Upload ONNX weights

#4
by Xenova HF staff - opened
onnx/model.onnx ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:4e0cbababab52b0f2b49fcc66fdc13505eb73f87d0d8bf8691954152ab54b74c
3
+ size 1298376458
onnx/model_quantized.onnx ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:5e91bb6e537ec96aa9cd621e2d5295cfdf91f935f179bdeb3a343c5bb159b963
3
+ size 327363708
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
+ "Reshape",
8
+ "Sub",
9
+ "Cast",
10
+ "Gather",
11
+ "Unsqueeze",
12
+ "Sqrt",
13
+ "Pow",
14
+ "MatMul",
15
+ "Erf",
16
+ "Div",
17
+ "Constant",
18
+ "Add",
19
+ "Shape",
20
+ "Slice",
21
+ "Concat",
22
+ "ReduceMean",
23
+ "Transpose",
24
+ "Softmax",
25
+ "Mul"
26
+ ],
27
+ "weight_type": "QInt8"
28
+ }
29
+ }
30
+ }