Upload folder using huggingface_hub
Browse files- config.json +1 -1
- generation_config.json +2 -2
- onnx/decoder_model.onnx +2 -2
- onnx/decoder_model_merged.onnx +2 -2
- onnx/decoder_model_merged_quantized.onnx +2 -2
- onnx/decoder_model_quantized.onnx +2 -2
- onnx/decoder_with_past_model.onnx +2 -2
- onnx/decoder_with_past_model_quantized.onnx +2 -2
- quantize_config.json +72 -72
- tokenizer.json +0 -0
config.json
CHANGED
@@ -57,7 +57,7 @@
|
|
57 |
"scale_embedding": true,
|
58 |
"share_encoder_decoder_embeddings": true,
|
59 |
"static_position_embeddings": true,
|
60 |
-
"transformers_version": "4.
|
61 |
"use_cache": true,
|
62 |
"vocab_size": 65001
|
63 |
}
|
|
|
57 |
"scale_embedding": true,
|
58 |
"share_encoder_decoder_embeddings": true,
|
59 |
"static_position_embeddings": true,
|
60 |
+
"transformers_version": "4.34.0.dev0",
|
61 |
"use_cache": true,
|
62 |
"vocab_size": 65001
|
63 |
}
|
generation_config.json
CHANGED
@@ -1,5 +1,4 @@
|
|
1 |
{
|
2 |
-
"_from_model_config": true,
|
3 |
"bad_words_ids": [
|
4 |
[
|
5 |
65000
|
@@ -12,5 +11,6 @@
|
|
12 |
"max_length": 512,
|
13 |
"num_beams": 4,
|
14 |
"pad_token_id": 65000,
|
15 |
-
"
|
|
|
16 |
}
|
|
|
1 |
{
|
|
|
2 |
"bad_words_ids": [
|
3 |
[
|
4 |
65000
|
|
|
11 |
"max_length": 512,
|
12 |
"num_beams": 4,
|
13 |
"pad_token_id": 65000,
|
14 |
+
"renormalize_logits": true,
|
15 |
+
"transformers_version": "4.34.0.dev0"
|
16 |
}
|
onnx/decoder_model.onnx
CHANGED
@@ -1,3 +1,3 @@
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
-
oid sha256:
|
3 |
-
size
|
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:d06abec46352b4ba6a024bd77e6a9521bc2bd2a00943d194673c0e5b48c91509
|
3 |
+
size 235603669
|
onnx/decoder_model_merged.onnx
CHANGED
@@ -1,3 +1,3 @@
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
-
oid sha256:
|
3 |
-
size
|
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:5cb22f9ac32429b8211ae45e5940d9c05dfd66cf5847fd349ac6e391cd671e1d
|
3 |
+
size 235839236
|
onnx/decoder_model_merged_quantized.onnx
CHANGED
@@ -1,3 +1,3 @@
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
-
oid sha256:
|
3 |
-
size
|
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:023be4f841f4c47cd65fffcbaa81c0d99d7f7e0138f7ba0e03fa220a4e688aff
|
3 |
+
size 60212804
|
onnx/decoder_model_quantized.onnx
CHANGED
@@ -1,3 +1,3 @@
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
-
oid sha256:
|
3 |
-
size
|
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:2c66a3981099b40edbb3a0d65e015d05964d42fecb9780f8422776eff5939112
|
3 |
+
size 59842102
|
onnx/decoder_with_past_model.onnx
CHANGED
@@ -1,3 +1,3 @@
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
-
oid sha256:
|
3 |
-
size
|
