Xenova HF staff commited on
Commit
2fcb351
1 Parent(s): 3b3d41e

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": 57642
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": 57642
60
  }
generation_config.json CHANGED
@@ -1,5 +1,4 @@
1
  {
2
- "_from_model_config": true,
3
  "bad_words_ids": [
4
  [
5
  57641
@@ -12,5 +11,6 @@
12
  "max_length": 512,
13
  "num_beams": 6,
14
  "pad_token_id": 57641,
15
- "transformers_version": "4.32.0.dev0"
 
16
  }
 
1
  {
 
2
  "bad_words_ids": [
3
  [
4
  57641
 
11
  "max_length": 512,
12
  "num_beams": 6,
13
  "pad_token_id": 57641,
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:e1693efe660470bd8adbfe3296b37b0bc1d9be33d8d6e369528c1e5f8308a23a
3
- size 338553760
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:07342521a547809bb38ab91a0939cc763c8373330f7670e9d86d8e7fc0065cb7
3
+ size 220503001
onnx/decoder_model_merged.onnx CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:6ac5a1d5e1401574c835a1c0acdac5074f57b72cb25160c0c84a5caac97d6ef3
3
- size 338789242
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:fad840794c922b7a83fe958fc2c88255c09bcf12cb1cb4099a31711017e34091
3
+ size 220738568
onnx/decoder_model_merged_quantized.onnx CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:d9c4fba6ac26690f6325dde5643a68b9d96bc214b44dbe85f5f9cd4c5903ed8d
3
- size 86217886
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:7915cb6d03674552b6f7a3bcd4fe04e4dedd2a16fc120b4bcab39a03c6e6ecb4
3
+ size 56415560
onnx/decoder_model_quantized.onnx CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:3139e0482d113a31a14514cf48b734121ab07fa252bfd42e97a8f784bbe75aa7
3
- size 85846530
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:0c5b9b864ee0628850d7f50ed24f8da3b0fddef285ffe23724d21fb0ac564c02
3
+ size 56044858
onnx/decoder_with_past_model.onnx CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:1684ed2ead95e4346e0d082dd77ef54eccd589c37125eae1ac4432b71bd8976e
3
- size 325903251
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:7ceccaef9e3fa65b633f1a5283c1d4b81761cd2bd18b86f2bede63a0dd1a48ca
3
+ size 207852492
onnx/decoder_with_past_model_quantized.onnx CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:645dd32b8c1dd06dd1c5149405a2b3963bb72d99915723d5d030f074b67baf23
3
- size 82575066
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:619507f34314419c196b4edd6fed2503730282bfd5df8bbba7ff7f743bb80e09
3
+ size 52773394
quantize_config.json CHANGED
@@ -4,120 +4,120 @@
4
  "per_model_config": {
5
  "encoder_model": {
6
  "op_types": [
7
- "Sqrt",
 
 
8
  "Unsqueeze",
9
- "Range",
 
10
  "Equal",
 
 
 
11
  "Sub",
 
 
 
 
 
 
12
  "ConstantOfShape",
 
 
13
  "Pow",
14
  "Constant",
15
- "Sigmoid",
16
- "Reshape",
17
- "MatMul",
18
- "Div",
19
- "Shape",
20
- "ReduceMean",
21
- "Cast",
22
- "Gather",
23
- "Where",
24
- "Add",
25
- "Expand",
26
- "Transpose",
27
- "Concat",
28
- "Softmax",
29
- "Mul"
30
  ],
31
  "weight_type": "QInt8"
32
  },
33
  "decoder_with_past_model": {
34
  "op_types": [
35
- "Sqrt",
 
 
36
  "Unsqueeze",
 
 
37
  "Equal",
 
 
 
38
  "Sub",
 
 
 
 
39
  "Range",
 
40
  "ConstantOfShape",
 
 
41
  "Pow",
42
  "Constant",
43
- "Sigmoid",
44
- "Reshape",
45
- "MatMul",
46
- "Div",
47
- "Shape",
48
- "ReduceMean",
49
- "Cast",
50
- "Gather",
51
- "Where",
52
- "Add",
53
- "Expand",
54
- "Transpose",
55
- "Concat",
56
- "Softmax",
57
- "Mul"
58
  ],
59
  "weight_type": "QInt8"
60
  },
61
- "decoder_model": {
62
  "op_types": [
 
 
 
63
  "Slice",
64
- "Sqrt",
65
  "Unsqueeze",
66
- "Range",
 
 
67
  "Equal",
 
 
 
 
68
  "Sub",
 
 
 
 
 
