xnohat commited on
Commit
166257d
1 Parent(s): da9d016

Upload folder using huggingface_hub

Browse files
onnx/decoder_model_merged.onnx CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:ebc84fd86752c37188c4ec26f231fb5a2e15be792389f135a9e59ce77c4a7b1e
3
  size 166832597
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:6db6d0d075544d2f0a6ed1de25aaca1db02b1d9cdb72382beaf0728959776fbd
3
  size 166832597
onnx/decoder_model_merged_quantized.onnx ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:425e1f4372b42e3989b9326e476d675bc522a0a6b2e38957c11aa79de0d533ce
3
+ size 42431407
onnx/decoder_model_quantized.onnx ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:43edad0c6c4b80f2f751915475ad24ab1a3942e412d343691e728e232635e7e2
3
+ size 42154656
onnx/decoder_with_past_model_quantized.onnx ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:27c15fe69162887323fba33d2663464bf41948ac947b4da16e2a134898b0bcce
3
+ size 38952123
onnx/encoder_model_quantized.onnx ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:878b39f5e367cf4c71feebaadd0b7c43fa2f97594fb297bc55a692d917427d59
3
+ size 35640020
quantize_config.json ADDED
@@ -0,0 +1,139 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "per_channel": true,
3
+ "reduce_range": true,
4
+ "per_model_config": {
5
+ "encoder_model": {
6
+ "op_types": [
7
+ "Shape",
8
+ "Reshape",
9
+ "Abs",
10
+ "Sqrt",
11
+ "Transpose",
12
+ "Gather",
13
+ "Pow",
14
+ "Softmax",
15
+ "Unsqueeze",
16
+ "Relu",
17
+ "Log",
18
+ "Less",
19
+ "Div",
20
+ "Range",
21
+ "ConstantOfShape",
22
+ "Greater",
23
+ "Concat",
24
+ "MatMul",
25
+ "Sub",
26
+ "Mul",
27
+ "Min",
28
+ "Add",
29
+ "Cast",
30
+ "Where",
31
+ "Constant",
32
+ "ReduceMean"
33
+ ],
34
+ "weight_type": "QInt8"
35
+ },
36
+ "decoder_with_past_model": {
37
+ "op_types": [
38
+ "Shape",
39
+ "Reshape",
40
+ "Sqrt",
41
+ "Transpose",
42
+ "Gather",
43
+ "Pow",
44
+ "Softmax",
45
+ "Unsqueeze",
46
+ "Relu",
47
+ "Slice",
48
+ "Log",
49
+ "Less",
50
+ "Div",
51
+ "Range",
52
+ "ConstantOfShape",
53
+ "Expand",
54
+ "Neg",
55
+ "Concat",
56
+ "MatMul",
57
+ "Sub",
58
+ "Mul",
59
+ "Min",
60
+ "LessOrEqual",
61
+ "Tile",
62
+ "Add",
63
+ "Cast",
64
+ "Where",
65
+ "Constant",
66
+ "ReduceMean"
67
+ ],
68
+ "weight_type": "QInt8"
69
+ },
70
+ "decoder_model": {
71
+ "op_types": [
72
+ "Shape",
73
+ "Reshape",
74
+ "Sqrt",
75
+ "Transpose",
76
+ "Gather",
77
+ "Pow",
78
+ "Softmax",
79
+ "Unsqueeze",
80
+ "Relu",
81
+ "Log",
82
+ "Less",
83
+ "Div",
84
+ "Range",
85
+ "ConstantOfShape",
86
+ "Expand",
87
+ "Neg",
88
+ "Concat",
89
+ "MatMul",
90
+ "Sub",
91
+ "Mul",
92
+ "Min",
93
+ "LessOrEqual",
94
+ "Tile",
95
+ "Add",
96
+ "Cast",
97
+ "Where",
98
+ "Constant",
99
+ "ReduceMean"
100
+ ],
101
+ "weight_type": "QInt8"
102
+ },
103
+ "decoder_model_merged": {
104
+ "op_types": [
105
+ "Shape",
106
+ "Reshape",
107
+ "Sqrt",
108
+ "Transpose",
109
+ "Gather",
110
+ "Pow",
111
+ "Softmax",
112
+ "Unsqueeze",
113
+ "Relu",
114
+ "Slice",
115
+ "Log",
116
+ "Less",
117
+ "Div",
118
+ "Range",
119
+ "ConstantOfShape",
120
+ "Expand",
121
+ "Neg",
122
+ "Concat",
123
+ "MatMul",
124
+ "Sub",
125
+ "Mul",
126
+ "Min",
127
+ "LessOrEqual",
128
+ "Tile",
129
+ "Add",
130
+ "Cast",
131
+ "Where",
132
+ "Constant",
133
+ "If",
134
+ "ReduceMean"
135
+ ],
136
+ "weight_type": "QInt8"
137
+ }
138
+ }
139
+ }