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:73bb5f83b98ec85ffcb8897a9290b027c1bd1fc2a2d7b40e540b1c13fb59d59a
|
3 |
+
size 222953160
|
onnx/decoder_with_past_model_quantized.onnx
CHANGED
@@ -1,3 +1,3 @@
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
-
oid sha256:
|
3 |
-
size
|
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:ae0e8420f002b7202ab1d5e1fc4f8843ba65bb99362520527089171f2b3351b5
|
3 |
+
size 56570638
|
quantize_config.json
CHANGED
@@ -2,122 +2,122 @@
|
|
2 |
"per_channel": true,
|
3 |
"reduce_range": true,
|
4 |
"per_model_config": {
|
5 |
-
"
|
6 |
"op_types": [
|
|
|
|
|
|
|
|
|
7 |
"Concat",
|
8 |
"Add",
|
9 |
-
"
|
10 |
-
"Gather",
|
11 |
"Sub",
|
|
|
|
|
|
|
12 |
"Cast",
|
13 |
-
"
|
14 |
-
"Constant",
|
15 |
-
"Sqrt",
|
16 |
-
"Div",
|
17 |
-
"Shape",
|
18 |
-
"ReduceMean",
|
19 |
-
"Softmax",
|
20 |
-
"Equal",
|
21 |
-
"Sigmoid",
|
22 |
"MatMul",
|
|
|
|
|
23 |
"Expand",
|
24 |
-
"Pow",
|
25 |
-
"Range",
|
26 |
-
"Transpose",
|
27 |
"Unsqueeze",
|
28 |
-
"
|
29 |
-
"
|
|
|
|
|
|
|
|
|
|
|
|
|
30 |
],
|
31 |
"weight_type": "QInt8"
|
32 |
},
|
33 |
-
"
|
34 |
"op_types": [
|
|
|
|
|
|
|
35 |
"Concat",
|
36 |
"Add",
|
37 |
-
"
|
38 |
-
"Gather",
|
39 |
"Sub",
|
|
|
|
|
|
|
40 |
"Cast",
|
41 |
-
"
|
42 |
-
"Constant",
|
43 |
-
"Sqrt",
|
44 |
-
"Div",
|
45 |
-
"Shape",
|
46 |
"ReduceMean",
|
47 |
-
"Softmax",
|
48 |
-
"Equal",
|
49 |
-
"Sigmoid",
|
50 |
"MatMul",
|
|
|
51 |
"Expand",
|
52 |
-
"Pow",
|
53 |
-
"Range",
|
54 |
-
"Transpose",
|
55 |
"Unsqueeze",
|
56 |
-
"
|
57 |
-
"
|
|
|
|
|
|
|
|
|
58 |
],
|
59 |
"weight_type": "QInt8"
|
60 |
},
|
61 |
-
"
|
62 |
"op_types": [
|
63 |
"Slice",
|
64 |
-
"
|
|
|
|
|
65 |
"Concat",
|
66 |
-
"Less",
|
67 |
"Add",
|
68 |
-
"
|
69 |
-
"Gather",
|
70 |
"Sub",
|
|
|
|
|
|
|
71 |
"Cast",
|
72 |
-
"
|
73 |
-
"
|
74 |
-
"Sqrt",
|
75 |
-
"Div",
|
76 |
-
"Shape",
|
77 |
-
"ReduceMean",
|
78 |
-
"Softmax",
|
79 |
-
"Equal",
|
80 |
-
"Sigmoid",
|
81 |
"MatMul",
|
|
|
|
|
82 |
"Expand",
|
83 |
-
"Pow",
|
84 |
-
"Range",
|
85 |
-
"Transpose",
|
86 |
"Unsqueeze",
|
87 |
-
"
|
88 |
-
"
|
|
|
|
|
|
|
|
|
|
|
|
|
89 |
],
|
90 |
"weight_type": "QInt8"
|
91 |
},
|
92 |
-
"
|
93 |
"op_types": [
|
94 |
-
"
|
95 |
-
"
|
|
|
96 |
"Concat",
|
97 |
-
"Less",
|
98 |
"Add",
|
99 |
-
"
|
100 |
-
"Gather",
|
101 |
"Sub",
|
|
|
|
|
|
|
102 |
"Cast",
|
103 |
-
"
|
104 |
-
"Constant",
|
105 |
-
"Sqrt",
|
106 |
-
"Div",
|
107 |
-
"If",
|
108 |
-
"Shape",
|
109 |
"ReduceMean",
|
110 |
-
"Softmax",
|
111 |
-
"Equal",
|
112 |
-
"Sigmoid",
|
113 |
"MatMul",
|
|
|
114 |
"Expand",
|
115 |
-
"Pow",
|
116 |
-
"Range",
|
117 |
-
"Transpose",
|
118 |
"Unsqueeze",
|
119 |
-
"
|
120 |
-
"
|
|
|
|
|
|
|
|
|
121 |
],
|
