Xenova HF staff commited on
Commit
88735d1
1 Parent(s): 03b97ad

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": 62518
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": 62518
60
  }
generation_config.json CHANGED
@@ -1,5 +1,4 @@
1
  {
2
- "_from_model_config": true,
3
  "bad_words_ids": [
4
  [
5
  62517
@@ -12,5 +11,6 @@
12
  "max_length": 512,
13
  "num_beams": 4,
14
  "pad_token_id": 62517,
15
- "transformers_version": "4.32.0.dev0"
 
16
  }
 
1
  {
 
2
  "bad_words_ids": [
3
  [
4
  62517
 
11
  "max_length": 512,
12
  "num_beams": 4,
13
  "pad_token_id": 62517,
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:b52cd4f14c3dd176917bd33db3f0cead8b13912488c38b409e181aa6e4b4492b
3
- size 358545360
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:2014f22836434baa54069071525be0dd2a540e5bff30862c4fa06456262ba234
3
+ size 230508553
onnx/decoder_model_merged.onnx CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:12ab2f2ce821aacfc16a73db855dcbae2e4264a790128d319751d66f75c0307c
3
- size 358780842
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:0686150fa652d177b0a9eee70f81851d921d2ad169535697a109e3c9878be581
3
+ size 230744120
onnx/decoder_model_merged_quantized.onnx CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:a2e3b48cb7ce2f3cbcc5f515431bd80eb14f6f37ac4a89813b061046aeed1226
3
- size 91254794
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:7efefcd793a5663bc204224a92a123524f3dfe2ffad59e5f73d6734d1b11ad04
3
+ size 58931576
onnx/decoder_model_quantized.onnx CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:e0b716fe5a5c22ecc17dd12e18ed424da84776ffa07ffa27a5b02d9fd2a50b68
3
- size 90883438
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:7e4e602c31130176b086cd4bf31d6d08d44a6a367617e910870e0ea3e5408574
3
+ size 58560874
onnx/decoder_with_past_model.onnx CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:566461b2841ecb55692cf8a6c7ebe9acb72cee05b2159d82852d081871c5a733
3
- size 345894851
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:8f6941f135f722de84bf26191c340a4fe7958aa10918872db284a9698795665d
3
+ size 217858044
onnx/decoder_with_past_model_quantized.onnx CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:99f14fe0f43ec1e1e19d8706712ce82ecdf6bfbd088a98f15101b6720d0dc574
3
- size 87611974
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:7946cca821e951629360b4bd12e7965318f777c06c4f0d5443c82197b3b8f169
3
+ size 55289410
quantize_config.json CHANGED
@@ -4,120 +4,120 @@
4
  "per_model_config": {
5
  "encoder_model": {
6
  "op_types": [
7
- "ConstantOfShape",
8
- "Gather",
9
- "Sqrt",
10
- "Reshape",
11
  "Div",
12
- "Expand",
13
  "Range",
14
  "Sub",
15
- "Sigmoid",
16
- "Softmax",
17
- "Pow",
18
- "Shape",
19
- "ReduceMean",
20
  "MatMul",
21
- "Unsqueeze",
22
- "Concat",
23
- "Cast",
24
- "Mul",
25
- "Constant",
26
- "Add",
27
  "Equal",
 
 
 
 
 
 
28
  "Transpose",
29
- "Where"
 
 
 
 
 
30
  ],
31
  "weight_type": "QInt8"
32
  },
33
  "decoder_with_past_model": {
34
  "op_types": [
35
- "ConstantOfShape",
36
- "Gather",
37
- "Sqrt",
38
- "Reshape",
39
  "Div",
 
40
  "Range",
41
- "Expand",
42
  "Sub",
43
- "Sigmoid",
44
- "Softmax",
45
- "Pow",
46
- "Shape",
47
- "ReduceMean",
48
  "MatMul",
49
- "Unsqueeze",
50
- "Concat",
51
- "Cast",
52
- "Mul",
53
- "Constant",
54
- "Add",
55
  "Equal",
 
 
 
 
 
 
56
  "Transpose",
57
- "Where"
 
 
 
 
 
58
  ],
59
  "weight_type": "QInt8"
60
  },
61
- "decoder_model": {
62
  "op_types": [
63
- "Slice",
64
- "ConstantOfShape",
65
- "Gather",
66
- "Sqrt",
67
- "Reshape",
68
  "Div",
69
- "Expand",
70
  "Range",
71
  "Sub",
72
- "Sigmoid",
73
- "Softmax",
74
- "Pow",
75
- "Shape",
76
- "ReduceMean",
77
  "MatMul",
78
- "Unsqueeze",
79
- "Concat",
 
