Xenova HF staff commited on
Commit
2bc4329
1 Parent(s): 930eb90

Upload folder using huggingface_hub

Browse files
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": "4.32.0.dev0",
58
  "use_cache": true,
59
  "vocab_size": 62509
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": 62509
60
  }
generation_config.json CHANGED
@@ -1,5 +1,4 @@
1
  {
2
- "_from_model_config": true,
3
  "bad_words_ids": [
4
  [
5
  62508
@@ -12,5 +11,6 @@
12
  "max_length": 512,
13
  "num_beams": 4,
14
  "pad_token_id": 62508,
15
- "transformers_version": "4.32.0.dev0"
 
16
  }
 
1
  {
 
2
  "bad_words_ids": [
3
  [
4
  62508
 
11
  "max_length": 512,
12
  "num_beams": 4,
13
  "pad_token_id": 62508,
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 sha256:2bca0470aa8d4868abf812c2ad6a4fdd3aa9b1d0c1ec183ae280eff204a9690b
3
- size 358508460
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:79d57be073450d1a68ba4ef78aedb4df00dfbf8521067b95b538484451614c08
3
+ size 230490085
onnx/decoder_model_merged.onnx CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:4c452aa2d8ee9a99bc1f966211541328cb211e7949a23fd73b084c74982e1e72
3
- size 358743942
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:16036548f1c5566fc913259a94f999b0d994663ec0bbdbf06f883d61895e7dd7
3
+ size 230725652
onnx/decoder_model_merged_quantized.onnx CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:aa0d02e9fd5aa0a9fb9791676d5b2b319461e7ef288607dd8cf6732b996bc02c
3
- size 91245497
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:811d43692ec5d65ff1dd9da69044b68cb867d225346f912309cd2be834115e66
3
+ size 58926932
onnx/decoder_model_quantized.onnx CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:ef816987acad595c3fbb505fc1808a23a8da31c9fa84cad05de4c067e55deaec
3
- size 90874141
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:caeed9603bc3802c1c74c8b4616bebc645fb82c0b21ef1c89465644f384aaa54
3
+ size 58556230
onnx/decoder_with_past_model.onnx CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:42635187056f16d198036b40afbfb5afebba6b97e2da224e3e7407dd68566770
3
- size 345857951
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:2d438f35185ace6b47c3da2067e1bc68ff8434250611ed44fd95f310be023ff8
3
+ size 217839576
onnx/decoder_with_past_model_quantized.onnx CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:3acdfb8f6c78ea6de59ed6eb9c8bbaad04da0db59d6b0e71256504b335143f62
3
- size 87602677
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:7050b3643713d6a95bafa576ee705beda32069cff8ba8a22f5fc3604b64631b6
3
+ size 55284766
quantize_config.json CHANGED
@@ -4,120 +4,120 @@
4
  "per_model_config": {
5
  "encoder_model": {
6
  "op_types": [
 
 
 
7
  "Transpose",
8
- "Reshape",
 
 
9
  "ReduceMean",
10
- "Sub",
11
- "Concat",
12
  "MatMul",
13
- "Equal",
14
- "Unsqueeze",
 
 
15
  "Sigmoid",
16
- "Mul",
17
- "Div",
18
- "ConstantOfShape",
19
  "Expand",
20
- "Range",
21
- "Gather",
22
- "Shape",
23
  "Softmax",
24
- "Sqrt",
25
- "Cast",
26
- "Pow",
27
- "Where",
28
  "Constant",
29
- "Add"
 
 
30
  ],
31
  "weight_type": "QInt8"
32
  },
33
  "decoder_with_past_model": {
34
  "op_types": [
 
 
 
35
  "Transpose",
36
- "Reshape",
 
 
37
  "ReduceMean",
38
- "Sub",
39
- "Concat",
40
  "MatMul",
41
- "Equal",
42
- "Unsqueeze",
 
 
43
  "Sigmoid",
44
- "Mul",
45
- "Div",
46
- "ConstantOfShape",
47
  "Expand",
48
- "Range",
49
- "Gather",
50
- "Shape",
51
  "Softmax",
52
- "Sqrt",
53
- "Cast",
54
- "Pow",
55
- "Where",
56
  "Constant",
57
- "Add"
 
 
58
  ],
59
  "weight_type": "QInt8"
60
  },
61
- "decoder_model": {
62
  "op_types": [
63
- "Slice",
64
- "Squeeze",
65
- "Transpose",
66
- "Reshape",
67
- "ReduceMean",
68
- "Sub",
69
- "Concat",
70
- "MatMul",
71
  "Equal",
 
 
 
