Xenova HF staff commited on
Commit
318c9f6
1 Parent(s): b01df13

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": 63430
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": 63430
60
  }
generation_config.json CHANGED
@@ -1,5 +1,4 @@
1
  {
2
- "_from_model_config": true,
3
  "bad_words_ids": [
4
  [
5
  63429
@@ -12,5 +11,6 @@
12
  "max_length": 512,
13
  "num_beams": 6,
14
  "pad_token_id": 63429,
15
- "transformers_version": "4.32.0.dev0"
 
16
  }
 
1
  {
 
2
  "bad_words_ids": [
3
  [
4
  63429
 
11
  "max_length": 512,
12
  "num_beams": 6,
13
  "pad_token_id": 63429,
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 sha256:7c18775d580b9cd37793cdf62fbb55bca1190fa0ee1c39d306cd044b60fae75c
3
- size 362284560
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:1642297bb5f9b396f88ea3af48a85f5d06507f83f7af033923671e33ebf2d12a
3
+ size 232379977
onnx/decoder_model_merged.onnx CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:96ea8ab3b0c270703cacaefe81edd7990533a2f13c7b98eb137cdd5df0e9d00f
3
- size 362520042
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:22cfe606ebea2efd22a70debe337d70a1905eda1c04e0cb3606de6eafb777d1d
3
+ size 232615544
onnx/decoder_model_merged_quantized.onnx CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:8fd50549e892b3bd215724642be3d40ed5e39e6086ca2f32a6385e733f7c7bb3
3
- size 92196890
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:9ad9da5bce473ca06a9b0a3dbf0793675827e5668ab501e5dc65dac8cecd93ba
3
+ size 59402168
onnx/decoder_model_quantized.onnx CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:c42f2387a2a7bbf3dbaa0af9997d3196e31908eb66866345c1ec2113a2966be6
3
- size 91825534
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:a459f0f27fcb96752c6b2c7cc14e576fe2049247c5dc310283fb783cce6908ca
3
+ size 59031466
onnx/decoder_with_past_model.onnx CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:59d291ef636941d5da7c70b3ceb8949a350718b7249ec81d50248c1ae43897a7
3
- size 349634051
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:19516253193fde563398f48b74dc03b0aea9e0e9e31dcfe7f8aeb1fe0256c7c9
3
+ size 219729468
onnx/decoder_with_past_model_quantized.onnx CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:503979002355a9e49f41f6e2adf35c274c536ff8d8b4c6315f517826ebb64f5e
3
- size 88554070
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:efc22e31b5e96362da3b8530412fda66e17972aff7fa40e609d347120c3eec46
3
+ size 55760002
quantize_config.json CHANGED
@@ -4,120 +4,120 @@
4
  "per_model_config": {
5
  "encoder_model": {
6
  "op_types": [
7
- "Equal",
8
- "Sigmoid",
9
- "Sub",
10
- "Mul",
11
- "Range",
12
  "Shape",
 
13
  "Reshape",
 
 
 
 
 
14
  "ReduceMean",
15
- "Div",
 
16
  "Gather",
17
  "Cast",
 
 
18
  "Sqrt",
19
- "Add",
20
- "ConstantOfShape",
21
- "MatMul",
22
- "Where",
23
  "Expand",
24
- "Softmax",
25
- "Transpose",
26
- "Constant",
27
- "Unsqueeze",
28
- "Concat",
29
- "Pow"
30
  ],
31
  "weight_type": "QInt8"
32
  },
33
  "decoder_with_past_model": {
34
  "op_types": [
35
- "Equal",
36
- "Sigmoid",
37
- "Sub",
38
- "Mul",
39
- "Range",
40
  "Shape",
 
41
  "Reshape",
 
 
 
 
 
