Hugging Face's logo

Xenova
/
distilbart-cnn-6-6

Summarization
Transformers.js
ONNX
bart
text2text-generation
Model card Files Community
2
Xenova HF staff commited on
Commit
1fd4a5e
·
1 Parent(s): 1ad05db

Upload folder using huggingface_hub

Browse files
Files changed (6) hide show
  1. onnx/decoder_model_merged.onnx +1 -1
  2. onnx/decoder_model_merged_quantized.onnx +2 -2
  3. onnx/decoder_model_quantized.onnx +2 -2
  4. onnx/decoder_with_past_model_quantized.onnx +2 -2
  5. onnx/encoder_model_quantized.onnx +2 -2
  6. quantize_config.json +71 -71
onnx/decoder_model_merged.onnx CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:2c15a6f30dad41c3265f6786dbd850b6c563d9e6cca0197a1517822ccf2cb44b
3
  size 819827967
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:697a94ad777226e8b7eb83ab3ffa6c23614e156063f96fc530a61b9ad1d42909
3
  size 819827967
onnx/decoder_model_merged_quantized.onnx CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:06a8b2aad1811a49576bb1fae43142bd91fc90204065a363191268ee84a7cd28
3
- size 206825235
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:9f13b225f3ac82972f03a43ab7419c489612acf9b66c47a97d0cd8390641c721
3
+ size 206174232
onnx/decoder_model_quantized.onnx CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:b704da825ee605a2576a88093fed2442978b00309d3a576337ae2fe1ac843fcf
3
- size 206437988
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:607ea7b1b4cdea8c574f4778fddf8399bb953c2f4dd52afb88188d0757a0a98c
3
+ size 205786985
onnx/decoder_with_past_model_quantized.onnx CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:8615325c7835f7734c6c6ed53264dc85f9954db1f176697cf033fc596fc7f52c
3
- size 193674668
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:7d68b0a160502ea0e6d5af0144fbaed6ebb78cdb1bbae4beeb4e37e4131e8810
3
+ size 193085165
onnx/encoder_model_quantized.onnx CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:14d9b2d8fb1b6ebf7133046eaba3ea0dc0539be0523caa14c3e9b22131518a02
3
- size 128819737
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:ffb6c0547b82b923abff346f03b2f0d75489bd71bdacc2fb773a9dc5d94b7d68
3
+ size 128543065
quantize_config.json CHANGED
@@ -1,123 +1,123 @@
1
  {
2
- "per_channel": true,
3
- "reduce_range": true,
4
  "per_model_config": {
5
  "decoder_model": {
6
  "op_types": [
7
  "Cast",
8
- "Transpose",
 
 
9
  "Slice",
10
- "Concat",
11
- "Unsqueeze",
12
- "Where",
 
13
  "Pow",
14
- "Expand",
15
  "MatMul",
16
- "Sub",
17
- "Div",
18
  "Equal",
19
- "Mul",
20
- "Gather",
21
- "Constant",
22
  "Softmax",
23
- "Shape",
24
- "Squeeze",
25
- "ConstantOfShape",
26
- "Range",
27
- "Less",
28
- "Add",
29
  "Sqrt",
30
- "Reshape",
 
 
 
 
31
  "ReduceMean",
32
- "Erf"
 
 
 
33
  ],
34
  "weight_type": "QInt8"
35
  },
36
  "decoder_model_merged": {
37
  "op_types": [
38
  "Cast",
39
- "If",
40
- "Transpose",
 
41
  "Slice",
42
- "Concat",
43
- "Unsqueeze",
44
- "Where",
 
45
  "Pow",
46
- "Expand",
47
  "MatMul",
48
- "Sub",
49
- "Div",
50
  "Equal",
51
- "Mul",
52
- "Gather",
53
- "Constant",
54
  "Softmax",
55
- "Shape",
56
- "Squeeze",
57
- "ConstantOfShape",
58
- "Range",
59
- "Less",
60
- "Add",
61
  "Sqrt",
62
- "Reshape",
 
 
 
 
63
  "ReduceMean",
64
- "Erf"
 
 
 
65
  ],
66
  "weight_type": "QInt8"
67
  },
68
  "decoder_with_past_model": {
69
  "op_types": [
 
70
  "Cast",
71
- "Transpose",
72
- "Concat",
73
- "Unsqueeze",
74
- "Where",
75
- "Pow",
76
- "Expand",
77
- "MatMul",
78
  "Sub",
 
 
79
  "Div",
 
 
 
