Upload ONNX weights

#2
by Xenova HF staff - opened
onnx/model.onnx ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:b42c3342c611b7248b2aa1b3ccf8031928067de99c886b57f177ea42c369d22b
3
+ size 1300477706
onnx/model_quantized.onnx ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:5595c39ea1e0f29a39ddf6cf977c24b2f80f21acd2b184c26212ada188a6b6d0
3
+ size 327892092
onnx/quantize_config.json ADDED
@@ -0,0 +1,30 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "per_channel": true,
3
+ "reduce_range": true,
4
+ "per_model_config": {
5
+ "model": {
6
+ "op_types": [
7
+ "Erf",
8
+ "Transpose",
9
+ "Cast",
10
+ "Softmax",
11
+ "Gather",
12
+ "Mul",
13
+ "ReduceMean",
14
+ "Unsqueeze",
15
+ "Reshape",
16
+ "Sqrt",
17
+ "Slice",
18
+ "Constant",
19
+ "MatMul",
20
+ "Concat",
21
+ "Div",
22
+ "Add",
23
+ "Pow",
24
+ "Sub",
25
+ "Shape"
26
+ ],
27
+ "weight_type": "QInt8"
28
+ }
29
+ }
30
+ }