Xenova HF staff commited on
Commit
a0e7036
·
1 Parent(s): d15c81a

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": 59471
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": 59471
60
  }
generation_config.json CHANGED
@@ -1,5 +1,4 @@
1
  {
2
- "_from_model_config": true,
3
  "bad_words_ids": [
4
  [
5
  59470
@@ -12,5 +11,6 @@
12
  "max_length": 512,
13
  "num_beams": 4,
14
  "pad_token_id": 59470,
15
- "transformers_version": "4.32.0.dev0"
 
16
  }
 
1
  {
 
2
  "bad_words_ids": [
3
  [
4
  59470
 
11
  "max_length": 512,
12
  "num_beams": 4,
13
  "pad_token_id": 59470,
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:fdfe119e8c1bf8c3cceab6266c20be9f4902ff99b93ab64c702fa68891c09b85
3
- size 346052660
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:7f47bdbd11a70f07799ec378c117f0f164e6b1b05b730d359c674ffe198fb798
3
+ size 224256109
onnx/decoder_model_merged.onnx CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:9f518161bccd0e4fdd4026fb8c0798ae2311e5538e6d44cb33b35c853674b77c
3
- size 346288142
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:e6ef394fa3c8f69a672e402de0f0b457e97f1562e40523c31f040556e731cf7c
3
+ size 224491676
onnx/decoder_model_merged_quantized.onnx CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:b8724d1ce7facad32a4aab3f9bf9df09cfdc1d1f230af3629f611d0ce32d04ea
3
- size 88107243
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:814051ef45250d1bcc1a961e4aa4f9d87bf825956180bfcaff7bbee36134c40a
3
+ size 57359324
onnx/decoder_model_quantized.onnx CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:28a0ab81b6ba98c744a00b73295e45c3ceff64d0f311eb2b2a9edf19152696fe
3
- size 87735887
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:9e7e4384a059a6db1654a50a54dab939481b77166c79344efa83d41d560f067c
3
+ size 56988622
onnx/decoder_with_past_model.onnx CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:7cbbc3bc23d6b8fc48e84cb4423b71bdf4ebaef6c9d91ef93b335dd30cc7fed0
3
- size 333402151
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:ddc88fe3b1d6e4f4772e16360e3ba4221b6f21db1d4c257bb15191375dce5e08
3
+ size 211605600
onnx/decoder_with_past_model_quantized.onnx CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:1926c2fe68fa4cfedd5ff21e6e6bbd2e8e95ebcce1def7530e0d2174982987b5
3
- size 84464423
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:9352f0bf6fc30fc3280f5c4fe8ecc89274e1445351e34b0713751aaafa4f23ee
3
+ size 53717158
quantize_config.json CHANGED
@@ -4,120 +4,120 @@
4
  "per_model_config": {
5
  "encoder_model": {
6
  "op_types": [
7
- "ConstantOfShape",
8
- "Constant",
9
- "Expand",
10
- "Transpose",
11
- "Cast",
12
- "Add",
13
- "Unsqueeze",
14
- "Div",
15
  "Sub",
16
- "Pow",
 
17
  "MatMul",
18
- "Mul",
19
- "Sigmoid",
 
 
20
  "ReduceMean",
21
- "Range",
22
  "Shape",
23
- "Equal",
24
- "Where",
 
25
  "Sqrt",
26
  "Concat",
27
- "Gather",
28
- "Softmax",
29
- "Reshape"
 
 
 
 
 
30
  ],
31
  "weight_type": "QInt8"
32
  },
33
  "decoder_with_past_model": {
34
  "op_types": [
35
- "ConstantOfShape",
36
- "Constant",
37
- "Expand",
38
- "Transpose",
39
- "Cast",
40
- "Add",
41
- "Unsqueeze",
42
- "Div",
43
  "Sub",
44
- "Pow",
 
45
  "MatMul",
46
- "Mul",
47
- "Sigmoid",
 
 
48
  "ReduceMean",
49
- "Range",
50
  "Shape",
51
- "Equal",
52
- "Where",
 
53
  "Sqrt",
54
  "Concat",
55
- "Gather",
56
- "Softmax",
57
- "Reshape"
 
 
 
 
 
58
  ],
59
  "weight_type": "QInt8"
60
  },
61
- "decoder_model": {
62
  "op_types": [
63
- "ConstantOfShape",
64
- "Slice",
 
65
  "Constant",
66
- "Expand",
67
- "Transpose",
68
- "Cast",
 
69
  "Add",
 
 
 
