Xenova HF staff commited on
Commit
a4477fe
1 Parent(s): 63e5e2b

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": 59056
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": 59056
60
  }
generation_config.json CHANGED
@@ -1,5 +1,4 @@
1
  {
2
- "_from_model_config": true,
3
  "bad_words_ids": [
4
  [
5
  59055
@@ -12,5 +11,6 @@
12
  "max_length": 512,
13
  "num_beams": 4,
14
  "pad_token_id": 59055,
15
- "transformers_version": "4.32.0.dev0"
 
16
  }
 
1
  {
 
2
  "bad_words_ids": [
3
  [
4
  59055
 
11
  "max_length": 512,
12
  "num_beams": 4,
13
  "pad_token_id": 59055,
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 sha256:56e1503c9341ab4cc3ecc9865be56fc469661c11ac11ee0e81b06f46cf0a33ed
3
- size 344351160
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:bf8e71b2cd58afe5d2f78fe579da22818350fcb6d8db0bd337bf886835cfb02f
3
+ size 223404529
onnx/decoder_model_merged.onnx CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:b19a600fdb012fa0b8f342b1ce1b344916fc124fbbd3f742ebc239cf5ba85c9c
3
- size 344586642
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:6641b0b1063c752c9db2899775cfba6f0691da2a8b1167402a9e9e66eb4a1051
3
+ size 223640096
onnx/decoder_model_merged_quantized.onnx CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:95050dbf22de26a787fdeef9576381a02b811c803a06ef0a9dbb5d96a133cf6e
3
- size 87678548
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:28fe02c5ada48511d856c655f49a03492dc62f1c8e92a80c96adeb29afd63d27
3
+ size 57145184
onnx/decoder_model_quantized.onnx CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:8f63fc4e1ef121b1a49480992bd1d250555676cd394dd05380464ebd30741e45
3
- size 87307192
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:d3bbbac0a4421a4378aabc600b82410601557e9f6199162657e9a4fcfb825a48
3
+ size 56774482
onnx/decoder_with_past_model.onnx CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:d37fa906412afa68e3376fb70e7487fa683d28716cef343febf7f194b8740c23
3
- size 331700651
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:4eeb96a389de040716858570432c937421c5fd5b91490b35abfffb37da871def
3
+ size 210754020
onnx/decoder_with_past_model_quantized.onnx CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:c895353f4cdcfe0d0527b7a6bd5f03bbbc7725d421411ce51399105efb1c0d72
3
- size 84035728
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:448eb4b1098bfdb357d307bd3dd988571042cd1bf73b8a9b83315cd14469972d
3
+ size 53503018
quantize_config.json CHANGED
@@ -4,120 +4,120 @@
4
  "per_model_config": {
5
  "encoder_model": {
6
  "op_types": [
7
- "ConstantOfShape",
8
- "Reshape",
 
 
9
  "Pow",
10
- "Where",
 
11
  "Cast",
 
 
12
  "Mul",
13
- "ReduceMean",
14
- "Sigmoid",
15
- "Constant",
16
  "Div",
17
- "Expand",
18
- "Sub",
19
- "Unsqueeze",
20
- "Concat",
21
  "Equal",
22
- "Softmax",
23
- "MatMul",
24
- "Shape",
25
- "Sqrt",
26
  "Transpose",
27
- "Add",
28
  "Gather",
29
- "Range"
 
 
 
 
 
 
30
  ],
31
  "weight_type": "QInt8"
32
  },
33
  "decoder_with_past_model": {
34
  "op_types": [
35
- "ConstantOfShape",
36
- "Reshape",
 
 
37
  "Pow",
38
- "Where",
 
39
  "Cast",
 
 
40
  "Mul",
41
- "ReduceMean",
42
- "Sigmoid",
43
- "Constant",
44
  "Div",
45
- "Expand",
46
- "Sub",
47
- "Unsqueeze",
48
- "Concat",
49
  "Equal",
50
- "Softmax",
51
- "MatMul",
52
- "Shape",
53
- "Sqrt",
54
  "Transpose",
55
- "Add",
56
  "Gather",
57
- "Range"
 
 
 
 
 
 
58
  ],
59
  "weight_type": "QInt8"
60
  },
61
- "decoder_model": {
62
  "op_types": [
63
- "ConstantOfShape",
64
- "Reshape",
 
 
 
 
65
  "Pow",
66
- "Where",
 
67
  "Cast",
 
 
68
  "Mul",
69
- "ReduceMean",
70
- "Sigmoid",
71
- "Constant",
72
  "Div",
73
- "Expand",
74
- "Sub",
75
- "Squeeze",
76
- "Unsqueeze",
77
- "Concat",
78
- "Less",
79
  "Equal",
80
- "Softmax",
81
- "MatMul",
82
- "Slice",
83
- "Shape",
84
- "Sqrt",
85
  "Transpose",
86
- "Add",
87
  "Gather",
88
- "Range"
 
