Xenova HF staff commited on
Commit
3bea4ba
1 Parent(s): c924a46

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": 58866
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": 58866
60
  }
generation_config.json CHANGED
@@ -1,5 +1,4 @@
1
  {
2
- "_from_model_config": true,
3
  "bad_words_ids": [
4
  [
5
  58865
@@ -12,5 +11,6 @@
12
  "max_length": 512,
13
  "num_beams": 4,
14
  "pad_token_id": 58865,
15
- "transformers_version": "4.32.0.dev0"
 
16
  }
 
1
  {
 
2
  "bad_words_ids": [
3
  [
4
  58865
 
11
  "max_length": 512,
12
  "num_beams": 4,
13
  "pad_token_id": 58865,
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 sha256:50039062be50cdf816cfdc60f7c9bfbaf5bfee9c84a140c14be04c1d718f99f1
3
- size 343572160
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:972300c09ea0b20f146d6f65c8e6959911e847fc57e19af33777401833a16eff
3
+ size 223014649
onnx/decoder_model_merged.onnx CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:c25c8420fe18c4aa71b3dc2cd4301be938ec92108bd6f4c4181b0bb2f652b8bd
3
- size 343807642
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:188c266e4d353303e8001b448af6bab02a28451f9b532dbeecba63c2e2411def
3
+ size 223250216
onnx/decoder_model_merged_quantized.onnx CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:35c2128e7bf2f578747f620add7fe01ae28607792554f5ffb768fe37eadba28b
3
- size 87482278
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:068f7a7db4dbef17a7d20b636fc1b5fdcaa0b5e234763ffdfc059af796bb27e6
3
+ size 57047144
onnx/decoder_model_quantized.onnx CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:89c336de269e3afbd18fe94eb186609cd1103eeb386be19bcfa686a09e2e6c90
3
- size 87110922
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:e045acb037f4337955472f7304fbff168ae5324c7dd433985eb37fa20a11228b
3
+ size 56676442
onnx/decoder_with_past_model.onnx CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:212a8e600774acb2b39012ee820fbe6ab6307212f5443806cc328483f31face3
3
- size 330921651
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:774bd236eeec6d0bbd48142fc7ef811b6827aa5209d6bf6b66593a98f48bf1ee
3
+ size 210364140
onnx/decoder_with_past_model_quantized.onnx CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:1692e3277f2aa055f5cdcd0d9dcc93f3757c795645d654f4240090e4591c6f72
3
- size 83839458
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:f2c6a1a70daf1618378caeda27deb79159f391db5d358a8f47ea5b1ddb557379
3
+ size 53404978
quantize_config.json CHANGED
@@ -4,120 +4,120 @@
4
  "per_model_config": {
5
  "encoder_model": {
6
  "op_types": [
7
- "Transpose",
8
- "Sqrt",
9
- "Constant",
10
- "Unsqueeze",
11
- "Equal",
12
- "Cast",
13
- "Reshape",
14
- "Pow",
15
  "Shape",
16
- "Softmax",
17
- "ReduceMean",
18
- "Where",
19
  "Mul",
20
  "Add",
21
- "Sub",
22
- "Concat",
23
- "MatMul",
24
- "ConstantOfShape",
25
  "Div",
26
  "Range",
27
- "Sigmoid",
 
 
 
 
 
 
 
 
 
28
  "Gather",
29
- "Expand"
 
 
 
 
 
30
  ],
31
  "weight_type": "QInt8"
32
  },
33
  "decoder_with_past_model": {
34
  "op_types": [
35
- "Transpose",
36
- "Sqrt",
37
- "Constant",
38
- "Unsqueeze",
39
- "Equal",
40
- "Cast",
41
- "Reshape",
42
- "Pow",
43
  "Shape",
44
- "Softmax",
45
- "ReduceMean",
46
- "Where",
47
  "Mul",
48
- "Sub",
49
  "Add",
50
- "Concat",
51
- "MatMul",
52
- "ConstantOfShape",
53
  "Div",
54
  "Range",
55
- "Sigmoid",
56
- "Gather",
57
- "Expand"
58
- ],
59
- "weight_type": "QInt8"
60
- },
61
- "decoder_model": {
62
- "op_types": [
63
  "Transpose",
64
- "Sqrt",
65
  "Constant",
66
- "Unsqueeze",
67
- "Equal",
68
  "Cast",
69
- "Reshape",
70
- "Pow",
71
- "Shape",
72
- "Softmax",
73
- "ReduceMean",
74
- "Squeeze",
75
- "Where",
76
- "Mul",
77
- "Add",
78
  "Sub",
 