 
70
  "Squeeze",
71
  "Unsqueeze",
72
- "Less",
 
 
73
  "Div",
74
- "Sub",
75
- "Pow",
76
- "MatMul",
77
- "Mul",
78
  "Sigmoid",
79
- "ReduceMean",
80
- "Range",
81
- "Shape",
82
  "Equal",
83
- "Where",
84
- "Sqrt",
85
- "Concat",
86
- "Gather",
87
- "Softmax",
88
- "Reshape"
89
  ],
90
  "weight_type": "QInt8"
91
  },
92
- "decoder_model_merged": {
93
  "op_types": [
94
- "ConstantOfShape",
95
- "Slice",
96
  "Constant",
97
- "Expand",
98
- "Transpose",
99
- "Cast",
 
100
  "Add",
 
 
 
 
101
  "Squeeze",
102
  "Unsqueeze",
103
- "Less",
 
 
104
  "Div",
105
- "Sub",
106
- "Pow",
107
- "MatMul",
108
- "Mul",
109
  "Sigmoid",
110
- "ReduceMean",
111
- "Range",
112
- "Shape",
113
  "Equal",
114
- "Where",
115
- "Sqrt",
116
- "Concat",
117
- "Gather",
118
- "If",
119
- "Softmax",
120
- "Reshape"
121
  ],
122
  "weight_type": "QInt8"
123
  }
 
4
  "per_model_config": {
5
  "encoder_model": {
6
  "op_types": [
 
 
 
 
 
 
 
 
7
  "Sub",
8
+ "Gather",
9
+ "Constant",
10
  "MatMul",
11
+ "Where",
12
+ "Reshape",
13
+ "Add",
14
+ "Softmax",
15
  "ReduceMean",
 
16
  "Shape",
17
+ "Pow",
18
+ "Unsqueeze",
19
+ "Cast",
20
  "Sqrt",
21
  "Concat",
22
+ "Div",
23
+ "Expand",
24
+ "Sigmoid",
25
+ "Mul",
26
+ "Transpose",
27
+ "ConstantOfShape",
28
+ "Equal",
29
+ "Range"
30
  ],
31
  "weight_type": "QInt8"
32
  },
33
  "decoder_with_past_model": {
34
  "op_types": [
 
 
 
 
 
 
 
 
35
  "Sub",
36
+ "Gather",
37
+ "Constant",
38
  "MatMul",
39
+ "Where",
40
+ "Reshape",
41
+ "Add",
42
+ "Softmax",
43
  "ReduceMean",
 
44
  "Shape",
45
+ "Pow",
46
+ "Unsqueeze",
47
+ "Cast",
48
  "Sqrt",
49
  "Concat",
50
+ "Div",
51
+ "Expand",
52
+ "Sigmoid",
53
+ "Mul",
54
+ "Transpose",
55
+ "ConstantOfShape",
56
+ "Equal",
57
+ "Range"
58
  ],
59
  "weight_type": "QInt8"
60
  },
61
+ "decoder_model_merged": {
62
  "op_types": [
63
+ "Sub",
64
+ "If",
65
+ "Gather",
66
  "Constant",
67
+ "MatMul",
68
+ "Where",
69
+ "Reshape",
70
+ "Less",
71
  "Add",
72
+ "Softmax",
73
+ "ReduceMean",
74
+ "Shape",
75
+ "Pow",
76
  "Squeeze",
77
  "Unsqueeze",
78
+ "Cast",
79
+ "Sqrt",
80
+ "Concat",
81
  "Div",
82
+ "Slice",
83
+ "Expand",
 
 
84
  "Sigmoid",
85
+ "Mul",
86
+ "Transpose",
87
+ "ConstantOfShape",
88
  "Equal",
89
+ "Range"
 
 
 
 
 
90
  ],
91
  "weight_type": "QInt8"
92
  },
93
+ "decoder_model": {
94
  "op_types": [
95
+ "Sub",
96
+ "Gather",
97
  "Constant",
98
+ "MatMul",
99
+ "Where",
100
+ "Reshape",
101
+ "Less",
102
  "Add",
103
+ "Softmax",
104
+ "ReduceMean",
105
+ "Shape",
106
+ "Pow",
107
  "Squeeze",
108
  "Unsqueeze",
109
+ "Cast",
110
+ "Sqrt",
111
+ "Concat",
112
  "Div",
113
+ "Slice",
114
+ "Expand",
 
 
115
  "Sigmoid",
116
+ "Mul",
117
+ "Transpose",
118
+ "ConstantOfShape",
119
  "Equal",
120
+ "Range"
 
 
 
 
 
 
121
  ],
122
  "weight_type": "QInt8"
123
  }
tokenizer.json CHANGED
The diff for this file is too large to render. See raw diff