Xenova HF staff commited on
Commit
c4d8042
1 Parent(s): 0be48a5

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": 61153
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": 61153
60
  }
generation_config.json CHANGED
@@ -1,5 +1,4 @@
1
  {
2
- "_from_model_config": true,
3
  "bad_words_ids": [
4
  [
5
  61152
@@ -12,5 +11,6 @@
12
  "max_length": 512,
13
  "num_beams": 4,
14
  "pad_token_id": 61152,
15
- "transformers_version": "4.32.0.dev0"
 
16
  }
 
1
  {
 
2
  "bad_words_ids": [
3
  [
4
  61152
 
11
  "max_length": 512,
12
  "num_beams": 4,
13
  "pad_token_id": 61152,
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:dc555a6f1b7a383209298937dcb03a9abd6a37bed7033b5b8dafba3a0a8dfb5b
3
- size 352948860
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:078406dc2a32f869faa857b5fdfacb22fcef3da41bd7c3c118c312332f4732d5
3
+ size 227707573
onnx/decoder_model_merged.onnx CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:5dfb6e4d047b19678ba162e0ecc384e384b62c05797a9c5a414f91d89d0c638e
3
- size 353184342
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:0e1f3e1cdaff1a85beda82021cc9aeac09bb9d380eedf3b9b5cfc979f8789e13
3
+ size 227943140
onnx/decoder_model_merged_quantized.onnx CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:f3d2ad2fa2095736bc7b83dea2da555d8e4071c5e54a45837af7e1517f30bda4
3
- size 89844749
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:d3f4dbe5fbdeae2668a189edfc17edd87876f6df7ff5282803d535c5980bd4a4
3
+ size 58227236
onnx/decoder_model_quantized.onnx CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:15b3f009f8f25662c7017df9c1995e0cbd288bf8e6005f6ddf2d54758119410d
3
- size 89473393
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:45c6444d3e25d077fe9c64217c11d2ff861a62e1405ca0f65c5a556c3a2a1eb4
3
+ size 57856534
onnx/decoder_with_past_model.onnx CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:ebd4c5b8bd57fab2e9d925f687774b5c8a67932eb81a0133fd7367412b2def99
3
- size 340298351
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:86a236fc66ea9a56eba4fe6d5448ac05f83841b11d2719e129c62823d96e594a
3
+ size 215057064
onnx/decoder_with_past_model_quantized.onnx CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:5cd6994449346d50a477173c14a6e4a5862f6778ab1812d26ced67b63dc493c1
3
- size 86201929
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:74393a9fffc9f2302b5d577897fb94d2193182c061f3ba83e159da2bababc94e
3
+ size 54585070
quantize_config.json CHANGED
@@ -4,120 +4,120 @@
4
  "per_model_config": {
5
  "encoder_model": {
6
  "op_types": [
7
- "Equal",
8
- "Gather",
9
- "Sub",
10
- "Cast",
11
- "Expand",
12
  "MatMul",
13
  "ConstantOfShape",
14
- "Sigmoid",
 
15
  "Concat",
16
- "Shape",
17
- "Pow",
18
  "Mul",
19
- "Add",
20
  "Softmax",
 
 
 
 
 
 
 
21
  "Range",
22
- "Constant",
 
23
  "Sqrt",
24
- "Reshape",
25
- "Where",
26
- "ReduceMean",
27
- "Div",
28
- "Unsqueeze",
29
- "Transpose"
30
  ],
31
  "weight_type": "QInt8"
32
  },
33
  "decoder_with_past_model": {
34
  "op_types": [
35
- "Equal",
36
- "Gather",
37
- "Sub",
38
- "Cast",
39
- "Expand",
40
  "MatMul",
41
  "ConstantOfShape",
42
- "Sigmoid",
 
43
  "Concat",
44
- "Shape",
45
- "Pow",
46
  "Mul",
47
- "Add",
48
  "Softmax",
 
 
 
 
 
 
 
49
  "Range",
50
- "Constant",
 
51
  "Sqrt",
52
- "Reshape",
53
- "Where",
54
- "ReduceMean",
55
- "Div",
56
- "Unsqueeze",
57
- "Transpose"
58
  ],
59
  "weight_type": "QInt8"
60
  },
61
- "decoder_model": {
62
  "op_types": [
63
- "Equal",
 
64
  "Slice",
65
- "Gather",
66
- "Sub",
67
- "Cast",
68
- "Expand",
69
  "MatMul",
70
  "ConstantOfShape",
71
- "Sigmoid",
72
- "Squeeze",
73
- "Concat",
74
- "Shape",
75
  "Less",
76
- "Pow",
77
  "Mul",
78
- "Add",
79
  "Softmax",
 