79
  "Concat",
80
- "MatMul",
81
- "ConstantOfShape",
82
- "Div",
83
- "Less",
84
- "Range",
85
- "Sigmoid",
86
  "Gather",
87
- "Expand",
88
- "Slice"
 
 
 
 
89
  ],
90
  "weight_type": "QInt8"
91
  },
92
  "decoder_model_merged": {
93
  "op_types": [
 
 
 
 
 
 
 
 
94
  "Transpose",
95
- "Sqrt",
96
  "Constant",
97
- "Unsqueeze",
98
- "Equal",
99
  "Cast",
 
 
 
 
 
 
 
100
  "Reshape",
101
  "Pow",
102
- "Shape",
103
- "Softmax",
104
- "ReduceMean",
105
  "Squeeze",
106
- "Where",
 
 
 
 
 
 
 
 
107
  "Mul",
108
  "Add",
 
 
 
 
 
 
 
109
  "Sub",
 
110
  "Concat",
 
 
 
 
 
 
 
 
111
  "MatMul",
112
- "ConstantOfShape",
113
- "Div",
114
  "Less",
115
- "If",
116
- "Range",
117
- "Sigmoid",
118
- "Gather",
119
- "Expand",
120
- "Slice"
121
  ],
122
  "weight_type": "QInt8"
123
  }
 
4
  "per_model_config": {
5
  "encoder_model": {
6
  "op_types": [
7
+ "ConstantOfShape",
 
 
 
 
 
 
 
8
  "Shape",
 
 
 
9
  "Mul",
10
  "Add",
 
 
 
 
11
  "Div",
12
  "Range",
13
+ "Transpose",
14
+ "Softmax",
15
+ "Constant",
16
+ "Cast",
17
+ "Sub",
18
+ "Expand",
19
+ "Concat",
20
+ "ReduceMean",
21
+ "Unsqueeze",
22
+ "Sqrt",
23
  "Gather",
24
+ "Reshape",
25
+ "Pow",
26
+ "Sigmoid",
27
+ "Equal",
28
+ "MatMul",
29
+ "Where"
30
  ],
31
  "weight_type": "QInt8"
32
  },
33
  "decoder_with_past_model": {
34
  "op_types": [
35
+ "ConstantOfShape",
 
 
 
 
 
 
 
36
  "Shape",
 
 
 
37
  "Mul",
 
38
  "Add",
 
 
 
39
  "Div",
40
  "Range",
 
 
 
 
 
 
 
 
41
  "Transpose",
42
+ "Softmax",
43
  "Constant",
 
 
44
  "Cast",
 
 
 
 
 
 
 
 
 
45
  "Sub",
46
+ "Expand",
47
  "Concat",
48
+ "ReduceMean",
49
+ "Unsqueeze",
50
+ "Sqrt",
 
 
 
51
  "Gather",
52
+ "Reshape",
53
+ "Pow",
54
+ "Sigmoid",
55
+ "Equal",
56
+ "MatMul",
57
+ "Where"
58
  ],
59
  "weight_type": "QInt8"
60
  },
61
  "decoder_model_merged": {
62
  "op_types": [
63
+ "ConstantOfShape",
64
+ "Shape",
65
+ "Mul",
66
+ "Add",
67
+ "Slice",
68
+ "Div",
69
+ "If",
70
+ "Range",
71
  "Transpose",
72
+ "Softmax",
73
  "Constant",
 
 
74
  "Cast",
75
+ "Sub",
76
+ "Expand",
77
+ "Concat",
78
+ "ReduceMean",
79
+ "Unsqueeze",
80
+ "Sqrt",
81
+ "Gather",
82
  "Reshape",
83
  "Pow",
84
+ "Sigmoid",
85
+ "Equal",
86
+ "MatMul",
87
  "Squeeze",
88
+ "Less",
89
+ "Where"
90
+ ],
91
+ "weight_type": "QInt8"
92
+ },
93
+ "decoder_model": {
94
+ "op_types": [
95
+ "ConstantOfShape",
96
+ "Shape",
97
  "Mul",
98
  "Add",
99
+ "Slice",
100
+ "Div",
101
+ "Range",
102
+ "Transpose",
103
+ "Softmax",
104
+ "Constant",
105
+ "Cast",
106
  "Sub",
107
+ "Expand",
108
  "Concat",
109
+ "ReduceMean",
110
+ "Unsqueeze",
111
+ "Sqrt",
112
+ "Gather",
113
+ "Reshape",
114
+ "Pow",
115
+ "Sigmoid",
116
+ "Equal",
117
  "MatMul",
118
+ "Squeeze",
 
119
  "Less",
120
+ "Where"
 
 
 
 
 
121
  ],
122
  "weight_type": "QInt8"
123
  }
tokenizer.json CHANGED
The diff for this file is too large to render. See raw diff