122 |
"weight_type": "QInt8"
|
123 |
}
|
|
|
2 |
"per_channel": true,
|
3 |
"reduce_range": true,
|
4 |
"per_model_config": {
|
5 |
+
"decoder_model": {
|
6 |
"op_types": [
|
7 |
+
"Slice",
|
8 |
+
"Where",
|
9 |
+
"Sqrt",
|
10 |
+
"Pow",
|
11 |
"Concat",
|
12 |
"Add",
|
13 |
+
"Equal",
|
|
|
14 |
"Sub",
|
15 |
+
"Gather",
|
16 |
+
"Mul",
|
17 |
+
"Range",
|
18 |
"Cast",
|
19 |
+
"ConstantOfShape",
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
20 |
"MatMul",
|
21 |
+
"ReduceMean",
|
22 |
+
"Constant",
|
23 |
"Expand",
|
|
|
|
|
|
|
24 |
"Unsqueeze",
|
25 |
+
"Sigmoid",
|
26 |
+
"Shape",
|
27 |
+
"Reshape",
|
28 |
+
"Div",
|
29 |
+
"Squeeze",
|
30 |
+
"Transpose",
|
31 |
+
"Softmax",
|
32 |
+
"Less"
|
33 |
],
|
34 |
"weight_type": "QInt8"
|
35 |
},
|
36 |
+
"encoder_model": {
|
37 |
"op_types": [
|
38 |
+
"Where",
|
39 |
+
"Sqrt",
|
40 |
+
"Pow",
|
41 |
"Concat",
|
42 |
"Add",
|
43 |
+
"Equal",
|
|
|
44 |
"Sub",
|
45 |
+
"Gather",
|
46 |
+
"Mul",
|
47 |
+
"Range",
|
48 |
"Cast",
|
49 |
+
"ConstantOfShape",
|
|
|
|
|
|
|
|
|
50 |
"ReduceMean",
|
|
|
|
|
|
|
51 |
"MatMul",
|
52 |
+
"Constant",
|
53 |
"Expand",
|
|
|
|
|
|
|
54 |
"Unsqueeze",
|
55 |
+
"Sigmoid",
|
56 |
+
"Shape",
|
57 |
+
"Reshape",
|
58 |
+
"Div",
|
59 |
+
"Transpose",
|
60 |
+
"Softmax"
|
61 |
],
|
62 |
"weight_type": "QInt8"
|
63 |
},
|
64 |
+
"decoder_model_merged": {
|
65 |
"op_types": [
|
66 |
"Slice",
|
67 |
+
"Where",
|
68 |
+
"Sqrt",
|
69 |
+
"Pow",
|
70 |
"Concat",
|
|
|
71 |
"Add",
|
72 |
+
"Equal",
|
|
|
73 |
"Sub",
|
74 |
+
"Gather",
|
75 |
+
"Mul",
|
76 |
+
"Range",
|
77 |
"Cast",
|
78 |
+
"ConstantOfShape",
|
79 |
+
"If",
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
80 |
"MatMul",
|
81 |
+
"ReduceMean",
|
82 |
+
"Constant",
|
83 |
"Expand",
|
|
|
|
|
|
|
84 |
"Unsqueeze",
|
85 |
+
"Sigmoid",
|
86 |
+
"Shape",
|
87 |
+
"Reshape",
|
88 |
+
"Div",
|
89 |
+
"Squeeze",
|
90 |
+
"Transpose",
|
91 |
+
"Softmax",
|
92 |
+
"Less"
|
93 |
],
|
94 |
"weight_type": "QInt8"
|
95 |
},
|
96 |
+
"decoder_with_past_model": {
|
97 |
"op_types": [
|
98 |
+
"Where",
|
99 |
+
"Sqrt",
|
100 |
+
"Pow",
|
101 |
"Concat",
|
|
|
102 |
"Add",
|
103 |
+
"Equal",
|
|
|
104 |
"Sub",
|
105 |
+
"Gather",
|
106 |
+
"Mul",
|
107 |
+
"Range",
|
108 |
"Cast",
|
109 |
+
"ConstantOfShape",
|
|
|
|
|
|
|
|
|
|
|
110 |
"ReduceMean",
|
|
|
|
|
|
|
111 |
"MatMul",
|
112 |
+
"Constant",
|
113 |
"Expand",
|
|
|
|
|
|
|
114 |
"Unsqueeze",
|
115 |
+
"Sigmoid",
|
116 |
+
"Shape",
|
117 |
+
"Reshape",
|
118 |
+
"Div",
|
119 |
+
"Transpose",
|
120 |
+
"Softmax"
|
121 |
],
|
122 |
"weight_type": "QInt8"
|
123 |
}
|
tokenizer.json
CHANGED
The diff for this file is too large to render.
See raw diff
|
|