Xenova HF staff commited on
Commit
4473434
1 Parent(s): 9460a1f

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": 58930
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": 58930
60
  }
generation_config.json CHANGED
@@ -1,5 +1,4 @@
1
  {
2
- "_from_model_config": true,
3
  "bad_words_ids": [
4
  [
5
  58929
@@ -12,5 +11,6 @@
12
  "max_length": 512,
13
  "num_beams": 4,
14
  "pad_token_id": 58929,
15
- "transformers_version": "4.32.0.dev0"
 
16
  }
 
1
  {
 
2
  "bad_words_ids": [
3
  [
4
  58929
 
11
  "max_length": 512,
12
  "num_beams": 4,
13
  "pad_token_id": 58929,
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 sha256:0cfadf5e1e922263a89371e470e367b9b6bdff5012d5e61153c419b1de5f2920
3
- size 343834560
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:61c14210375742dbb2765d9c0dd261e45df9600aaf372e338f60feeacc22d663
3
+ size 223145977
onnx/decoder_model_merged.onnx CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:15188d56ebd5ba46db57bb2b18d36b4e1a3bda24262a1ad68eaaeb591b23c6fd
3
- size 344070042
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:584a5a50e0e134d3c026e68ddd7cd0f346fc2e47a39b624ec4e40cc7585bb597
3
+ size 223381544
onnx/decoder_model_merged_quantized.onnx CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:cee0860865697e4bd058497f4c621a7b5842071cd68d12c2e9f2e5de844484b3
3
- size 87548389
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:871e5ed856e41b95f1a7b32ff4cdb683e1092963bf2300d502298a1fcb4fb0ed
3
+ size 57080167
onnx/decoder_model_quantized.onnx CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:3fa355af99bdebe67703dd86bd70e500dfbec3ba9d71a99dbe026328eb791df1
3
- size 87177033
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:feae6f1b8b5c7219d40338e618537505f1c842f7f122f24aa3dbb49691b8e7af
3
+ size 56709465
onnx/decoder_with_past_model.onnx CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:c4f5f0ff0b0347610bca9acc8ea138010a12b45e880f9e33ae368d05af17c281
3
- size 331184051
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:6d9c89d8347cbf70d62a8aa8385c6973746df24bbd5bd7881545d36ef18ed0ce
3
+ size 210495468
onnx/decoder_with_past_model_quantized.onnx CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:2bbb976dcef67e793e2b9ace946e28204d1b3b6566f6484f61aa380dd8b96c32
3
- size 83905569
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:7cb41e55406661d1ba525b289c2bf5d7c81c4ac842d5a1b2c1ddb13fdaf2c5bc
3
+ size 53438001
quantize_config.json CHANGED
@@ -4,120 +4,120 @@
4
  "per_model_config": {
5
  "encoder_model": {
6
  "op_types": [
7
- "Shape",
 
8
  "Mul",
 
 
 
 
9
  "Pow",
10
- "Softmax",
11
  "MatMul",
12
- "Sub",
13
- "Sigmoid",
 
 
14
  "Equal",
15
  "Add",
 
16
  "Div",
17
- "Concat",
18
- "Sqrt",
19
- "Transpose",
20
- "Expand",
21
- "ConstantOfShape",
22
  "Constant",
23
- "ReduceMean",
24
- "Where",
25
- "Cast",
26
- "Range",
27
  "Reshape",
28
- "Gather",
29
- "Unsqueeze"
 
 
30
  ],
31
  "weight_type": "QInt8"
32
  },
33
  "decoder_with_past_model": {
34
  "op_types": [
35
- "Shape",
 
36
  "Mul",
 
 
 
 
37
  "Pow",
38
- "Softmax",
39
  "MatMul",
40
- "Sub",
41
- "Sigmoid",
 
 
42
  "Equal",
 
43
  "Add",
44
  "Div",
45
- "Concat",
46
- "Sqrt",
47
- "Transpose",
48
- "Expand",
49
- "ConstantOfShape",
50
  "Constant",
51
- "ReduceMean",
52
- "Where",
53
- "Cast",
54
- "Range",
55
  "Reshape",
56
- "Gather",
57
- "Unsqueeze"
 
 
58
  ],
59
  "weight_type": "QInt8"
60
  },
61
- "decoder_model": {
62
  "op_types": [
63
- "Shape",
 
 
64
  "Mul",
 
 
 
 
 
65
  "Pow",
66
- "Softmax",
67
  "MatMul",
68
- "Sub",
69
- "Squeeze",
70
- "Sigmoid",
71
- "Less",
72
- "Slice",
73
  "Equal",
74
  "Add",
 
