Xenova HF staff commited on
Commit
581e16a
1 Parent(s): 04d4a27

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": 56434
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": 56434
60
  }
generation_config.json CHANGED
@@ -1,5 +1,4 @@
1
  {
2
- "_from_model_config": true,
3
  "bad_words_ids": [
4
  [
5
  56433
@@ -12,5 +11,6 @@
12
  "max_length": 512,
13
  "num_beams": 6,
14
  "pad_token_id": 56433,
15
- "transformers_version": "4.32.0.dev0"
 
16
  }
 
1
  {
 
2
  "bad_words_ids": [
3
  [
4
  56433
 
11
  "max_length": 512,
12
  "num_beams": 6,
13
  "pad_token_id": 56433,
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 sha256:6d85d949b65e908dcf7af7b5848d343415cd8c62710df3ab1074aee9ce9e7c57
3
- size 333600960
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:b944309a802898654ff633df3253e1e27b0caed0a8a650d410d23fa915b22d82
3
+ size 218024185
onnx/decoder_model_merged.onnx CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:1c24586422327c9c745d5071986e5229faafb88b55ed33015c3b5bdc3bd5e397
3
- size 333836442
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:302d30373d3afef8be538d692b5a30ed447fcfebc91a46c708aad3904c465d5a
3
+ size 218259752
onnx/decoder_model_merged_quantized.onnx CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:a819a7c76954d6e4113e6ded64df0b13193064a9a7c7f0196ca3472515d999ce
3
- size 84970022
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:29973afcd575cc61d17133d432fa1a81d8ba1f4b717b247384c2eb11342a472d
3
+ size 55792232
onnx/decoder_model_quantized.onnx CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:af147403443d7fd5891adf217b5b4f736453552069793ee3e17320469c9d0b53
3
- size 84598666
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:31965a3c359076de33f9676ec0ec88312b34c64a35d648e99f81bf79ca883ad4
3
+ size 55421530
onnx/decoder_with_past_model.onnx CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:53294902e59d8964d5ba25f3f61d1bb4f7524925a2e79746fbcdf2ebc57ce288
3
- size 320950451
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:5b43010f0af60c3c10f89b5b8c067709e2b2860cbe772934da884e7b6d53b1ab
3
+ size 205373676
onnx/decoder_with_past_model_quantized.onnx CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:b4bec4d45701fdf5cea644816fcbf5036c9de97d07ac44077307657cb02375e2
3
- size 81327202
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:b2d6c7231763e09e3abd1c2d95677024f76b3953753a3f2a306e24482d902db1
3
+ size 52150066
quantize_config.json CHANGED
@@ -4,120 +4,120 @@
4
  "per_model_config": {
5
  "encoder_model": {
6
  "op_types": [
 
 
 
 
 
 
7
  "Where",
8
- "Reshape",
9
- "Softmax",
10
  "Equal",
11
- "ReduceMean",
 
 
 
 
 
12
  "Gather",
 
 
13
  "Expand",
14
  "Sqrt",
15
- "Range",
16
- "Sigmoid",
17
- "Concat",
18
- "Pow",
19
- "ConstantOfShape",
20
- "Add",
21
  "Cast",
22
- "Constant",
23
- "MatMul",
24
- "Unsqueeze",
25
- "Sub",
26
- "Transpose",
27
  "Mul",
28
- "Div",
29
- "Shape"
30
  ],
31
  "weight_type": "QInt8"
32
  },
33
  "decoder_with_past_model": {
34
  "op_types": [
 
 
 
 
 
 
35
  "Where",
36
- "Reshape",
37
- "Softmax",
38
  "Equal",
39
- "ReduceMean",
 
 
 
 
 
40
  "Gather",
 
 
41
  "Expand",
42
  "Sqrt",
43
- "Range",
44
- "Sigmoid",
45
- "Concat",
46
- "Pow",
47
- "ConstantOfShape",
48
- "Add",
49
  "Cast",
50
- "Constant",
51
- "MatMul",
52
- "Unsqueeze",
53
- "Sub",
54
- "Transpose",
55
  "Mul",
56
- "Div",
57
- "Shape"
58
  ],
59
  "weight_type": "QInt8"
60
  },
61
- "decoder_model": {
62
  "op_types": [
 
 
 
 
 
 
 
63
  "Where",
64
- "Reshape",
65
- "Softmax",
66
  "Equal",
67
- "ReduceMean",
 
 
 
 
 
