Xenova HF staff commited on
Commit
778861e
1 Parent(s): 7642a91

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": 59351
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": 59351
60
  }
generation_config.json CHANGED
@@ -1,5 +1,4 @@
1
  {
2
- "_from_model_config": true,
3
  "bad_words_ids": [
4
  [
5
  59350
@@ -12,5 +11,6 @@
12
  "max_length": 512,
13
  "num_beams": 6,
14
  "pad_token_id": 59350,
15
- "transformers_version": "4.32.0.dev0"
 
16
  }
 
1
  {
 
2
  "bad_words_ids": [
3
  [
4
  59350
 
11
  "max_length": 512,
12
  "num_beams": 6,
13
  "pad_token_id": 59350,
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 sha256:538b42ad1c00b1b76ccb398a8d64d3a6840784efd94e26d03aa43b4470478a08
3
- size 345560660
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:83d64a4245c0f353120f61080b9e4ee39dabf47c2b7933a86f62da58a8cbc79e
3
+ size 224009869
onnx/decoder_model_merged.onnx CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:0893ced2fad86faebc3295722bbbb3725c20b97ce4b13fa22e816502ff5c51b0
3
- size 345796142
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:df78df821bb9a8f6ff6281df6be2c3d4b13e1b9cdb7d83fb8a22769f53392e2b
3
+ size 224245436
onnx/decoder_model_merged_quantized.onnx CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:6392d9cf48fe503aebc417887b75c9b422856cbcb36253f46516f1f1866c1004
3
- size 87983283
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:062ee6c4f6c2318114239a681849667492667e21b5617458e73a27b1d2db23b7
3
+ size 57297404
onnx/decoder_model_quantized.onnx CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:a4a68a1fcc23f5548955fda29b221837e9a2d798a301536b66b206a11ed361da
3
- size 87611927
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:50678339d48e2aeb222c1f104a590b9d840fd43e46e155c6eeb333c29eb0643a
3
+ size 56926702
onnx/decoder_with_past_model.onnx CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:a20cfc70c0ac2f15982b32c006bc98781eb45fe10cde6eaf7c8f7f4b87ce9972
3
- size 332910151
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:a2a93531ac605c6c09c8ed5a18215c5ddcb241f2a9f3e5a0696d4e55e9fb6c33
3
+ size 211359360
onnx/decoder_with_past_model_quantized.onnx CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:e35a0c85cbd149805db3d1220e45727b110f1831b86346e0da8eee515e956f00
3
- size 84340463
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:3a0f9462d91293a64bb1d34e106985ea4358eabd3f079ecf9e6058fd25d992ef
3
+ size 53655238
quantize_config.json CHANGED
@@ -4,120 +4,120 @@
4
  "per_model_config": {
5
  "encoder_model": {
6
  "op_types": [
7
- "Softmax",
8
  "Where",
9
- "Unsqueeze",
10
- "Div",
11
- "Sqrt",
12
- "MatMul",
13
- "Reshape",
14
- "Transpose",
15
- "Sigmoid",
16
- "Pow",
17
- "Constant",
18
- "Mul",
19
- "ReduceMean",
20
- "Sub",
21
- "Equal",
22
- "Gather",
23
  "Shape",
24
  "Expand",
25
- "ConstantOfShape",
26
- "Concat",
27
  "Add",
 
 
 
 
 
28
  "Cast",
29
- "Range"
 
 
 
 
 
 
 
 
 
 
30
  ],
31
  "weight_type": "QInt8"
32
  },
33
  "decoder_with_past_model": {
34
  "op_types": [
35
- "Softmax",
36
  "Where",
37
- "Unsqueeze",
38
- "Div",
39
- "Sqrt",
40
- "MatMul",
41
- "Reshape",
42
- "Transpose",
43
- "Sigmoid",
44
- "Pow",
45
- "Constant",
46
- "Mul",
47
- "ReduceMean",
48
- "Sub",
49
- "Equal",
50
- "Gather",
51
  "Shape",
52
  "Expand",
53
- "ConstantOfShape",
54
- "Concat",
55
  "Add",
 
 
 
 
 
 
56
  "Cast",
57
- "Range"
 
 
 
 
 
 
 
 
 
58
  ],
59
  "weight_type": "QInt8"
60
  },
61
- "decoder_model": {
62
  "op_types": [
63
- "Softmax",
64
- "Where",
65
- "Unsqueeze",
66
- "Div",
67
  "Less",
 
 
 
 
 
 
 
 
 
68
  "Sqrt",
69
- "MatMul",
70
  "Squeeze",
 
 
 
 
 
 
 
