Xenova HF staff commited on
Commit
74e1b53
1 Parent(s): e4442ed

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": 62389
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": 62389
60
  }
generation_config.json CHANGED
@@ -1,5 +1,4 @@
1
  {
2
- "_from_model_config": true,
3
  "bad_words_ids": [
4
  [
5
  62388
@@ -12,5 +11,6 @@
12
  "max_length": 512,
13
  "num_beams": 6,
14
  "pad_token_id": 62388,
15
- "transformers_version": "4.32.0.dev0"
 
16
  }
 
1
  {
 
2
  "bad_words_ids": [
3
  [
4
  62388
 
11
  "max_length": 512,
12
  "num_beams": 6,
13
  "pad_token_id": 62388,
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 sha256:0ca6e09fea9157191b2a5e317ba53d6db4100bacfe491f48776dbe24acba1e12
3
- size 358016460
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:0da4b2f54359971b40ffc3cc7b9091f6faf9f4660f5345ef8372b740c60bb33b
3
+ size 230243845
onnx/decoder_model_merged.onnx CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:947e6acf243694e16b52905ae8f7ee3491ef9f7f93c8514dd29bb9e3025648df
3
- size 358251942
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:fb5600bc5abc3467e7ac97fdeef9cab24be6e5dcec500190fa30ea411877fada
3
+ size 230479412
onnx/decoder_model_merged_quantized.onnx CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:8d57bcabcdc6824551a08cdfb8bb77c99c2e2d00a9de9d57e1244c897bdecf49
3
- size 91121537
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:b32325078cf0370f3741a2b22f2bc11a94e219c6ecc56a27eaee1e0f40e24b04
3
+ size 58865012
onnx/decoder_model_quantized.onnx CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:bf77d257ff05bfcfb84e80560b5eb0d1f47eb818fc2aba8c82d4d6bd7d0a2c21
3
- size 90750181
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:70a8a968a4fe8cbf75733fb287957e53c204f17513ff7445e5a68e66874f1ec4
3
+ size 58494310
onnx/decoder_with_past_model.onnx CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:d2ef457004af425130e8b26f03a08d029ecaaabdcf488546570503573feec14b
3
- size 345365951
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:e24da5c9207fd603c71758f43027e2ff128bf135aff0bc31bae8d1ecf5f7672c
3
+ size 217593336
onnx/decoder_with_past_model_quantized.onnx CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:41ada98f277b2adae08e5ad14af8ad651fc884830e297da0752fdfb134ecfbf3
3
- size 87478717
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:3a58984a99bd23abe1e110f0e4adb340b65d15091a62d88103ef2d2051fd97e0
3
+ size 55222846
quantize_config.json CHANGED
@@ -4,120 +4,120 @@
4
  "per_model_config": {
5
  "encoder_model": {
6
  "op_types": [
7
- "Pow",
8
- "Cast",
9
- "Transpose",
10
- "Add",
11
- "Shape",
12
- "Mul",
13
  "Reshape",
 
 
14
  "Sqrt",
 
 
 
 
 
 
 
 
15
  "Expand",
 
16
  "Range",
17
  "Constant",
18
- "Gather",
 
19
  "Unsqueeze",
20
- "ConstantOfShape",
21
- "Sigmoid",
22
  "Div",
23
- "Sub",
24
- "Equal",
25
- "ReduceMean",
26
- "Where",
27
- "Softmax",
28
- "Concat",
29
- "MatMul"
30
  ],
31
  "weight_type": "QInt8"
32
  },
33
  "decoder_with_past_model": {
34
  "op_types": [
35
- "Pow",
36
- "Cast",
37
- "Transpose",
38
- "Add",
39
- "Shape",
40
- "Mul",
41
  "Reshape",
 
 
42
  "Sqrt",
 
 
 
 
 
 
 
 
43
  "Expand",
 
44
  "Range",
45
  "Constant",
46
- "Gather",
 
47
  "Unsqueeze",
48
- "ConstantOfShape",
49
- "Sigmoid",
50
  "Div",
51
- "Sub",
52
- "Equal",
53
- "ReduceMean",
54
- "Where",
55
- "Softmax",
56
- "Concat",
57
- "MatMul"
58
  ],
59
  "weight_type": "QInt8"
60
  },
61
- "decoder_model": {
62
  "op_types": [
63
- "Pow",
 
 
 
 
 
 
64
  "Cast",
65
- "Transpose",
 
 
66
  "Slice",
67
- "Add",
68
  "Shape",
69
- "Mul",
70
- "Squeeze",
71
- "Reshape",
72
- "Sqrt",
73
  "Expand",
 
