Xenova HF staff commited on
Commit
86ac47c
1 Parent(s): ff043a0

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:e918f5de08de4cb6fcae6db006f9618bf55015a517903bae5b9576ac71a2fd6b
3
  size 1399393
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:7b5440b416feb4f24695692b176225f518deafed3c017cea8ba6d5fe0151cf81
3
  size 1399393
onnx/decoder_model_merged_quantized.onnx CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:46bbf11c9e9fb14f36e2e9428b3eead8e90342f6554369d8aa1ff34606635e03
3
- size 821504874
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:a93ddbe77bb8359c687418bf7592b46f5bd3e5ea2fe8dbf62ea682c73b8cc3ea
3
+ size 819144015
onnx/decoder_model_quantized.onnx CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:a83014d55c5fa9fc9264040513b9b7bd5eff1d9d4da78d66acfae8410c222efa
3
- size 820477030
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:e3f054fdc6afd82cbbf5f23b6f32770fff798d7620e2457c838fb5b5a0facec3
3
+ size 818116171
onnx/decoder_with_past_model_quantized.onnx CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:f981dcad0360bd9da6f48d95f457fd0a9590c29d4ef9fb9114b95e42f125328f
3
- size 820479158
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:7ffe8bf5750be07cc010c1d018e16e45d13fe8f2ecdde7ae904693fe6a31055a
3
+ size 818118299
quantize_config.json ADDED
@@ -0,0 +1,104 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "per_channel": false,
3
+ "reduce_range": false,
4
+ "per_model_config": {
5
+ "decoder_model": {
6
+ "op_types": [
7
+ "ScatterND",
8
+ "ReduceMean",
9
+ "Transpose",
10
+ "Expand",
11
+ "CumSum",
12
+ "Softmax",
13
+ "Gather",
14
+ "Reshape",
15
+ "Tanh",
16
+ "Unsqueeze",
17
+ "Cast",
18
+ "Pow",
19
+ "Equal",
20
+ "Concat",
21
+ "Constant",
22
+ "Slice",
23
+ "Sub",
24
+ "Range",
25
+ "Sqrt",
26
+ "Not",
27
+ "Mul",
28
+ "Div",
29
+ "Less",
30
+ "Or",
31
+ "MatMul",
32
+ "Shape",
33
+ "Add",
34
+ "Where",
35
+ "ConstantOfShape"
36
+ ],
37
+ "weight_type": "QInt8"
38
+ },
39
+ "decoder_model_merged": {
40
+ "op_types": [
41
+ "ScatterND",
42
+ "ReduceMean",
43
+ "Transpose",
44
+ "Expand",
45
+ "CumSum",
46
+ "Softmax",
47
+ "Gather",
48
+ "Reshape",
49
+ "Tanh",
50
+ "Unsqueeze",
51
+ "Cast",
52
+ "Pow",
53
+ "Equal",
54
+ "Concat",
55
+ "Constant",
56
+ "Slice",
57
+ "Sub",
58
+ "Range",
59
+ "Sqrt",
60
+ "Not",
61
+ "Mul",
62
+ "If",
63
+ "Div",
64
+ "Less",
65
+ "Or",
66
+ "MatMul",
67
+ "Shape",
68
+ "Add",
69
+ "Where",
70
+ "ConstantOfShape"
71
+ ],
72
+ "weight_type": "QInt8"
73
+ },
74
+ "decoder_with_past_model": {
75
+ "op_types": [
76
+ "ReduceMean",
77
+ "Transpose",
78
+ "Expand",
79
+ "CumSum",
80
+ "Softmax",
81
+ "Gather",
82
+ "Reshape",
83
+ "Tanh",
84
+ "Unsqueeze",
85
+ "Cast",
86
+ "Pow",
87
+ "Equal",
88
+ "Concat",
89
+ "Constant",
90
+ "Sub",
91
+ "Sqrt",
92
+ "Not",
93
+ "Mul",
94
+ "Div",
95
+ "MatMul",
96
+ "Shape",
97
+ "Add",
98
+ "Where",
99
+ "ConstantOfShape"
100
+ ],
101
+ "weight_type": "QInt8"
102
+ }
103
+ }
104
+ }