71
  "Reshape",
72
- "Transpose",
73
- "Sigmoid",
74
- "Pow",
75
  "Constant",
76
- "Mul",
77
- "ReduceMean",
78
- "Sub",
79
  "Slice",
80
- "Equal",
81
- "Gather",
82
- "Shape",
83
- "Expand",
84
- "ConstantOfShape",
85
- "Concat",
86
- "Add",
87
- "Cast",
88
- "Range"
89
  ],
90
  "weight_type": "QInt8"
91
  },
92
- "decoder_model_merged": {
93
  "op_types": [
94
- "Softmax",
95
- "Where",
96
- "Unsqueeze",
97
- "Div",
98
  "Less",
 
 
 
 
 
 
 
 
 
99
  "Sqrt",
100
- "MatMul",
101
  "Squeeze",
 
 
 
 
 
 
102
  "Reshape",
103
- "Transpose",
104
- "Sigmoid",
105
- "Pow",
106
  "Constant",
107
- "Mul",
108
- "If",
109
- "ReduceMean",
110
- "Sub",
111
  "Slice",
112
- "Equal",
113
- "Gather",
114
- "Shape",
115
- "Expand",
116
- "ConstantOfShape",
117
- "Concat",
118
- "Add",
119
- "Cast",
120
- "Range"
121
  ],
122
  "weight_type": "QInt8"
123
  }
 
4
  "per_model_config": {
5
  "encoder_model": {
6
  "op_types": [
7
+ "ConstantOfShape",
8
  "Where",
 
 
 
 
 
 
 
 
 
 
 
 
 
 
9
  "Shape",
10
  "Expand",
11
+ "Pow",
 
12
  "Add",
13
+ "Gather",
14
+ "Sigmoid",
15
+ "Concat",
16
+ "Sqrt",
17
+ "ReduceMean",
18
  "Cast",
19
+ "Equal",
20
+ "Sub",
21
+ "Softmax",
22
+ "Div",
23
+ "Mul",
24
+ "Reshape",
25
+ "Constant",
26
+ "Transpose",
27
+ "Unsqueeze",
28
+ "Range",
29
+ "MatMul"
30
  ],
31
  "weight_type": "QInt8"
32
  },
33
  "decoder_with_past_model": {
34
  "op_types": [
35
+ "ConstantOfShape",
36
  "Where",
 
 
 
 
 
 
 
 
 
 
 
 
 
 
37
  "Shape",
38
  "Expand",
39
+ "Pow",
 
40
  "Add",
41
+ "Gather",
42
+ "Sigmoid",
43
+ "Concat",
44
+ "Sqrt",
45
+ "ReduceMean",
46
+ "Equal",
47
  "Cast",
48
+ "Sub",
49
+ "Softmax",
50
+ "Div",
51
+ "Mul",
52
+ "Reshape",
53
+ "Constant",
54
+ "Transpose",
55
+ "Unsqueeze",
56
+ "Range",
57
+ "MatMul"
58
  ],
59
  "weight_type": "QInt8"
60
  },
61
+ "decoder_model_merged": {
62
  "op_types": [
 
 
 
 
63
  "Less",
64
+ "ConstantOfShape",
65
+ "Where",
66
+ "Shape",
67
+ "Expand",
68
+ "Pow",
69
+ "Add",
70
+ "Gather",
71
+ "Sigmoid",
72
+ "Concat",
73
  "Sqrt",
74
+ "ReduceMean",
75
  "Squeeze",
76
+ "If",
77
+ "Cast",
78
+ "Equal",
79
+ "Sub",
80
+ "Softmax",
81
+ "Div",
82
+ "Mul",
83
  "Reshape",
 
 
 
84
  "Constant",
 
 
 
85
  "Slice",
86
+ "Transpose",
87
+ "Unsqueeze",
88
+ "Range",
89
+ "MatMul"
 
 
 
 
 
90
  ],
91
  "weight_type": "QInt8"
92
  },
93
+ "decoder_model": {
94
  "op_types": [
 
 
 
 
95
  "Less",
96
+ "ConstantOfShape",
97
+ "Where",
98
+ "Shape",
99
+ "Expand",
100
+ "Pow",
101
+ "Add",
102
+ "Gather",
103
+ "Sigmoid",
104
+ "Concat",
105
  "Sqrt",
106
+ "ReduceMean",
107
  "Squeeze",
108
+ "Cast",
109
+ "Equal",
110
+ "Sub",
111
+ "Softmax",
112
+ "Div",
113
+ "Mul",
114
  "Reshape",
 
 
 
115
  "Constant",
 
 
 
 
116
  "Slice",
117
+ "Transpose",
118
+ "Unsqueeze",
119
+ "Range",
120
+ "MatMul"
 
 
 
 
 
121
  ],
122
  "weight_type": "QInt8"
123
  }
tokenizer.json CHANGED
The diff for this file is too large to render. See raw diff