Xenova HF staff commited on
Commit
dd3afdf
1 Parent(s): 210faab

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": "4.33.0.dev0",
58
  "use_cache": true,
59
  "vocab_size": 53685
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": 53685
60
  }
generation_config.json CHANGED
@@ -12,5 +12,5 @@
12
  "num_beams": 4,
13
  "pad_token_id": 53684,
14
  "renormalize_logits": true,
15
- "transformers_version": "4.33.0.dev0"
16
  }
 
12
  "num_beams": 4,
13
  "pad_token_id": 53684,
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 sha256:2f0d893453f8a43d23b576dd6bf3fa1115f1518e3b91d9c40c14dbccb43b7343
3
- size 322330060
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:6749fbaa057790f6aa24044d6c4c8e9e9866c4c09643c7a626a212423ef62ef6
3
+ size 212383237
onnx/decoder_model_merged.onnx CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:39f01ded3e54e0bd052238352ffd14242b6a59a843fa1a6c51413ea4f232c3f1
3
- size 322565542
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:baf572c55c5201340cd91802ae3c3032c8dfcdc7f25ed27446a6a644cffce622
3
+ size 212618804
onnx/decoder_model_merged_quantized.onnx CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:1db1532aa7e2ac44e4fc4b5b7a239133c13589aff7ae5af2ebe9ef1df812a123
3
- size 82130305
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:1cfc6a84f9c189e6928c9710b281fae3f1f42d4ffbba109e7b4ec8e31ceb06dd
3
+ size 54373748
onnx/decoder_model_quantized.onnx CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:7e20ba038bf624f634480c1ad20e466d8b2c83d976ff97040df0af794004632e
3
- size 81758949
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:cde3e02aaca72cfaf2267f6c31c4dcc1b323d769bf90c273f2a1c1d56198ddba
3
+ size 54003046
onnx/decoder_with_past_model.onnx CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:283f351d005aba8e3922fed2d646f898dadbed4809602d62469f9ed4e6b62eba
3
- size 309679551
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:3efadf66d7039a6a5c42f565f9e99ac1f270f46535ab7363e0004ec635891b4d
3
+ size 199732728
onnx/decoder_with_past_model_quantized.onnx CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:e6bcc438c2493a8e6aaa818acf203ddd2884785fcbc9ab711fcdba4d29c4aa25
3
- size 78487485
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:e8f3945a5b393a3533930f920d0e98c5cb4d3804646611534360aa918b2489f2
3
+ size 50731582
quantize_config.json CHANGED
@@ -2,122 +2,122 @@
2
  "per_channel": true,
3
  "reduce_range": true,
4
  "per_model_config": {
5
- "decoder_model_merged": {
6
  "op_types": [
7
- "Sqrt",
8
- "Equal",
9
- "Cast",
10
- "Mul",
11
- "Slice",
12
- "Gather",
13
- "Constant",
14
- "ReduceMean",
15
- "Div",
16
- "Squeeze",
17
- "MatMul",
18
- "ConstantOfShape",
19
- "Reshape",
20
  "Pow",
21
- "Transpose",
22
- "Range",
23
- "Sub",
24
  "Shape",
25
- "If",
26
- "Unsqueeze",
 
27
  "Concat",
28
  "Add",
 
 
 
 
 
 
29
  "Where",
 
 
 
 
 
 
 
30
  "Less",
31
- "Sigmoid",
32
- "Expand",
33
- "Softmax"
 
34
  ],
35
  "weight_type": "QInt8"
36
  },
37
- "decoder_model": {
38
  "op_types": [
39
- "Sqrt",
40
- "Equal",
41
- "Cast",
42
- "Mul",
43
- "Slice",
44
- "Gather",
45
- "Constant",
46
- "ReduceMean",
47
- "Div",
48
- "Squeeze",
49
- "MatMul",
50
- "ConstantOfShape",
51
- "Reshape",
52
  "Pow",
53
- "Transpose",
54
- "Range",
55
- "Sub",
56
  "Shape",
57
- "Unsqueeze",
 
 
58
  "Concat",
59
  "Add",
60
- "Where",
61
- "Less",
62
- "Sigmoid",
 
63
  "Expand",
64
- "Softmax"
 
 
 
 
 
 
 
 
 
 
65
  ],
66
  "weight_type": "QInt8"
67
  },
68
- "decoder_with_past_model": {
69
  "op_types": [
70
- "Sqrt",
71
- "Equal",
72
- "Cast",
73
- "Mul",
74
- "Gather",
75
- "Constant",
76
- "ReduceMean",
77
- "Div",
78
- "MatMul",
79
- "ConstantOfShape",
80
- "Reshape",
81
  "Pow",
82
- "Transpose",
83
- "Range",
84
- "Sub",
85
  "Shape",
86
- "Unsqueeze",
 
