 
 
 
 
 
102
  "MatMul",
103
+ "Sub",
 
 
 
 
 
 
104
  "Mul",
 
 
 
 
105
  "Concat",
106
+ "Sigmoid",
107
+ "Add",
108
+ "Div",
109
+ "Cast",
110
+ "ReduceMean",
111
+ "Equal",
112
+ "Gather",
113
+ "Where",
114
+ "ConstantOfShape",
115
+ "Range",
116
+ "Sqrt",
117
+ "Reshape",
118
+ "Shape",
119
+ "Constant",
120
+ "Unsqueeze"
121
  ],
122
  "weight_type": "QInt8"
123
  }
tokenizer.json CHANGED
The diff for this file is too large to render. See raw diff