File size: 3,120 Bytes
8cc1bed
 
 
 
 
 
 
4b71e58
 
 
 
 
 
 
 
 
8cc1bed
4b71e58
 
8cc1bed
4b71e58
 
 
8cc1bed
 
4b71e58
 
 
8cc1bed
4b71e58
8cc1bed
 
 
 
 
 
 
4b71e58
 
 
 
 
 
 
 
 
8cc1bed
4b71e58
 
8cc1bed
4b71e58
 
 
8cc1bed
 
4b71e58
 
 
8cc1bed
4b71e58
8cc1bed
 
 
 
 
 
4b71e58
 
 
 
 
 
 
 
 
8cc1bed
4b71e58
 
8cc1bed
4b71e58
 
 
8cc1bed
 
4b71e58
 
 
8cc1bed
4b71e58
8cc1bed
 
 
 
 
4b71e58
 
 
 
 
8cc1bed
 
4b71e58
 
 
8cc1bed
4b71e58
 
 
 
 
8cc1bed
 
 
4b71e58
8cc1bed
 
 
 
 
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
{
    "per_channel": false,
    "reduce_range": false,
    "per_model_config": {
        "decoder_model": {
            "op_types": [
                "Pow",
                "Split",
                "Softmax",
                "Reshape",
                "Range",
                "Slice",
                "Mul",
                "Constant",
                "Gemm",
                "Shape",
                "Sub",
                "Concat",
                "Tanh",
                "ConstantOfShape",
                "LayerNormalization",
                "Cast",
                "Squeeze",
                "Where",
                "Div",
                "Gather",
                "Transpose",
                "MatMul",
                "Unsqueeze",
                "Add"
            ],
            "weight_type": "QInt8"
        },
        "decoder_model_merged": {
            "op_types": [
                "Pow",
                "If",
                "Split",
                "Softmax",
                "Reshape",
                "Range",
                "Slice",
                "Mul",
                "Constant",
                "Gemm",
                "Shape",
                "Sub",
                "Concat",
                "Tanh",
                "ConstantOfShape",
                "LayerNormalization",
                "Cast",
                "Squeeze",
                "Where",
                "Div",
                "Gather",
                "Transpose",
                "MatMul",
                "Unsqueeze",
                "Add"
            ],
            "weight_type": "QInt8"
        },
        "decoder_with_past_model": {
            "op_types": [
                "Pow",
                "Split",
                "Softmax",
                "Reshape",
                "Range",
                "Slice",
                "Mul",
                "Constant",
                "Gemm",
                "Shape",
                "Sub",
                "Concat",
                "Tanh",
                "ConstantOfShape",
                "LayerNormalization",
                "Cast",
                "Squeeze",
                "Where",
                "Div",
                "Gather",
                "Transpose",
                "MatMul",
                "Unsqueeze",
                "Add"
            ],
            "weight_type": "QInt8"
        },
        "encoder_model": {
            "op_types": [
                "Softmax",
                "Reshape",
                "Conv",
                "Expand",
                "Slice",
                "MatMul",
                "Mul",
                "Constant",
                "Erf",
                "Shape",
                "Concat",
                "ConstantOfShape",
                "LayerNormalization",
                "Equal",
                "Where",
                "Gather",
                "Transpose",
                "Div",
                "Unsqueeze",
                "Add"
            ],
            "weight_type": "QUInt8"
        }
    }
}