 
87
  "Concat",
88
  "Add",
89
- "Where",
90
- "Sigmoid",
 
 
 
91
  "Expand",
92
- "Softmax"
 
 
 
 
 
 
 
 
 
 
 
 
 
93
  ],
94
  "weight_type": "QInt8"
95
  },
96
- "encoder_model": {
97
  "op_types": [
98
- "Sqrt",
99
- "Equal",
100
- "Cast",
101
- "Mul",
102
- "Gather",
103
- "Constant",
104
- "ReduceMean",
105
- "Div",
106
- "MatMul",
107
- "ConstantOfShape",
108
- "Reshape",
109
  "Pow",
110
- "Transpose",
111
- "Range",
112
- "Sub",
113
  "Shape",
114
- "Unsqueeze",
115
- "Concat",
 
116
  "Add",
117
- "Where",
118
- "Sigmoid",
 
 
 
119
  "Expand",
120
- "Softmax"
 
 
 
 
 
 
 
 
 
 
121
  ],
122
  "weight_type": "QInt8"
123
  }
 
2
  "per_channel": true,
3
  "reduce_range": true,
4
  "per_model_config": {
5
+ "decoder_model": {
6
  "op_types": [
 
 
 
 
 
 
 
 
 
 
 
 
 
7
  "Pow",
 
 
 
8
  "Shape",
9
+ "ReduceMean",
10
+ "Mul",
11
+ "Reshape",
12
  "Concat",
13
  "Add",
14
+ "Sqrt",
15
+ "Sub",
16
+ "Slice",
17
+ "Div",
18
+ "Cast",
19
+ "Expand",
20
  "Where",
21
+ "Range",
22
+ "Softmax",
23
+ "MatMul",
24
+ "Unsqueeze",
25
+ "Transpose",
26
+ "Squeeze",
27
+ "ConstantOfShape",
28
  "Less",
29
+ "Gather",
30
+ "Equal",
31
+ "Constant",
32
+ "Sigmoid"
33
  ],
34
  "weight_type": "QInt8"
35
  },
36
+ "encoder_model": {
37
  "op_types": [
 
 
 
 
 
 
 
 
 
 
 
 
 
38
  "Pow",
 
 
 
39
  "Shape",
40
+ "ReduceMean",
41
+ "Mul",
42
+ "Reshape",
43
  "Concat",
44
  "Add",
45
+ "Sqrt",
46
+ "Sub",
47
+ "Div",
48
+ "Cast",
49
  "Expand",
50
+ "Where",
51
+ "Range",
52
+ "Softmax",
53
+ "MatMul",
54
+ "Unsqueeze",
55
+ "Transpose",
56
+ "ConstantOfShape",
57
+ "Gather",
58
+ "Equal",
59
+ "Constant",
60
+ "Sigmoid"
61
  ],
62
  "weight_type": "QInt8"
63
  },
64
+ "decoder_model_merged": {
65
  "op_types": [
 
 
 
 
 
 
 
 
 
 
 
66
  "Pow",
 
 
 
67
  "Shape",
68
+ "ReduceMean",
69
+ "Mul",
70
+ "Reshape",
71
  "Concat",
72
  "Add",
73
+ "Sqrt",
74
+ "Sub",
75
+ "Slice",
76
+ "Div",
77
+ "Cast",
78
  "Expand",
79
+ "Where",
80
+ "Range",
81
+ "Softmax",
82
+ "MatMul",
83
+ "Unsqueeze",
84
+ "Transpose",
85
+ "Squeeze",
86
+ "If",
87
+ "ConstantOfShape",
88
+ "Less",
89
+ "Gather",
90
+ "Equal",
91
+ "Constant",
92
+ "Sigmoid"
93
  ],
94
  "weight_type": "QInt8"
95
  },
96
+ "decoder_with_past_model": {
97
  "op_types": [
 
 
 
 
 
 
 
 
 
 
 
98
  "Pow",
 
 
 
99
  "Shape",
100
+ "ReduceMean",
101
+ "Mul",
102
+ "Reshape",
103
  "Add",
104
+ "Concat",
105
+ "Sqrt",
106
+ "Sub",
107
+ "Div",
108
+ "Cast",
109
  "Expand",
110
+ "Where",
111
+ "Range",
112
+ "Softmax",
113
+ "MatMul",
114
+ "Unsqueeze",
115
+ "Transpose",
116
+ "ConstantOfShape",
117
+ "Gather",
118
+ "Equal",
119
+ "Constant",
120
+ "Sigmoid"
121
  ],
122
  "weight_type": "QInt8"
123
  }