 
 
 
 
 
 
 
80
  "Range",
81
- "Constant",
 
 
82
  "Sqrt",
83
- "Reshape",
84
- "Where",
85
- "ReduceMean",
86
- "Div",
87
- "Unsqueeze",
88
- "Transpose"
89
  ],
90
  "weight_type": "QInt8"
91
  },
92
- "decoder_model_merged": {
93
  "op_types": [
94
- "If",
95
- "Equal",
96
  "Slice",
97
- "Gather",
98
- "Sub",
99
- "Cast",
100
- "Expand",
101
  "MatMul",
102
  "ConstantOfShape",
103
- "Sigmoid",
104
- "Squeeze",
105
- "Concat",
106
- "Shape",
107
  "Less",
108
- "Pow",
109
  "Mul",
110
- "Add",
111
  "Softmax",
 
 
 
 
 
 
 
 
112
  "Range",
113
- "Constant",
 
114
  "Sqrt",
115
- "Reshape",
116
- "Where",
117
- "ReduceMean",
118
- "Div",
119
- "Unsqueeze",
120
- "Transpose"
121
  ],
122
  "weight_type": "QInt8"
123
  }
 
4
  "per_model_config": {
5
  "encoder_model": {
6
  "op_types": [
7
+ "Div",
8
+ "Sigmoid",
9
+ "Reshape",
 
 
10
  "MatMul",
11
  "ConstantOfShape",
12
+ "ReduceMean",
13
+ "Sub",
14
  "Concat",
 
 
15
  "Mul",
16
+ "Constant",
17
  "Softmax",
18
+ "Equal",
19
+ "Unsqueeze",
20
+ "Add",
21
+ "Gather",
22
+ "Transpose",
23
+ "Pow",
24
+ "Expand",
25
  "Range",
26
+ "Shape",
27
+ "Cast",
28
  "Sqrt",
29
+ "Where"
 
 
 
 
 
30
  ],
31
  "weight_type": "QInt8"
32
  },
33
  "decoder_with_past_model": {
34
  "op_types": [
35
+ "Div",
36
+ "Sigmoid",
37
+ "Reshape",
 
 
38
  "MatMul",
39
  "ConstantOfShape",
40
+ "ReduceMean",
41
+ "Sub",
42
  "Concat",
 
 
43
  "Mul",
44
+ "Constant",
45
  "Softmax",
46
+ "Equal",
47
+ "Unsqueeze",
48
+ "Add",
49
+ "Gather",
50
+ "Transpose",
51
+ "Pow",
52
+ "Expand",
53
  "Range",
54
+ "Shape",
55
+ "Cast",
56
  "Sqrt",
57
+ "Where"
 
 
 
 
 
58
  ],
59
  "weight_type": "QInt8"
60
  },
61
+ "decoder_model_merged": {
62
  "op_types": [
63
+ "Div",
64
+ "Sigmoid",
65
  "Slice",
66
+ "Reshape",
 
 
 
67
  "MatMul",
68
  "ConstantOfShape",
69
+ "ReduceMean",
70
+ "Sub",
 
 
71
  "Less",
72
+ "Concat",
73
  "Mul",
74
+ "Constant",
75
  "Softmax",
76
+ "Squeeze",
77
+ "Equal",
78
+ "Unsqueeze",
79
+ "Add",
80
+ "Gather",
81
+ "Transpose",
82
+ "Pow",
83
+ "Expand",
84
  "Range",
85
+ "Shape",
86
+ "Cast",
87
+ "If",
88
  "Sqrt",
89
+ "Where"
 
 
 
 
 
90
  ],
91
  "weight_type": "QInt8"
92
  },
93
+ "decoder_model": {
94
  "op_types": [
95
+ "Div",
96
+ "Sigmoid",
97
  "Slice",
98
+ "Reshape",
 
 
 
99
  "MatMul",
100
  "ConstantOfShape",
101
+ "ReduceMean",
102
+ "Sub",
 
 
103
  "Less",
104
+ "Concat",
105
  "Mul",
106
+ "Constant",
107
  "Softmax",
108
+ "Squeeze",
109
+ "Equal",
110
+ "Unsqueeze",
111
+ "Add",
112
+ "Gather",
113
+ "Transpose",
114
+ "Pow",
115
+ "Expand",
116
  "Range",
117
+ "Shape",
118
+ "Cast",
119
  "Sqrt",
120
+ "Where"
 
 
 
 
 
121
  ],
122
  "weight_type": "QInt8"
123
  }
tokenizer.json CHANGED
The diff for this file is too large to render. See raw diff