 
69
  "ConstantOfShape",
 
 
70
  "Pow",
71
  "Constant",
72
- "Sigmoid",
73
- "Reshape",
74
- "Less",
75
- "MatMul",
76
- "Div",
77
- "Shape",
78
- "ReduceMean",
79
- "Cast",
80
- "Gather",
81
- "Where",
82
- "Add",
83
- "Expand",
84
- "Transpose",
85
- "Concat",
86
- "Softmax",
87
- "Mul",
88
- "Squeeze"
89
  ],
90
  "weight_type": "QInt8"
91
  },
92
- "decoder_model_merged": {
93
  "op_types": [
 
 
 
94
  "Slice",
95
- "Sqrt",
96
  "Unsqueeze",
97
- "Range",
 
 
98
  "Equal",
 
 
 
99
  "Sub",
 
 
 
 
 
 
100
  "ConstantOfShape",
 
 
101
  "Pow",
102
  "Constant",
103
- "Sigmoid",
104
- "Reshape",
105
- "Less",
106
- "MatMul",
107
- "Div",
108
- "Shape",
109
- "ReduceMean",
110
- "Cast",
111
- "Gather",
112
- "If",
113
- "Where",
114
- "Add",
115
- "Expand",
116
- "Transpose",
117
- "Concat",
118
- "Softmax",
119
- "Mul",
120
- "Squeeze"
121
  ],
122
  "weight_type": "QInt8"
123
  }
 
4
  "per_model_config": {
5
  "encoder_model": {
6
  "op_types": [
7
+ "Add",
8
+ "Mul",
9
+ "ReduceMean",
10
  "Unsqueeze",
11
+ "MatMul",
12
+ "Transpose",
13
  "Equal",
14
+ "Expand",
15
+ "Sqrt",
16
+ "Div",
17
  "Sub",
18
+ "Where",
19
+ "Gather",
20
+ "Concat",
21
+ "Cast",
22
+ "Range",
23
+ "Sigmoid",
24
  "ConstantOfShape",
25
+ "Shape",
26
+ "Reshape",
27
  "Pow",
28
  "Constant",
29
+ "Softmax"
 
 
 
 
 
 
 
 
 
 
 
 
 
 
30
  ],
31
  "weight_type": "QInt8"
32
  },
33
  "decoder_with_past_model": {
34
  "op_types": [
35
+ "Add",
36
+ "Mul",
37
+ "ReduceMean",
38
  "Unsqueeze",
39
+ "MatMul",
40
+ "Transpose",
41
  "Equal",
42
+ "Expand",
43
+ "Sqrt",
44
+ "Div",
45
  "Sub",
46
+ "Where",
47
+ "Gather",
48
+ "Concat",
49
+ "Cast",
50
  "Range",
51
+ "Sigmoid",
52
  "ConstantOfShape",
53
+ "Shape",
54
+ "Reshape",
55
  "Pow",
56
  "Constant",
57
+ "Softmax"
 
 
 
 
 
 
 
 
 
 
 
 
 
 
58
  ],
59
  "weight_type": "QInt8"
60
  },
61
+ "decoder_model_merged": {
62
  "op_types": [
63
+ "Add",
64
+ "Mul",
65
+ "Squeeze",
66
  "Slice",
67
+ "ReduceMean",
68
  "Unsqueeze",
69
+ "MatMul",
70
+ "Less",
71
+ "If",
72
  "Equal",
73
+ "Transpose",
74
+ "Expand",
75
+ "Sqrt",
76
+ "Div",
77
  "Sub",
78
+ "Where",
79
+ "Gather",
80
+ "Concat",
81
+ "Cast",
82
+ "Range",
83
+ "Sigmoid",
84
  "ConstantOfShape",
85
+ "Shape",
86
+ "Reshape",
87
  "Pow",
88
  "Constant",
89
+ "Softmax"
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
90
  ],
91
  "weight_type": "QInt8"
92
  },
93
+ "decoder_model": {
94
  "op_types": [
95
+ "Add",
96
+ "Mul",
97
+ "Squeeze",
98
  "Slice",
99
+ "ReduceMean",
100
  "Unsqueeze",
101
+ "MatMul",
102
+ "Less",
103
+ "Transpose",
104
  "Equal",
105
+ "Expand",
106
+ "Sqrt",
107
+ "Div",
108
  "Sub",
109
+ "Where",
110
+ "Gather",
111
+ "Concat",
112
+ "Cast",
113
+ "Range",
114
+ "Sigmoid",
115
  "ConstantOfShape",
116
+ "Shape",
117
+ "Reshape",
118
  "Pow",
119
  "Constant",
120
+ "Softmax"
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
121
  ],
122
  "weight_type": "QInt8"
123
  }
tokenizer.json CHANGED
The diff for this file is too large to render. See raw diff