Xenova HF staff commited on
Commit
dc7cf45
·
1 Parent(s): a7d6c79

Upload folder using huggingface_hub

Browse files
config.json CHANGED
@@ -53,7 +53,7 @@
53
  "scale_embedding": true,
54
  "share_encoder_decoder_embeddings": true,
55
  "static_position_embeddings": true,
56
- "transformers_version": "4.32.0.dev0",
57
  "use_cache": true,
58
  "vocab_size": 65001
59
  }
 
53
  "scale_embedding": true,
54
  "share_encoder_decoder_embeddings": true,
55
  "static_position_embeddings": true,
56
+ "transformers_version": "4.34.0.dev0",
57
  "use_cache": true,
58
  "vocab_size": 65001
59
  }
generation_config.json CHANGED
@@ -1,5 +1,4 @@
1
  {
2
- "_from_model_config": true,
3
  "bad_words_ids": [
4
  [
5
  65000
@@ -12,5 +11,6 @@
12
  "max_length": 512,
13
  "num_beams": 4,
14
  "pad_token_id": 65000,
15
- "transformers_version": "4.32.0.dev0"
 
16
  }
 
1
  {
 
2
  "bad_words_ids": [
3
  [
4
  65000
 
11
  "max_length": 512,
12
  "num_beams": 4,
13
  "pad_token_id": 65000,
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:efd85b3ea1ad8a7482a306144d48bc5c1c22c5536921c1eb5a02aabd64041b5b
3
- size 368725660
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:8bd0a586aac3cf44f89fe59319b5fd2297968c00dabd4f0c2a5fd227848891e1
3
+ size 235603669
onnx/decoder_model_merged.onnx CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:9b7f52b2928ea2148dadaf7ac79fb9e752c63a8acfab62a9763ba77781abb0f4
3
- size 368961142
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:9010a890e540f69db9a8fa2708b11acc41189b685abb59755ef8383ff5399ce0
3
+ size 235839236
onnx/decoder_model_merged_quantized.onnx CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:14479d793e08b11cda59f3634f490f28ad2fc64f53dfad046853e16a794d9fd6
3
- size 93819733
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:58759988d510656029ef9ea744efdda9fd66dbfb689333abe7fb29bca7b239f3
3
+ size 60212804
onnx/decoder_model_quantized.onnx CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:4cb7e6d213215b29586154c99b17461ef0a9d535a25bb2c6ebcff0d13355db9c
3
- size 93448377
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:4c8c3a0502f385dce8dd2a6b4546aaa69423c7097b5f0b0c77c3f097ce68ddde
3
+ size 59842102
onnx/decoder_with_past_model.onnx CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:464008a53bf8942ad80f8482503d2e84dec241e4a2e92d5009e517f6545d0edb
3
- size 356075151
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:aeda5ce52ef5a2eb77c782d8be8d2e6aac2ab655f41c0164bdef2d1037e0bd9f
3
+ size 222953160
onnx/decoder_with_past_model_quantized.onnx CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:ec0cfa5ad9a54f64db7df4e39c44af79cfd1f4deb428fdf4ae2cc91d342b41ea
3
- size 90176913
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:97c8776c2ee02d070370bf3135ade8d3e4d8f1bd45308c82c160b78972839ed3
3
+ size 56570638
quantize_config.json CHANGED
@@ -4,120 +4,120 @@
4
  "per_model_config": {
5
  "encoder_model": {
6
  "op_types": [
7
- "Concat",
8
- "Equal",
 
 
9
  "Expand",
10
- "Sigmoid",
11
  "Add",
 
 
12
  "Constant",
13
  "MatMul",
 
14
  "Cast",
 
 
15
  "Unsqueeze",
16
- "Reshape",
17
- "Shape",
18
  "Sub",
 
19
  "ConstantOfShape",
20
- "Where",
21
- "Transpose",
22
- "Pow",
23
  "Mul",
24
- "Div",
25
- "Softmax",
26
- "ReduceMean",
27
- "Sqrt",
28
- "Gather",
29
- "Range"
30
  ],
31
  "weight_type": "QInt8"
32
  },
33
  "decoder_with_past_model": {
34
  "op_types": [
35
- "Concat",
36
- "Equal",
 
 
37
  "Expand",
38
- "Sigmoid",
39
  "Add",
 
 
40
  "Constant",
41
  "MatMul",
 
42
  "Cast",
 
 
43
  "Unsqueeze",
44
- "Reshape",
45
- "Shape",
46
  "Sub",
 
47
  "ConstantOfShape",
48
- "Where",
49
- "Transpose",
50
- "Pow",
51
  "Mul",
52
- "Div",
53
- "Softmax",
54
- "ReduceMean",
55
- "Sqrt",
56
- "Gather",
57
- "Range"
58
  ],
59
  "weight_type": "QInt8"
60
  },
61
- "decoder_model": {
62
  "op_types": [
63
- "Concat",
64
- "Equal",
 