72
  "Less",
73
- "Unsqueeze",
74
- "Sigmoid",
75
  "Mul",
76
- "Div",
77
- "ConstantOfShape",
78
- "Expand",
79
  "Range",
80
- "Gather",
81
- "Shape",
82
- "Softmax",
83
- "Sqrt",
84
  "Cast",
85
- "Pow",
86
  "Where",
 
 
 
 
 
 
 
 
 
87
  "Constant",
88
- "Add"
 
 
89
  ],
90
  "weight_type": "QInt8"
91
  },
92
- "decoder_model_merged": {
93
  "op_types": [
94
- "Slice",
95
- "Squeeze",
96
- "Transpose",
97
- "Reshape",
98
- "ReduceMean",
99
- "Sub",
100
- "Concat",
101
- "MatMul",
102
  "Equal",
 
 
 
103
  "Less",
104
- "Unsqueeze",
105
- "Sigmoid",
106
  "Mul",
107
- "Div",
108
- "ConstantOfShape",
109
- "Expand",
110
  "Range",
111
- "If",
112
- "Gather",
113
- "Shape",
114
- "Softmax",
115
- "Sqrt",
116
  "Cast",
117
- "Pow",
118
  "Where",
 
 
 
 
 
 
 
 
119
  "Constant",
120
- "Add"
 
 
121
  ],
122
  "weight_type": "QInt8"
123
  }
 
4
  "per_model_config": {
5
  "encoder_model": {
6
  "op_types": [
7
+ "Equal",
8
+ "Sub",
9
+ "Div",
10
  "Transpose",
11
+ "Add",
12
+ "Mul",
13
+ "Pow",
14
  "ReduceMean",
15
+ "Range",
 
16
  "MatMul",
17
+ "Cast",
18
+ "Gather",
19
+ "Where",
20
+ "Reshape",
21
  "Sigmoid",
22
+ "Concat",
23
+ "Sqrt",
 
24
  "Expand",
 
 
 
25
  "Softmax",
 
 
 
 
26
  "Constant",
27
+ "Unsqueeze",
28
+ "Shape",
29
+ "ConstantOfShape"
30
  ],
31
  "weight_type": "QInt8"
32
  },
33
  "decoder_with_past_model": {
34
  "op_types": [
35
+ "Equal",
36
+ "Sub",
37
+ "Div",
38
  "Transpose",
39
+ "Add",
40
+ "Mul",
41
+ "Pow",
42
  "ReduceMean",
43
+ "Range",
 
44
  "MatMul",
45
+ "Cast",
46
+ "Gather",
47
+ "Where",
48
+ "Reshape",
49
  "Sigmoid",
50
+ "Concat",
51
+ "Sqrt",
 
52
  "Expand",
 
 
 
53
  "Softmax",
 
 
 
 
54
  "Constant",
55
+ "Unsqueeze",
56
+ "Shape",
57
+ "ConstantOfShape"
58
  ],
59
  "weight_type": "QInt8"
60
  },
61
+ "decoder_model_merged": {
62
  "op_types": [
 
 
 
 
 
 
 
 
63
  "Equal",
64
+ "Sub",
65
+ "Div",
66
+ "Transpose",
67
  "Less",
68
+ "Add",
 
69
  "Mul",
70
+ "Pow",
71
+ "ReduceMean",
 
72
  "Range",
73
+ "MatMul",
 
 
 
74
  "Cast",
75
+ "Gather",
76
  "Where",
77
+ "Reshape",
78
+ "Sigmoid",
79
+ "Concat",
80
+ "Slice",
81
+ "If",
82
+ "Sqrt",
83
+ "Expand",
84
+ "Squeeze",
85
+ "Softmax",
86
  "Constant",
87
+ "Unsqueeze",
88
+ "Shape",
89
+ "ConstantOfShape"
90
  ],
91
  "weight_type": "QInt8"
92
  },
93
+ "decoder_model": {
94
  "op_types": [
 
 
 
 
 
 
 
 
95
  "Equal",
96
+ "Sub",
97
+ "Div",
98
+ "Transpose",
99
  "Less",
100
+ "Add",
 
101
  "Mul",
102
+ "Pow",
103
+ "ReduceMean",
 
104
  "Range",
105
+ "MatMul",
 
 
 
 
106
  "Cast",
107
+ "Gather",
108
  "Where",
109
+ "Reshape",
110
+ "Sigmoid",
111
+ "Concat",
112
+ "Slice",
113
+ "Sqrt",
114
+ "Expand",
115
+ "Squeeze",
116
+ "Softmax",
117
  "Constant",
118
+ "Unsqueeze",
119
+ "Shape",
120
+ "ConstantOfShape"
121
  ],
122
  "weight_type": "QInt8"
123
  }
tokenizer.json CHANGED
The diff for this file is too large to render. See raw diff