Xenova HF staff commited on
Commit
ec06818
1 Parent(s): f8c2cd1

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": 54796
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": 54796
60
  }
generation_config.json CHANGED
@@ -1,5 +1,4 @@
1
  {
2
- "_from_model_config": true,
3
  "bad_words_ids": [
4
  [
5
  54795
@@ -12,5 +11,6 @@
12
  "max_length": 512,
13
  "num_beams": 6,
14
  "pad_token_id": 54795,
15
- "transformers_version": "4.32.0.dev0"
 
16
  }
 
1
  {
 
2
  "bad_words_ids": [
3
  [
4
  54795
 
11
  "max_length": 512,
12
  "num_beams": 6,
13
  "pad_token_id": 54795,
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 sha256:8f8652dcb4f97783dc359b41dc68dfe1bdbc9bc4b27452d7a26172959d9d8302
3
- size 326885160
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:b4181595162594d4892841ec258b85cd332ee819cadac70e19d18f7a21249dd7
3
+ size 214663009
onnx/decoder_model_merged.onnx CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:ecfaa7ab0570414dffe8de9cede6f793d4ee0d12d1a30ed8a48073fb8fd4fbae
3
- size 327120642
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:c01f168e0ab9310956c84d7028a0c423ce7788a3649d69da979d6d8734c21287
3
+ size 214898576
onnx/decoder_model_merged_quantized.onnx CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:0285dfd6c878e25fd2b2de7c96829efc6a3799dc57f009304797b30dee444146
3
- size 83277968
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:d65eeab370c2873bdf22a7b2b311f0f636aa693b8edd55c60b6b1db8473268dd
3
+ size 54947024
onnx/decoder_model_quantized.onnx CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:022a72f9230d4f33ca5e14f498c9f6bc638fa7c54154951af1ace3f3b50ddd68
3
- size 82906612
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:e72a8b1062477950b6464b1fcfb37a99c196618e5bb67222451472ea52e98090
3
+ size 54576322
onnx/decoder_with_past_model.onnx CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:13f40ad7a07e95078a80f2830cb9be96e237b60c441d9270014ddf806dfef270
3
- size 314234651
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:52edf64e6841c37980f40beb79683ba4976f4d71de23325e7e97357ea7249f7f
3
+ size 202012500
onnx/decoder_with_past_model_quantized.onnx CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:64c41be3c50bce5d65bd4b146e1cd94a1c4072deb08fc1ba5076e732cb688d37
3
- size 79635148
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:f00cababe9d29731078f007db796826d3584788a8e171d02f82a899cfbd0a01e
3
+ size 51304858
quantize_config.json CHANGED
@@ -4,120 +4,120 @@
4
  "per_model_config": {
5
  "encoder_model": {
6
  "op_types": [
 
7
  "Equal",
8
- "Sqrt",
9
- "Constant",
10
- "Transpose",
11
- "Sigmoid",
12
- "Pow",
13
- "ConstantOfShape",
14
- "Cast",
15
- "Expand",
16
  "Add",
17
- "Softmax",
 
 
18
  "Mul",
19
- "Gather",
20
  "MatMul",
21
- "Range",
22
  "Div",
23
- "Concat",
24
  "Shape",
25
- "Reshape",
 
 
 
 
26
  "Unsqueeze",
27
- "Where",
28
- "ReduceMean",
29
- "Sub"
 
30
  ],
31
  "weight_type": "QInt8"
32
  },
33
  "decoder_with_past_model": {
34
  "op_types": [
 
35
  "Equal",
36
- "Sqrt",
37
- "Constant",
38
- "Transpose",
39
- "Sigmoid",
40
- "Pow",
41
- "ConstantOfShape",
42
- "Cast",
43
- "Expand",
44
  "Add",
45
- "Softmax",
 
46
  "Mul",
47
- "Gather",
48
  "MatMul",
49
- "Range",
50
  "Div",
51
- "Concat",
52
  "Shape",
53
- "Reshape",
 
 
 
 
54
  "Unsqueeze",
55
- "Where",
56
- "ReduceMean",
57
- "Sub"
 
58
  ],
59
  "weight_type": "QInt8"
60
  },
61
- "decoder_model": {
62
  "op_types": [
 
 
 
63
  "Equal",
64
- "Sqrt",
65
- "Constant",
66
- "Transpose",
67
- "Sigmoid",
68
- "Pow",
69
- "ConstantOfShape",
70
- "Cast",
71
- "Expand",
72
  "Add",
 
 
 