 
68
  "Gather",
 
 
69
  "Expand",
70
  "Sqrt",
71
- "Range",
72
- "Sigmoid",
73
- "Concat",
74
- "Pow",
75
- "ConstantOfShape",
76
- "Add",
77
- "Less",
78
  "Cast",
79
- "Slice",
80
- "Constant",
81
- "MatMul",
82
- "Unsqueeze",
83
- "Sub",
84
- "Transpose",
85
  "Mul",
86
- "Squeeze",
87
- "Div",
88
- "Shape"
89
  ],
90
  "weight_type": "QInt8"
91
  },
92
- "decoder_model_merged": {
93
  "op_types": [
 
 
 
 
 
 
94
  "Where",
95
- "If",
96
- "Reshape",
97
- "Softmax",
98
  "Equal",
99
- "ReduceMean",
 
 
 
 
 
 
100
  "Gather",
 
 
101
  "Expand",
102
  "Sqrt",
103
- "Range",
104
- "Sigmoid",
105
- "Concat",
106
- "Pow",
107
- "ConstantOfShape",
108
- "Add",
109
- "Less",
110
  "Cast",
111
- "Slice",
112
- "Constant",
113
- "MatMul",
114
- "Unsqueeze",
115
- "Sub",
116
- "Transpose",
117
  "Mul",
118
- "Squeeze",
119
- "Div",
120
- "Shape"
121
  ],
122
  "weight_type": "QInt8"
123
  }
 
4
  "per_model_config": {
5
  "encoder_model": {
6
  "op_types": [
7
+ "ConstantOfShape",
8
+ "Sigmoid",
9
+ "Transpose",
10
+ "MatMul",
11
+ "Pow",
12
+ "Unsqueeze",
13
  "Where",
 
 
14
  "Equal",
15
+ "Softmax",
16
+ "Add",
17
+ "Concat",
18
+ "Constant",
19
+ "Range",
20
+ "Div",
21
  "Gather",
22
+ "ReduceMean",
23
+ "Sub",
24
  "Expand",
25
  "Sqrt",
 
 
 
 
 
 
26
  "Cast",
27
+ "Shape",
 
 
 
 
28
  "Mul",
29
+ "Reshape"
 
30
  ],
31
  "weight_type": "QInt8"
32
  },
33
  "decoder_with_past_model": {
34
  "op_types": [
35
+ "ConstantOfShape",
36
+ "Sigmoid",
37
+ "Transpose",
38
+ "MatMul",
39
+ "Pow",
40
+ "Unsqueeze",
41
  "Where",
 
 
42
  "Equal",
43
+ "Softmax",
44
+ "Add",
45
+ "Concat",
46
+ "Constant",
47
+ "Range",
48
+ "Div",
49
  "Gather",
50
+ "ReduceMean",
51
+ "Sub",
52
  "Expand",
53
  "Sqrt",
 
 
 
 
 
 
54
  "Cast",
55
+ "Shape",
 
 
 
 
56
  "Mul",
57
+ "Reshape"
 
58
  ],
59
  "weight_type": "QInt8"
60
  },
61
+ "decoder_model_merged": {
62
  "op_types": [
63
+ "If",
64
+ "ConstantOfShape",
65
+ "Sigmoid",
66
+ "Transpose",
67
+ "MatMul",
68
+ "Pow",
69
+ "Unsqueeze",
70
  "Where",
 
 
71
  "Equal",
72
+ "Softmax",
73
+ "Add",
74
+ "Concat",
75
+ "Constant",
76
+ "Range",
77
+ "Div",
78
+ "Slice",
79
  "Gather",
80
+ "ReduceMean",
81
+ "Sub",
82
  "Expand",
83
  "Sqrt",
 
 
 
 
 
 
 
84
  "Cast",
85
+ "Shape",
 
 
 
 
 
86
  "Mul",
87
+ "Reshape",
88
+ "Less",
89
+ "Squeeze"
90
  ],
91
  "weight_type": "QInt8"
92
  },
93
+ "decoder_model": {
94
  "op_types": [
95
+ "ConstantOfShape",
96
+ "Sigmoid",
97
+ "Transpose",
98
+ "MatMul",
99
+ "Pow",
100
+ "Unsqueeze",
101
  "Where",
 
 
 
102
  "Equal",
103
+ "Softmax",
104
+ "Add",
105
+ "Concat",
106
+ "Constant",
107
+ "Range",
108
+ "Div",
109
+ "Slice",
110
  "Gather",
111
+ "ReduceMean",
112
+ "Sub",
113
  "Expand",
114
  "Sqrt",
 
 
 
 
 
 
 
115
  "Cast",
116
+ "Shape",
 
 
 
 
 
117
  "Mul",
118
+ "Reshape",
119
+ "Less",
120
+ "Squeeze"
121
  ],
122
  "weight_type": "QInt8"
123
  }
tokenizer.json CHANGED
The diff for this file is too large to render. See raw diff