Xenova HF staff commited on
Commit
15f975b
1 Parent(s): cdaedc4

Upload folder using huggingface_hub

Browse files
config.json CHANGED
@@ -3,6 +3,7 @@
3
  "architectures": [
4
  "T5ForConditionalGeneration"
5
  ],
 
6
  "d_ff": 2048,
7
  "d_kv": 64,
8
  "d_model": 512,
@@ -53,7 +54,7 @@
53
  "prefix": "translate English to Romanian: "
54
  }
55
  },
56
- "transformers_version": "4.29.2",
57
  "use_cache": true,
58
  "vocab_size": 32128
59
  }
3
  "architectures": [
4
  "T5ForConditionalGeneration"
5
  ],
6
+ "classifier_dropout": 0.0,
7
  "d_ff": 2048,
8
  "d_kv": 64,
9
  "d_model": 512,
54
  "prefix": "translate English to Romanian: "
55
  }
56
  },
57
+ "transformers_version": "4.34.0.dev0",
58
  "use_cache": true,
59
  "vocab_size": 32128
60
  }
generation_config.json CHANGED
@@ -3,5 +3,5 @@
3
  "decoder_start_token_id": 0,
4
  "eos_token_id": 1,
5
  "pad_token_id": 0,
6
- "transformers_version": "4.29.2"
7
  }
3
  "decoder_start_token_id": 0,
4
  "eos_token_id": 1,
5
  "pad_token_id": 0,
6
+ "transformers_version": "4.34.0.dev0"
7
  }
onnx/decoder_model.onnx CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:380ec9b11e607cc87169efd74c04634724ae40cc0e4ed22964a28f235701e7ab
3
- size 232468834
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:937c37fbaf0a0e539e6019f604abc5285f4d13196aa2643674210449d8dbca66
3
+ size 166670419
onnx/decoder_model_merged.onnx CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:6ff0c5b93a304e8b08ee7f97e6bb0ed13d3a4343a3fb7c71e99b2cbbcf831068
3
- size 232631325
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:e7366705cc2418efb8211fce9aea2050845edff8c74066903d608c024e54fa6d
3
+ size 166832597
onnx/decoder_model_merged_quantized.onnx CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:851c4f09c24d06dd56423d42991c6cc1baae3d2f54ce0315200acb9af08aa8e6
3
- size 59043179
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:022ea05ec1c067113e03e4513c7a4c498044b316959de0a7c1dd1950e3308aa6
3
+ size 42431407
onnx/decoder_model_quantized.onnx CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:bae5d8a47b7d1ab4462be4779471f69ab6e72625a150eb64d206ee1eab6ed133
3
- size 58765649
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:fe424ae0e0f10b6219a2a839c894df55c64158897bfd7dda6db155179cdb6bf4
3
+ size 42154655
onnx/decoder_with_past_model.onnx CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:21793391fb0645b01924b80ece6307c1947a6ae725c82611323aa6d459a62fe3
3
- size 219874970
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:05cca57c04b30f90263eb6cc4f460b90ab33b45d54368623dcc61d0154c8d5f5
3
+ size 154076559
onnx/decoder_with_past_model_quantized.onnx CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:5f02802ba62544115d0ca9b3acb3b08564c37ecf2fa897b901c08c08d9088dc3
3
- size 55563102
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:5be205c8cd47b6a32d5e04abb36ff5aa07dd5613c87fe3191a81bc0a99c8f0ba
3
+ size 38952122
quantize_config.json ADDED
@@ -0,0 +1,139 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "per_channel": true,
3
+ "reduce_range": true,
4
+ "per_model_config": {
5
+ "decoder_model": {
6
+ "op_types": [
7
+ "Constant",
8
+ "Concat",
9
+ "Shape",
10
+ "Transpose",
11
+ "Sqrt",
12
+ "Cast",
13
+ "Gather",
14
+ "LessOrEqual",
15
+ "MatMul",
16
+ "Less",
17
+ "Tile",
18
+ "Div",
19
+ "Unsqueeze",
20
+ "Reshape",
21
+ "Sub",
22
+ "Log",
23
+ "Min",
24
+ "Neg",
25
+ "Where",
26
+ "Add",
27
+ "Relu",
28
+ "Softmax",
29
+ "Range",
30
+ "Expand",
31
+ "ReduceMean",
32
+ "Pow",
33
+ "ConstantOfShape",
34
+ "Mul"
35
+ ],
36
+ "weight_type": "QInt8"
37
+ },
38
+ "encoder_model": {
39
+ "op_types": [
40
+ "Constant",
41
+ "Concat",
42
+ "Shape",
43
+ "Transpose",
44
+ "Sqrt",
45
+ "Cast",
46
+ "Gather",
47
+ "MatMul",
48
+ "Less",
49
+ "Div",
50
+ "Unsqueeze",
51
+ "Reshape",
52
+ "Sub",
53
+ "Abs",
54
+ "Log",
55
+ "Min",
56
+ "Where",
57
+ "Add",
58
+ "Greater",
59
+ "Relu",
60
+ "Softmax",
61
+ "Range",
62
+ "ReduceMean",
63
+ "Pow",
64
+ "ConstantOfShape",
65
+ "Mul"
66
+ ],
67
+ "weight_type": "QInt8"
68
+ },
69
+ "decoder_model_merged": {
70
+ "op_types": [
71
+ "Constant",
72
+ "Concat",
73
+ "If",
74
+ "Shape",
75
+ "Transpose",
76
+ "Sqrt",
77
+ "Cast",
78
+ "Gather",
79
+ "LessOrEqual",
80
+ "MatMul",
81
+ "Less",
82
+ "Tile",
83
+ "Slice",
84
+ "Div",
85
+ "Unsqueeze",
86
+ "Reshape",
87
+ "Sub",
88
+ "Log",
89
+ "Min",
90
+ "Neg",
91
+ "Where",
92
+ "Add",
93
+ "Relu",
94
+ "Softmax",
95
+ "Range",
96
+ "Expand",
97
+ "ReduceMean",
98
+ "Pow",
99
+ "ConstantOfShape",
100
+ "Mul"
101
+ ],
102
+ "weight_type": "QInt8"
103
+ },
104
+ "decoder_with_past_model": {
105
+ "op_types": [
106
+ "Constant",
107
+ "Concat",
108
+ "Shape",
109
+ "Transpose",
110
+ "Sqrt",
111
+ "Cast",
112
+ "Gather",
113
+ "LessOrEqual",
114
+ "MatMul",
115
+ "Less",
116
+ "Tile",
117
+ "Slice",
118
+ "Div",
119
+ "Unsqueeze",
120
+ "Reshape",
121
+ "Sub",
122
+ "Log",
123
+ "Min",
124
+ "Neg",
125
+ "Where",
126
+ "Add",
127
+ "Relu",
128
+ "Softmax",
129
+ "Range",
130
+ "Expand",
131
+ "ReduceMean",
132
+ "Pow",
133
+ "ConstantOfShape",
134
+ "Mul"
135
+ ],
136
+ "weight_type": "QInt8"
137
+ }
138
+ }
139
+ }