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 +76 -76
- tokenizer.json +0 -0
config.json
CHANGED
@@ -54,7 +54,7 @@
|
|
54 |
"scale_embedding": true,
|
55 |
"share_encoder_decoder_embeddings": true,
|
56 |
"static_position_embeddings": true,
|
57 |
-
"transformers_version": "4.
|
58 |
"use_cache": true,
|
59 |
"vocab_size": 65001
|
60 |
}
|
|
|
54 |
"scale_embedding": true,
|
55 |
"share_encoder_decoder_embeddings": true,
|
56 |
"static_position_embeddings": true,
|
57 |
+
"transformers_version": "4.34.0.dev0",
|
58 |
"use_cache": true,
|
59 |
"vocab_size": 65001
|
60 |
}
|
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": 6,
|
14 |
"pad_token_id": 65000,
|
15 |
-
"
|
|
|
16 |
}
|
|
|
1 |
{
|
|
|
2 |
"bad_words_ids": [
|
3 |
[
|
4 |
65000
|
|
|
11 |
"max_length": 512,
|
12 |
"num_beams": 6,
|
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:fd368f87cc14f1c12c0ea1fc550f8125cd65b8a4649b628ecc76d05140a83405
|
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:f1851afa4323c5218cc8da0585c4f4e73aa1ea675de62138016ca4173ba85aff
|
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:c6b7f04ff1ba0fbd1bf6852599b4c0cad6fe512d57cd887f44ef36cf705424cb
|
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:debcc3054b9ff9aaed972bd6e251a459abe1e2f1402da1efa63163dc2f46ef73
|
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 sha256:4f88148695d27ae34d261134d916d5a4a255141221f2df391927a901a02c2661
|
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:fbc484152fad41788c729d184058261b2a55d84eed270efa59f0ece477494e00
|
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 |
"Pow",
|
8 |
-
"
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
9 |
"Transpose",
|
10 |
-
"
|
11 |
-
"Concat",
|
12 |
-
"Range",
|
13 |
-
"Where",
|
14 |
"Softmax",
|
15 |
-
"
|
16 |
"Gather",
|
|
|
17 |
"Cast",
|
18 |
-
"
|
19 |
-
"
|
20 |
-
"
|
21 |
-
"Mul",
|
22 |
-
"Div",
|
23 |
"Sqrt",
|
24 |
-
"Add",
|
25 |
-
"Sigmoid",
|
26 |
-
"Sub",
|
27 |
-
"Unsqueeze",
|
28 |
"Expand",
|
29 |
-
"
|
30 |
],
|
31 |
"weight_type": "QInt8"
|
32 |
},
|
33 |
-
"
|
34 |
"op_types": [
|
|
|
35 |
"Pow",
|
36 |
-
"
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
37 |
"Transpose",
|
38 |
-
"
|
39 |
-
"Concat",
|
40 |
-
"Range",
|
41 |
-
"Where",
|
42 |
"Softmax",
|
43 |
-
"
|
44 |
"Gather",
|
45 |
-
"
|
46 |
"Cast",
|
47 |
-
"
|
48 |
-
"
|
49 |
-
"Mul",
|
50 |
-
"Div",
|
51 |
"Sqrt",
|
52 |
-
"
|
53 |
-
"Sigmoid",
|
54 |
-
"Sub",
|
55 |
-
"Unsqueeze",
|
56 |
-
"Expand",
|
57 |
-
"Constant"
|
58 |
],
|
59 |
"weight_type": "QInt8"
|
60 |
},
|
61 |
-
"
|
62 |
"op_types": [
|
|
|
63 |
"Pow",
|
64 |
-
"
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
65 |
"Transpose",
|
66 |
-
"
|
67 |
-
"
|
68 |
-
"Range",
|
69 |
-
"Where",
|
70 |
"Softmax",
|
71 |
-
"
|
72 |
"Gather",
|
73 |
-
"
|
74 |
"Cast",
|
75 |
-
"
|
76 |
-
"
|
77 |
-
"MatMul",
|
78 |
-
"Mul",
|
79 |
"Less",
|
80 |
-
"Div",
|
81 |
-
"Slice",
|
82 |
"Sqrt",
|
83 |
-
"Add",
|
84 |
-
"Sigmoid",
|
85 |
-
"Sub",
|
86 |
-
"Unsqueeze",
|
87 |
"Expand",
|
88 |
-
"
|
89 |
],
|
90 |