 
 
 
 
 
 
 
89
  ],
90
  "weight_type": "QInt8"
91
  },
92
- "decoder_model_merged": {
93
  "op_types": [
94
- "ConstantOfShape",
95
- "Reshape",
 
 
 
96
  "Pow",
97
- "Where",
 
98
  "Cast",
 
 
99
  "Mul",
100
- "ReduceMean",
101
- "Sigmoid",
102
- "Constant",
103
  "Div",
104
- "Expand",
105
- "Sub",
106
- "Squeeze",
107
- "Unsqueeze",
108
- "Concat",
109
- "If",
110
- "Less",
111
  "Equal",
112
- "Softmax",
113
- "MatMul",
114
- "Slice",
115
- "Shape",
116
- "Sqrt",
117
  "Transpose",
118
- "Add",
119
  "Gather",
120
- "Range"
 
 
 
 
 
 
 
 
121
  ],
122
  "weight_type": "QInt8"
123
  }
 
4
  "per_model_config": {
5
  "encoder_model": {
6
  "op_types": [
7
+ "Add",
8
+ "Shape",
9
+ "Range",
10
+ "MatMul",
11
  "Pow",
12
+ "Reshape",
13
+ "Expand",
14
  "Cast",
15
+ "Unsqueeze",
16
+ "Where",
17
  "Mul",
 
 
 
18
  "Div",
 
 
 
 
19
  "Equal",
 
 
 
 
20
  "Transpose",
21
+ "ConstantOfShape",
22
  "Gather",
23
+ "Constant",
24
+ "Sub",
25
+ "Sqrt",
26
+ "Sigmoid",
27
+ "ReduceMean",
28
+ "Softmax",
29
+ "Concat"
30
  ],
31
  "weight_type": "QInt8"
32
  },
33
  "decoder_with_past_model": {
34
  "op_types": [
35
+ "Add",
36
+ "Shape",
37
+ "Range",
38
+ "MatMul",
39
  "Pow",
40
+ "Reshape",
41
+ "Expand",
42
  "Cast",
43
+ "Unsqueeze",
44
+ "Where",
45
  "Mul",
 
 
 
46
  "Div",
 
 
 
 
47
  "Equal",
 
 
 
 
48
  "Transpose",
49
+ "ConstantOfShape",
50
  "Gather",
51
+ "Constant",
52
+ "Sub",
53
+ "Sqrt",
54
+ "Sigmoid",
55
+ "ReduceMean",
56
+ "Softmax",
57
+ "Concat"
58
  ],
59
  "weight_type": "QInt8"
60
  },
61
+ "decoder_model_merged": {
62
  "op_types": [
63
+ "Add",
64
+ "Shape",
65
+ "Range",
66
+ "If",
67
+ "MatMul",
68
+ "Slice",
69
  "Pow",
70
+ "Reshape",
71
+ "Expand",
72
  "Cast",
73
+ "Unsqueeze",
74
+ "Where",
75
  "Mul",
 
 
 
76
  "Div",
 
 
 
 
 
 
77
  "Equal",
 
 
 
 
 
78
  "Transpose",
79
+ "ConstantOfShape",
80
  "Gather",
81
+ "Less",
82
+ "Squeeze",
83
+ "Constant",
84
+ "Sub",
85
+ "Sqrt",
86
+ "Sigmoid",
87
+ "ReduceMean",
88
+ "Softmax",
89
+ "Concat"
90
  ],
91
  "weight_type": "QInt8"
92
  },
93
+ "decoder_model": {
94
  "op_types": [
95
+ "Add",
96
+ "Shape",
97
+ "Range",
98
+ "MatMul",
99
+ "Slice",
100
  "Pow",
101
+ "Reshape",
102
+ "Expand",
103
  "Cast",
104
+ "Unsqueeze",
105
+ "Where",
106
  "Mul",
 
 
 
107
  "Div",
 
 
 
 
 
 
 
108
  "Equal",
 
 
 
 
 
109
  "Transpose",
110
+ "ConstantOfShape",
111
  "Gather",
112
+ "Less",
113
+ "Squeeze",
114
+ "Constant",
115
+ "Sub",
116
+ "Sqrt",
117
+ "Sigmoid",
118
+ "ReduceMean",
119
+ "Softmax",
120
+ "Concat"
121
  ],
122
  "weight_type": "QInt8"
123
  }
tokenizer.json CHANGED
The diff for this file is too large to render. See raw diff