73
  "Mul",
74
- "Gather",
75
- "Softmax",
76
  "MatMul",
77
- "Range",
78
- "Squeeze",
79
  "Div",
80
- "Concat",
81
  "Shape",
82
- "Reshape",
 
 
 
 
83
  "Unsqueeze",
84
- "Slice",
85
- "Where",
86
- "ReduceMean",
87
- "Less",
88
- "Sub"
89
  ],
90
  "weight_type": "QInt8"
91
  },
92
- "decoder_model_merged": {
93
  "op_types": [
 
 
94
  "Equal",
95
- "Sqrt",
96
- "If",
97
- "Constant",
98
- "Transpose",
99
- "Sigmoid",
100
- "Pow",
101
- "ConstantOfShape",
102
- "Cast",
103
- "Expand",
104
  "Add",
 
 
 
105
  "Mul",
106
- "Gather",
107
- "Softmax",
108
  "MatMul",
109
- "Range",
110
- "Squeeze",
111
  "Div",
112
- "Concat",
113
  "Shape",
114
- "Reshape",
 
 
 
 
115
  "Unsqueeze",
116
- "Slice",
117
- "Where",
118
- "ReduceMean",
119
- "Less",
120
- "Sub"
121
  ],
122
  "weight_type": "QInt8"
123
  }
 
4
  "per_model_config": {
5
  "encoder_model": {
6
  "op_types": [
7
+ "Softmax",
8
  "Equal",
 
 
 
 
 
 
 
 
9
  "Add",
10
+ "Where",
11
+ "Pow",
12
+ "Constant",
13
  "Mul",
14
+ "Expand",
15
  "MatMul",
16
+ "Reshape",
17
  "Div",
18
+ "ReduceMean",
19
  "Shape",
20
+ "Gather",
21
+ "Sub",
22
+ "Sqrt",
23
+ "Sigmoid",
24
+ "ConstantOfShape",
25
  "Unsqueeze",
26
+ "Cast",
27
+ "Range",
28
+ "Transpose",
29
+ "Concat"
30
  ],
31
  "weight_type": "QInt8"
32
  },
33
  "decoder_with_past_model": {
34
  "op_types": [
35
+ "Softmax",
36
  "Equal",
37
+ "Where",
 
 
 
 
 
 
 
38
  "Add",
39
+ "Pow",
40
+ "Constant",
41
  "Mul",
42
+ "Expand",
43
  "MatMul",
44
+ "Reshape",
45
  "Div",
46
+ "ReduceMean",
47
  "Shape",
48
+ "Gather",
49
+ "Sub",
50
+ "Sqrt",
51
+ "Sigmoid",
52
+ "ConstantOfShape",
53
  "Unsqueeze",
54
+ "Cast",
55
+ "Range",
56
+ "Transpose",
57
+ "Concat"
58
  ],
59
  "weight_type": "QInt8"
60
  },
61
+ "decoder_model_merged": {
62
  "op_types": [
63
+ "Slice",
64
+ "Softmax",
65
+ "If",
66
  "Equal",
67
+ "Less",
 
 
 
 
 
 
 
68
  "Add",
69
+ "Where",
70
+ "Pow",
71
+ "Constant",
72
  "Mul",
73
+ "Expand",
 
74
  "MatMul",
75
+ "Reshape",
 
76
  "Div",
77
+ "ReduceMean",
78
  "Shape",
79
+ "Gather",
80
+ "Sub",
81
+ "Sqrt",
82
+ "Sigmoid",
83
+ "ConstantOfShape",
84
  "Unsqueeze",
85
+ "Cast",
86
+ "Range",
87
+ "Transpose",
88
+ "Squeeze",
89
+ "Concat"
90
  ],
91
  "weight_type": "QInt8"
92
  },
93
+ "decoder_model": {
94
  "op_types": [
95
+ "Slice",
96
+ "Softmax",
97
  "Equal",
98
+ "Less",
 
 
 
 
 
 
 
 
99
  "Add",
100
+ "Where",
101
+ "Pow",
102
+ "Constant",
103
  "Mul",
104
+ "Expand",
 
105
  "MatMul",
106
+ "Reshape",
 
107
  "Div",
108
+ "ReduceMean",
109
  "Shape",
110
+ "Gather",
111
+ "Sub",
112
+ "Sqrt",
113
+ "Sigmoid",
114
+ "ConstantOfShape",
115
  "Unsqueeze",
116
+ "Cast",
117
+ "Range",
118
+ "Transpose",
119
+ "Squeeze",
120
+ "Concat"
121
  ],
122
  "weight_type": "QInt8"
123
  }
tokenizer.json CHANGED
The diff for this file is too large to render. See raw diff