80
  "Equal",
81
- "Mul",
82
- "Gather",
83
- "Constant",
84
  "Softmax",
85
- "Shape",
86
- "ConstantOfShape",
87
- "Range",
88
- "Add",
89
  "Sqrt",
90
- "Reshape",
 
 
 
 
91
  "ReduceMean",
92
- "Erf"
 
 
93
  ],
94
  "weight_type": "QInt8"
95
  },
96
  "encoder_model": {
97
  "op_types": [
 
98
  "Cast",
99
- "Transpose",
100
- "Concat",
101
- "Unsqueeze",
102
- "Where",
103
- "Pow",
104
- "Expand",
105
- "MatMul",
106
  "Sub",
 
 
107
  "Div",
 
 
 
108
  "Equal",
109
- "Mul",
110
- "Gather",
111
- "Constant",
112
  "Softmax",
113
- "Shape",
114
- "ConstantOfShape",
115
- "Range",
116
- "Add",
117
  "Sqrt",
118
- "Reshape",
 
 
 
 
119
  "ReduceMean",
120
- "Erf"
 
 
121
  ],
122
  "weight_type": "QInt8"
123
  }
 
1
  {
2
+ "per_channel": false,
3
+ "reduce_range": false,
4
  "per_model_config": {
5
  "decoder_model": {
6
  "op_types": [
7
  "Cast",
8
+ "Reshape",
9
+ "Gather",
10
+ "Sub",
11
  "Slice",
12
+ "Erf",
13
+ "Less",
14
+ "Mul",
15
+ "Div",
16
  "Pow",
 
17
  "MatMul",
18
+ "Add",
 
19
  "Equal",
 
 
 
20
  "Softmax",
21
+ "Transpose",
 
 
 
 
 
22
  "Sqrt",
23
+ "ConstantOfShape",
24
+ "Concat",
25
+ "Unsqueeze",
26
+ "Expand",
27
+ "Constant",
28
  "ReduceMean",
29
+ "Range",
30
+ "Squeeze",
31
+ "Where",
32
+ "Shape"
33
  ],
34
  "weight_type": "QInt8"
35
  },
36
  "decoder_model_merged": {
37
  "op_types": [
38
  "Cast",
39
+ "Reshape",
40
+ "Gather",
41
+ "Sub",
42
  "Slice",
43
+ "Erf",
44
+ "Less",
45
+ "Mul",
46
+ "Div",
47
  "Pow",
 
48
  "MatMul",
49
+ "Add",
50
+ "If",
51
  "Equal",
 
 
 
52
  "Softmax",
53
+ "Transpose",
 
 
 
 
 
54
  "Sqrt",
55
+ "ConstantOfShape",
56
+ "Concat",
57
+ "Unsqueeze",
58
+ "Expand",
59
+ "Constant",
60
  "ReduceMean",
61
+ "Range",
62
+ "Squeeze",
63
+ "Where",
64
+ "Shape"
65
  ],
66
  "weight_type": "QInt8"
67
  },
68
  "decoder_with_past_model": {
69
  "op_types": [
70
+ "Reshape",
71
  "Cast",
72
+ "Gather",
 
 
 
 
 
 
73
  "Sub",
74
+ "Erf",
75
+ "Mul",
76
  "Div",
77
+ "Pow",
78
+ "MatMul",
79
+ "Add",
80
  "Equal",
 
 
 
81
  "Softmax",
82
+ "Transpose",
 
 
 
83
  "Sqrt",
84
+ "ConstantOfShape",
85
+ "Concat",
86
+ "Unsqueeze",
87
+ "Expand",
88
+ "Constant",
89
  "ReduceMean",
90
+ "Range",
91
+ "Where",
92
+ "Shape"
93
  ],
94
  "weight_type": "QInt8"
95
  },
96
  "encoder_model": {
97
  "op_types": [
98
+ "Reshape",
99
  "Cast",
100
+ "Gather",
 
 
 
 
 
 
101
  "Sub",
102
+ "Erf",
103
+ "Mul",
104
  "Div",
105
+ "Pow",
106
+ "MatMul",
107
+ "Add",
108
  "Equal",
 
 
 
109
  "Softmax",
110
+ "Transpose",
 
 
 
111
  "Sqrt",
112
+ "ConstantOfShape",
113
+ "Concat",
114
+ "Unsqueeze",
115
+ "Expand",
116
+ "Constant",
117
  "ReduceMean",
118
+ "Range",
119
+ "Where",
120
+ "Shape"
121
  ],
122
  "weight_type": "QInt8"
123
  }