74
  "Range",
75
  "Constant",
76
- "Gather",
 
77
  "Unsqueeze",
78
- "ConstantOfShape",
79
- "Sigmoid",
80
  "Div",
81
- "Sub",
82
- "Equal",
83
- "ReduceMean",
84
- "Where",
85
- "Less",
86
- "Softmax",
87
- "Concat",
88
- "MatMul"
89
  ],
90
  "weight_type": "QInt8"
91
  },
92
- "decoder_model_merged": {
93
  "op_types": [
94
- "Pow",
 
 
 
 
 
 
95
  "Cast",
96
- "Transpose",
 
 
97
  "Slice",
98
- "Add",
99
  "Shape",
100
- "Mul",
101
- "Squeeze",
102
- "Reshape",
103
- "Sqrt",
104
  "Expand",
 
105
  "Range",
106
  "Constant",
107
- "Gather",
 
108
  "Unsqueeze",
109
- "ConstantOfShape",
110
- "Sigmoid",
111
  "Div",
112
- "Sub",
113
- "If",
114
- "Equal",
115
- "ReduceMean",
116
- "Where",
117
- "Less",
118
- "Softmax",
119
- "Concat",
120
- "MatMul"
121
  ],
122
  "weight_type": "QInt8"
123
  }
 
4
  "per_model_config": {
5
  "encoder_model": {
6
  "op_types": [
 
 
 
 
 
 
7
  "Reshape",
8
+ "Concat",
9
+ "Sigmoid",
10
  "Sqrt",
11
+ "Equal",
12
+ "ReduceMean",
13
+ "MatMul",
14
+ "Cast",
15
+ "Softmax",
16
+ "Mul",
17
+ "Shape",
18
+ "Sub",
19
  "Expand",
20
+ "Where",
21
  "Range",
22
  "Constant",
23
+ "Add",
24
+ "Transpose",
25
  "Unsqueeze",
 
 
26
  "Div",
27
+ "Pow",
28
+ "Gather",
29
+ "ConstantOfShape"
 
 
 
 
30
  ],
31
  "weight_type": "QInt8"
32
  },
33
  "decoder_with_past_model": {
34
  "op_types": [
 
 
 
 
 
 
35
  "Reshape",
36
+ "Concat",
37
+ "Sigmoid",
38
  "Sqrt",
39
+ "Equal",
40
+ "ReduceMean",
41
+ "MatMul",
42
+ "Cast",
43
+ "Softmax",
44
+ "Mul",
45
+ "Shape",
46
+ "Sub",
47
  "Expand",
48
+ "Where",
49
  "Range",
50
  "Constant",
51
+ "Add",
52
+ "Transpose",
53
  "Unsqueeze",
 
 
54
  "Div",
55
+ "Pow",
56
+ "Gather",
57
+ "ConstantOfShape"
 
 
 
 
58
  ],
59
  "weight_type": "QInt8"
60
  },
61
+ "decoder_model_merged": {
62
  "op_types": [
63
+ "Reshape",
64
+ "Concat",
65
+ "Sigmoid",
66
+ "Sqrt",
67
+ "Equal",
68
+ "ReduceMean",
69
+ "MatMul",
70
  "Cast",
71
+ "Softmax",
72
+ "Mul",
73
+ "Less",
74
  "Slice",
 
75
  "Shape",
76
+ "Sub",
 
 
 
77
  "Expand",
78
+ "Where",
79
  "Range",
80
  "Constant",
81
+ "Add",
82
+ "Transpose",
83
  "Unsqueeze",
84
+ "Squeeze",
 
85
  "Div",
86
+ "Pow",
87
+ "Gather",
88
+ "ConstantOfShape",
89
+ "If"
 
 
 
 
90
  ],
91
  "weight_type": "QInt8"
92
  },
93
+ "decoder_model": {
94
  "op_types": [
95
+ "Reshape",
96
+ "Concat",
97
+ "Sigmoid",
98
+ "Sqrt",
99
+ "Equal",
100
+ "ReduceMean",
101
+ "MatMul",
102
  "Cast",
103
+ "Softmax",
104
+ "Mul",
105
+ "Less",
106
  "Slice",
 
107
  "Shape",
108
+ "Sub",
 
 
 
109
  "Expand",
110
+ "Where",
111
  "Range",
112
  "Constant",
113
+ "Add",
114
+ "Transpose",
115
  "Unsqueeze",
116
+ "Squeeze",
 
117
  "Div",
118
+ "Pow",
119
+ "Gather",
120
+ "ConstantOfShape"
 
 
 
 
 
 
121
  ],
122
  "weight_type": "QInt8"
123
  }
tokenizer.json CHANGED
The diff for this file is too large to render. See raw diff