Upload folder using huggingface_hub
Browse files- onnx/model.onnx +3 -0
- onnx/model_bnb4.onnx +3 -0
- onnx/model_fp16.onnx +3 -0
- onnx/model_int8.onnx +3 -0
- onnx/model_q4.onnx +3 -0
- onnx/model_quantized.onnx +3 -0
- onnx/model_uint8.onnx +3 -0
- quantize_config.json +90 -0
onnx/model.onnx
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:e44631a1a62f85c35f8f147f086a697e556c881079d29f786426bdd8934e3c90
|
3 |
+
size 30291205
|
onnx/model_bnb4.onnx
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:e9063679d074d7516e97d537a2f13d8b316281a5fc617566e67aaf566993add8
|
3 |
+
size 30291224
|
onnx/model_fp16.onnx
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:4d489e89eb5250b7db458f835ee0eb27016e0af05e16afbdda7cc6fef36a8fcd
|
3 |
+
size 15229540
|
onnx/model_int8.onnx
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:1d16cba523e28b6fa1af7730c434788080cdcefe3193851d13aca27fc4a54572
|
3 |
+
size 7990828
|
onnx/model_q4.onnx
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:e9063679d074d7516e97d537a2f13d8b316281a5fc617566e67aaf566993add8
|
3 |
+
size 30291224
|
onnx/model_quantized.onnx
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:e37777e83a9ef29901d91e595eb909c026fd2dbbf9a4bd2899a8fd7ffb9f5f93
|
3 |
+
size 7990872
|
onnx/model_uint8.onnx
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:e37777e83a9ef29901d91e595eb909c026fd2dbbf9a4bd2899a8fd7ffb9f5f93
|
3 |
+
size 7990872
|
quantize_config.json
ADDED
@@ -0,0 +1,90 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"fp16": {},
|
3 |
+
"q8": {
|
4 |
+
"per_model_config": {
|
5 |
+
"model": {
|
6 |
+
"op_types": [
|
7 |
+
"Add",
|
8 |
+
"BatchNormalization",
|
9 |
+
"Constant",
|
10 |
+
"Conv",
|
11 |
+
"Div",
|
12 |
+
"Erf",
|
13 |
+
"Flatten",
|
14 |
+
"Gemm",
|
15 |
+
"GlobalAveragePool",
|
16 |
+
"Identity",
|
17 |
+
"Mul",
|
18 |
+
"ReduceMean",
|
19 |
+
"Relu",
|
20 |
+
"Sigmoid",
|
21 |
+
"Sub"
|
22 |
+
],
|
23 |
+
"weight_type": "QUInt8"
|
24 |
+
}
|
25 |
+
},
|
26 |
+
"per_channel": false,
|
27 |
+
"reduce_range": false
|
28 |
+
},
|
29 |
+
"int8": {
|
30 |
+
"per_model_config": {
|
31 |
+
"model": {
|
32 |
+
"op_types": [
|
33 |
+
"Add",
|
34 |
+
"BatchNormalization",
|
35 |
+
"Constant",
|
36 |
+
"Conv",
|
37 |
+
"Div",
|
38 |
+
"Erf",
|
39 |
+
"Flatten",
|
40 |
+
"Gemm",
|
41 |
+
"GlobalAveragePool",
|
42 |
+
"Identity",
|
43 |
+
"Mul",
|
44 |
+
"ReduceMean",
|
45 |
+
"Relu",
|
46 |
+
"Sigmoid",
|
47 |
+
"Sub"
|
48 |
+
],
|
49 |
+
"weight_type": "QInt8"
|
50 |
+
}
|
51 |
+
},
|
52 |
+
"per_channel": false,
|
53 |
+
"reduce_range": false
|
54 |
+
},
|
55 |
+
"uint8": {
|
56 |
+
"per_model_config": {
|
57 |
+
"model": {
|
58 |
+
"op_types": [
|
59 |
+
"Add",
|
60 |
+
"BatchNormalization",
|
61 |
+
"Constant",
|
62 |
+
"Conv",
|
63 |
+
"Div",
|
64 |
+
"Erf",
|
65 |
+
"Flatten",
|
66 |
+
"Gemm",
|
67 |
+
"GlobalAveragePool",
|
68 |
+
"Identity",
|
69 |
+
"Mul",
|
70 |
+
"ReduceMean",
|
71 |
+
"Relu",
|
72 |
+
"Sigmoid",
|
73 |
+
"Sub"
|
74 |
+
],
|
75 |
+
"weight_type": "QUInt8"
|
76 |
+
}
|
77 |
+
},
|
78 |
+
"per_channel": false,
|
79 |
+
"reduce_range": false
|
80 |
+
},
|
81 |
+
"q4": {
|
82 |
+
"block_size": 32,
|
83 |
+
"is_symmetric": true,
|
84 |
+
"accuracy_level": null
|
85 |
+
},
|
86 |
+
"bnb4": {
|
87 |
+
"block_size": 64,
|
88 |
+
"quant_type": 1
|
89 |
+
}
|
90 |
+
}
|