howdyaendra commited on
Commit
21e6dfc
1 Parent(s): 65682e2

Adding Onnx weights

Browse files
onnx/model.onnx ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:b3e15a7b913e06fd48da762eb5af2a5fe2a22f41b9051471116c188146b8d300
3
+ size 343491582
onnx/model_quantized.onnx ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:645665e61eb5fd55e373c52fbc8b0e53d4695d9b353f6631b7c4a54a6fedac23
3
+ size 87488948
quantize_config.json ADDED
@@ -0,0 +1,35 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "per_channel": true,
3
+ "reduce_range": true,
4
+ "per_model_config": {
5
+ "model": {
6
+ "op_types": [
7
+ "Slice",
8
+ "Gather",
9
+ "Equal",
10
+ "Mul",
11
+ "Constant",
12
+ "Sqrt",
13
+ "Div",
14
+ "Expand",
15
+ "ConstantOfShape",
16
+ "Shape",
17
+ "Add",
18
+ "Gemm",
19
+ "Reshape",
20
+ "Where",
21
+ "Erf",
22
+ "Concat",
23
+ "Transpose",
24
+ "ReduceMean",
25
+ "MatMul",
26
+ "Conv",
27
+ "Pow",
28
+ "Unsqueeze",
29
+ "Sub",
30
+ "Softmax"
31
+ ],
32
+ "weight_type": "QUInt8"
33
+ }
34
+ }
35
+ }