 
80
  "Squeeze",
 
 
81
  "Less",
82
- "Cast",
83
  "Mul",
84
- "Constant",
85
- "Add",
86
- "Equal",
87
  "Transpose",
88
- "Where"
 
 
 
 
 
89
  ],
90
  "weight_type": "QInt8"
91
  },
92
- "decoder_model_merged": {
93
  "op_types": [
94
- "Slice",
95
- "ConstantOfShape",
96
- "Gather",
97
- "Sqrt",
98
- "Reshape",
99
  "Div",
100
- "Expand",
101
  "Range",
102
  "Sub",
103
- "Sigmoid",
104
- "Softmax",
105
- "Pow",
106
- "Shape",
107
- "ReduceMean",
108
  "MatMul",
109
- "Unsqueeze",
110
- "Concat",
 
 
111
  "Squeeze",
 
 
112
  "Less",
113
- "Cast",
114
  "Mul",
115
- "Constant",
116
- "Add",
117
- "Equal",
118
  "Transpose",
119
- "If",
120
- "Where"
 
 
 
 
121
  ],
122
  "weight_type": "QInt8"
123
  }
 
4
  "per_model_config": {
5
  "encoder_model": {
6
  "op_types": [
7
+ "Softmax",
8
+ "Cast",
9
+ "Unsqueeze",
 
10
  "Div",
11
+ "Gather",
12
  "Range",
13
  "Sub",
 
 
 
 
 
14
  "MatMul",
15
+ "Pow",
 
 
 
 
 
16
  "Equal",
17
+ "Sigmoid",
18
+ "Where",
19
+ "Expand",
20
+ "Add",
21
+ "Mul",
22
+ "Concat",
23
  "Transpose",
24
+ "ReduceMean",
25
+ "Sqrt",
26
+ "Constant",
27
+ "Shape",
28
+ "ConstantOfShape",
29
+ "Reshape"
30
  ],
31
  "weight_type": "QInt8"
32
  },
33
  "decoder_with_past_model": {
34
  "op_types": [
35
+ "Softmax",
36
+ "Cast",
37
+ "Unsqueeze",
 
38
  "Div",
39
+ "Gather",
40
  "Range",
 
41
  "Sub",
 
 
 
 
 
42
  "MatMul",
43
+ "Pow",
 
 
 
 
 
44
  "Equal",
45
+ "Sigmoid",
46
+ "Where",
47
+ "Expand",
48
+ "Add",
49
+ "Mul",
50
+ "Concat",
51
  "Transpose",
52
+ "ReduceMean",
53
+ "Sqrt",
54
+ "Constant",
55
+ "Shape",
56
+ "ConstantOfShape",
57
+ "Reshape"
58
  ],
59
  "weight_type": "QInt8"
60
  },
61
+ "decoder_model_merged": {
62
  "op_types": [
63
+ "Softmax",
64
+ "Cast",
65
+ "Unsqueeze",
 
 
66
  "Div",
67
+ "Gather",
68
  "Range",
69
  "Sub",
70
+ "If",
 
 
 
 
71
  "MatMul",
72
+ "Pow",
73
+ "Equal",
74
+ "Sigmoid",
75
+ "Where",
76
  "Squeeze",
77
+ "Expand",
78
+ "Add",
79
  "Less",
 
80
  "Mul",
81
+ "Slice",
82
+ "Concat",
 
83
  "Transpose",
84
+ "ReduceMean",
85
+ "Sqrt",
86
+ "Constant",
87
+ "Shape",
88
+ "ConstantOfShape",
89
+ "Reshape"
90
  ],
91
  "weight_type": "QInt8"
92
  },
93
+ "decoder_model": {
94
  "op_types": [
95
+ "Softmax",
96
+ "Cast",
97
+ "Unsqueeze",
 
 
98
  "Div",
99
+ "Gather",
100
  "Range",
101
  "Sub",
 
 
 
 
 
102
  "MatMul",
103
+ "Pow",
104
+ "Equal",
105
+ "Sigmoid",
106
+ "Where",
107
  "Squeeze",
108
+ "Expand",
109
+ "Add",
110
  "Less",
 
111
  "Mul",
112
+ "Slice",
113
+ "Concat",
 
114
  "Transpose",
115
+ "ReduceMean",
116
+ "Sqrt",
117
+ "Constant",
118
+ "Shape",
119
+ "ConstantOfShape",
120
+ "Reshape"
121
  ],
122
  "weight_type": "QInt8"
123
  }
tokenizer.json CHANGED
The diff for this file is too large to render. See raw diff