"weight_type": "QInt8"
|
91 |
},
|
92 |
-
"
|
93 |
"op_types": [
|
|
|
94 |
"Pow",
|
95 |
-
"
|
96 |
-
"
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
97 |
"Transpose",
|
98 |
-
"
|
99 |
-
"Concat",
|
100 |
-
"Range",
|
101 |
-
"Where",
|
102 |
"Softmax",
|
103 |
-
"
|
104 |
"Gather",
|
105 |
-
"
|
106 |
"Cast",
|
107 |
-
"
|
108 |
-
"
|
109 |
-
"MatMul",
|
110 |
-
"Mul",
|
111 |
-
"Less",
|
112 |
-
"Div",
|
113 |
-
"Slice",
|
114 |
"Sqrt",
|
115 |
-
"
|
116 |
-
"Sigmoid",
|
117 |
-
"Sub",
|
118 |
-
"Unsqueeze",
|
119 |
-
"Expand",
|
120 |
-
"Constant"
|
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 |
+
"Where",
|
8 |
"Pow",
|
9 |
+
"Div",
|
10 |
+
"Reshape",
|
11 |
+
"Unsqueeze",
|
12 |
+
"Equal",
|
13 |
+
"Shape",
|
14 |
+
"Constant",
|
15 |
+
"Slice",
|
16 |
+
"Sigmoid",
|
17 |
+
"MatMul",
|
18 |
+
"Mul",
|
19 |
+
"Sub",
|
20 |
"Transpose",
|
21 |
+
"Add",
|
|
|
|
|
|
|
22 |
"Softmax",
|
23 |
+
"Range",
|
24 |
"Gather",
|
25 |
+
"ReduceMean",
|
26 |
"Cast",
|
27 |
+
"ConstantOfShape",
|
28 |
+
"Concat",
|
29 |
+
"Less",
|
|
|
|
|
30 |
"Sqrt",
|
|
|
|
|
|
|
|
|
31 |
"Expand",
|
32 |
+
"Squeeze"
|
33 |
],
|
34 |
"weight_type": "QInt8"
|
35 |
},
|
36 |
+
"encoder_model": {
|
37 |
"op_types": [
|
38 |
+
"Where",
|
39 |
"Pow",
|
40 |
+
"Div",
|
41 |
+
"Reshape",
|
42 |
+
"Unsqueeze",
|
43 |
+
"Equal",
|
44 |
+
"Shape",
|
45 |
+
"Constant",
|
46 |
+
"Sigmoid",
|
47 |
+
"MatMul",
|
48 |
+
"Mul",
|
49 |
+
"Sub",
|
50 |
"Transpose",
|
51 |
+
"Add",
|
|
|
|
|
|
|
52 |
"Softmax",
|
53 |
+
"Range",
|
54 |
"Gather",
|
55 |
+
"ReduceMean",
|
56 |
"Cast",
|
57 |
+
"ConstantOfShape",
|
58 |
+
"Concat",
|
|
|
|
|
59 |
"Sqrt",
|
60 |
+
"Expand"
|
|
|
|
|
|
|
|
|
|
|
61 |
],
|
62 |
"weight_type": "QInt8"
|
63 |
},
|
64 |
+
"decoder_model_merged": {
|
65 |
"op_types": [
|
66 |
+
"Where",
|
67 |
"Pow",
|
68 |
+
"Div",
|
69 |
+
"Reshape",
|
70 |
+
"Unsqueeze",
|
71 |
+
"Equal",
|
72 |
+
"Shape",
|
73 |
+
"Constant",
|
74 |
+
"Slice",
|
75 |
+
"Sigmoid",
|
76 |
+
"MatMul",
|
77 |
+
"Mul",
|
78 |
+
"Sub",
|
79 |
"Transpose",
|
80 |
+
"Add",
|
81 |
+
"If",
|
|
|
|
|
82 |
"Softmax",
|
83 |
+
"Range",
|
84 |
"Gather",
|
85 |
+
"ReduceMean",
|
86 |
"Cast",
|
87 |
+
"ConstantOfShape",
|
88 |
+
"Concat",
|
|
|
|
|
89 |
"Less",
|
|
|
|
|
90 |
"Sqrt",
|
|
|
|
|
|
|
|
|
91 |
"Expand",
|
92 |
+
"Squeeze"
|
93 |
],
|
94 |
"weight_type": "QInt8"
|
95 |
},
|
96 |
+
"decoder_with_past_model": {
|
97 |
"op_types": [
|
98 |
+
"Where",
|
99 |
"Pow",
|
100 |
+
"Div",
|
101 |
+
"Reshape",
|
102 |
+
"Unsqueeze",
|
103 |
+
"Equal",
|
104 |
+
"Shape",
|
105 |
+
"Constant",
|
106 |
+
"Sigmoid",
|
107 |
+
"MatMul",
|
108 |
+
"Mul",
|
109 |
+
"Sub",
|
110 |
"Transpose",
|
111 |
+
"Add",
|
|
|
|
|
|
|
112 |
"Softmax",
|
113 |
+
"Range",
|
114 |
"Gather",
|
115 |
+
"ReduceMean",
|
116 |
"Cast",
|
117 |
+
"ConstantOfShape",
|
118 |
+
"Concat",
|
|
|
|
|
|
|
|
|
|
|
119 |
"Sqrt",
|
120 |
+
"Expand"
|
|
|
|
|
|
|
|
|
|
|
121 |
],
|
122 |
"weight_type": "QInt8"
|
123 |
}
|
tokenizer.json
CHANGED
The diff for this file is too large to render.
See raw diff
|
|