75
  "Div",
76
- "Concat",
77
- "Sqrt",
78
- "Transpose",
79
- "Expand",
80
- "ConstantOfShape",
81
  "Constant",
82
- "ReduceMean",
83
- "Where",
84
- "Cast",
85
- "Range",
86
  "Reshape",
87
- "Gather",
88
- "Unsqueeze"
 
 
 
 
89
  ],
90
  "weight_type": "QInt8"
91
  },
92
- "decoder_model_merged": {
93
  "op_types": [
94
- "Shape",
 
 
95
  "Mul",
 
 
 
 
 
96
  "Pow",
97
- "Softmax",
98
  "MatMul",
99
- "Sub",
100
- "Squeeze",
101
- "Sigmoid",
102
- "Less",
103
- "Slice",
104
  "Equal",
105
  "Add",
 
106
  "Div",
107
- "Concat",
108
- "Sqrt",
109
- "If",
110
- "Transpose",
111
- "Expand",
112
- "ConstantOfShape",
113
  "Constant",
114
- "ReduceMean",
115
- "Where",
116
- "Cast",
117
- "Range",
118
  "Reshape",
119
- "Gather",
120
- "Unsqueeze"
 
 
 
121
  ],
122
  "weight_type": "QInt8"
123
  }
 
4
  "per_model_config": {
5
  "encoder_model": {
6
  "op_types": [
7
+ "Transpose",
8
+ "ReduceMean",
9
  "Mul",
10
+ "ConstantOfShape",
11
+ "Shape",
12
+ "Range",
13
+ "Concat",
14
  "Pow",
 
15
  "MatMul",
16
+ "Gather",
17
+ "Unsqueeze",
18
+ "Sqrt",
19
+ "Softmax",
20
  "Equal",
21
  "Add",
22
+ "Sub",
23
  "Div",
 
 
 
 
 
24
  "Constant",
 
 
 
 
25
  "Reshape",
26
+ "Cast",
27
+ "Expand",
28
+ "Sigmoid",
29
+ "Where"
30
  ],
31
  "weight_type": "QInt8"
32
  },
33
  "decoder_with_past_model": {
34
  "op_types": [
35
+ "Transpose",
36
+ "ReduceMean",
37
  "Mul",
38
+ "ConstantOfShape",
39
+ "Shape",
40
+ "Range",
41
+ "Concat",
42
  "Pow",
 
43
  "MatMul",
44
+ "Gather",
45
+ "Unsqueeze",
46
+ "Sqrt",
47
+ "Softmax",
48
  "Equal",
49
+ "Sub",
50
  "Add",
51
  "Div",
 
 
 
 
 
52
  "Constant",
 
 
 
 
53
  "Reshape",
54
+ "Cast",
55
+ "Expand",
56
+ "Sigmoid",
57
+ "Where"
58
  ],
59
  "weight_type": "QInt8"
60
  },
61
+ "decoder_model_merged": {
62
  "op_types": [
63
+ "Transpose",
64
+ "Slice",
65
+ "ReduceMean",
66
  "Mul",
67
+ "Less",
68
+ "ConstantOfShape",
69
+ "Shape",
70
+ "Range",
71
+ "Concat",
72
  "Pow",
 
73
  "MatMul",
74
+ "Gather",
75
+ "Unsqueeze",
76
+ "Sqrt",
77
+ "Softmax",
 
78
  "Equal",
79
  "Add",
80
+ "Sub",
81
  "Div",
 
 
 
 
 
82
  "Constant",
 
 
 
 
83
  "Reshape",
84
+ "If",
85
+ "Cast",
86
+ "Squeeze",
87
+ "Expand",
88
+ "Sigmoid",
89
+ "Where"
90
  ],
91
  "weight_type": "QInt8"
92
  },
93
+ "decoder_model": {
94
  "op_types": [
95
+ "Transpose",
96
+ "Slice",
97
+ "ReduceMean",
98
  "Mul",
99
+ "Less",
100
+ "ConstantOfShape",
101
+ "Shape",
102
+ "Range",
103
+ "Concat",
104
  "Pow",
 
105
  "MatMul",
106
+ "Gather",
107
+ "Unsqueeze",
108
+ "Sqrt",
109
+ "Softmax",
 
110
  "Equal",
111
  "Add",
112
+ "Sub",
113
  "Div",
 
 
 
 
 
 
114
  "Constant",
 
 
 
 
115
  "Reshape",
116
+ "Cast",
117
+ "Squeeze",
118
+ "Expand",
119
+ "Sigmoid",
120
+ "Where"
121
  ],
122
  "weight_type": "QInt8"
123
  }
tokenizer.json CHANGED
The diff for this file is too large to render. See raw diff