 
65
  "Expand",
66
- "Slice",
67
- "Sigmoid",
68
  "Add",
 
 
 
69
  "Constant",
70
- "Less",
71
  "MatMul",
 
 
72
  "Cast",
73
- "Squeeze",
 
74
  "Unsqueeze",
75
- "Reshape",
76
- "Shape",
77
  "Sub",
 
 
78
  "ConstantOfShape",
79
- "Where",
80
- "Transpose",
81
- "Pow",
82
  "Mul",
83
- "Div",
84
- "Softmax",
85
- "ReduceMean",
86
- "Sqrt",
87
- "Gather",
88
- "Range"
89
  ],
90
  "weight_type": "QInt8"
91
  },
92
- "decoder_model_merged": {
93
  "op_types": [
94
- "Concat",
95
- "Equal",
 
 
96
  "Expand",
97
- "Slice",
98
- "Sigmoid",
99
- "If",
100
  "Add",
 
 
 
101
  "Constant",
102
- "Less",
103
  "MatMul",
 
 
104
  "Cast",
105
- "Squeeze",
 
106
  "Unsqueeze",
107
- "Reshape",
108
- "Shape",
109
  "Sub",
 
110
  "ConstantOfShape",
111
- "Where",
112
- "Transpose",
113
- "Pow",
114
  "Mul",
115
- "Div",
116
- "Softmax",
117
- "ReduceMean",
118
- "Sqrt",
119
- "Gather",
120
- "Range"
121
  ],
122
  "weight_type": "QInt8"
123
  }
 
4
  "per_model_config": {
5
  "encoder_model": {
6
  "op_types": [
7
+ "Sqrt",
8
+ "Shape",
9
+ "Div",
10
+ "Reshape",
11
  "Expand",
12
+ "Pow",
13
  "Add",
14
+ "Equal",
15
+ "Where",
16
  "Constant",
17
  "MatMul",
18
+ "ReduceMean",
19
  "Cast",
20
+ "Range",
21
+ "Transpose",
22
  "Unsqueeze",
23
+ "Gather",
 
24
  "Sub",
25
+ "Sigmoid",
26
  "ConstantOfShape",
27
+ "Concat",
 
 
28
  "Mul",
29
+ "Softmax"
 
 
 
 
 
30
  ],
31
  "weight_type": "QInt8"
32
  },
33
  "decoder_with_past_model": {
34
  "op_types": [
35
+ "Sqrt",
36
+ "Shape",
37
+ "Div",
38
+ "Reshape",
39
  "Expand",
40
+ "Pow",
41
  "Add",
42
+ "Equal",
43
+ "Where",
44
  "Constant",
45
  "MatMul",
46
+ "ReduceMean",
47
  "Cast",
48
+ "Range",
49
+ "Transpose",
50
  "Unsqueeze",
51
+ "Gather",
 
52
  "Sub",
53
+ "Sigmoid",
54
  "ConstantOfShape",
55
+ "Concat",
 
 
56
  "Mul",
57
+ "Softmax"
 
 
 
 
 
58
  ],
59
  "weight_type": "QInt8"
60
  },
61
+ "decoder_model_merged": {
62
  "op_types": [
63
+ "Sqrt",
64
+ "Shape",
65
+ "Div",
66
+ "Reshape",
67
  "Expand",
68
+ "Pow",
 
69
  "Add",
70
+ "Equal",
71
+ "Softmax",
72
+ "Where",
73
  "Constant",
 
74
  "MatMul",
75
+ "ReduceMean",
76
+ "Less",
77
  "Cast",
78
+ "Range",
79
+ "Transpose",
80
  "Unsqueeze",
81
+ "Gather",
 
82
  "Sub",
83
+ "If",
84
+ "Sigmoid",
85
  "ConstantOfShape",
86
+ "Squeeze",
87
+ "Concat",
 
88
  "Mul",
89
+ "Slice"
 
 
 
 
 
90
  ],
91
  "weight_type": "QInt8"
92
  },
93
+ "decoder_model": {
94
  "op_types": [
95
+ "Sqrt",
96
+ "Shape",
97
+ "Div",
98
+ "Reshape",
99
  "Expand",
100
+ "Pow",
 
 
101
  "Add",
102
+ "Equal",
103
+ "Softmax",
104
+ "Where",
105
  "Constant",
 
106
  "MatMul",
107
+ "ReduceMean",
108
+ "Less",
109
  "Cast",
110
+ "Range",
111
+ "Transpose",
112
  "Unsqueeze",
113
+ "Gather",
 
114
  "Sub",
115
+ "Sigmoid",
116
  "ConstantOfShape",
117
+ "Squeeze",
118
+ "Concat",
 
119
  "Mul",
120
+ "Slice"
 
 
 
 
 
121
  ],
122
  "weight_type": "QInt8"
123
  }
tokenizer.json CHANGED
The diff for this file is too large to render. See raw diff