42
  "ReduceMean",
43
- "Div",
 
44
  "Gather",
45
  "Cast",
 
 
46
  "Sqrt",
47
- "Add",
48
- "ConstantOfShape",
49
- "MatMul",
50
- "Where",
51
  "Expand",
52
- "Softmax",
53
- "Transpose",
54
- "Constant",
55
- "Unsqueeze",
56
- "Concat",
57
- "Pow"
58
  ],
59
  "weight_type": "QInt8"
60
  },
61
- "decoder_model": {
62
  "op_types": [
63
- "Equal",
64
- "Sigmoid",
65
- "Sub",
66
- "Mul",
67
- "Range",
68
  "Shape",
69
- "Reshape",
70
- "ReduceMean",
71
- "Less",
72
  "Div",
73
- "Gather",
74
- "Cast",
75
- "Sqrt",
76
- "Slice",
77
- "Add",
78
  "Squeeze",
 
 
79
  "ConstantOfShape",
80
  "MatMul",
 
81
  "Where",
82
- "Expand",
83
- "Transpose",
84
  "Softmax",
85
- "Constant",
86
  "Unsqueeze",
87
- "Concat",
88
- "Pow"
 
 
 
 
 
 
 
 
89
  ],
90
  "weight_type": "QInt8"
91
  },
92
- "decoder_model_merged": {
93
  "op_types": [
94
- "Equal",
95
- "Sigmoid",
96
- "Sub",
97
- "Mul",
98
- "Range",
99
  "Shape",
100
- "Reshape",
101
- "ReduceMean",
102
- "Less",
103
  "Div",
104
- "Gather",
105
- "Cast",
106
- "Sqrt",
107
- "Slice",
108
- "Add",
109
  "Squeeze",
 
 
110
  "ConstantOfShape",
111
  "MatMul",
 
112
  "Where",
113
- "If",
114
- "Expand",
115
- "Transpose",
116
  "Softmax",
117
- "Constant",
118
  "Unsqueeze",
119
- "Concat",
120
- "Pow"
 
 
 
 
 
 
 
 
121
  ],
122
  "weight_type": "QInt8"
123
  }
 
4
  "per_model_config": {
5
  "encoder_model": {
6
  "op_types": [
 
 
 
 
 
7
  "Shape",
8
+ "Div",
9
  "Reshape",
10
+ "Concat",
11
+ "Constant",
12
+ "Range",
13
+ "ConstantOfShape",
14
+ "MatMul",
15
  "ReduceMean",
16
+ "Where",
17
+ "Add",
18
  "Gather",
19
  "Cast",
20
+ "Unsqueeze",
21
+ "Transpose",
22
  "Sqrt",
23
+ "Sigmoid",
24
+ "Mul",
25
+ "Pow",
26
+ "Equal",
27
  "Expand",
28
+ "Sub",
29
+ "Softmax"
 
 
 
 
30
  ],
31
  "weight_type": "QInt8"
32
  },
33
  "decoder_with_past_model": {
34
  "op_types": [
 
 
 
 
 
35
  "Shape",
36
+ "Div",
37
  "Reshape",
38
+ "Concat",
39
+ "Constant",
40
+ "Range",
41
+ "ConstantOfShape",
42
+ "MatMul",
43
  "ReduceMean",
44
+ "Where",
45
+ "Add",
46
  "Gather",
47
  "Cast",
48
+ "Unsqueeze",
49
+ "Transpose",
50
  "Sqrt",
51
+ "Sigmoid",
52
+ "Mul",
53
+ "Pow",
54
+ "Equal",
55
  "Expand",
56
+ "Sub",
57
+ "Softmax"
 
 
 
 
58
  ],
59
  "weight_type": "QInt8"
60
  },
61
+ "decoder_model_merged": {
62
  "op_types": [
 
 
 
 
 
63
  "Shape",
64
+ "If",
 
 
65
  "Div",
66
+ "Reshape",
67
+ "Concat",
 
 
 
68
  "Squeeze",
69
+ "Constant",
70
+ "Range",
71
  "ConstantOfShape",
72
  "MatMul",
73
+ "ReduceMean",
74
  "Where",
75
+ "Add",
76
+ "Gather",
77
  "Softmax",
78
+ "Slice",
79
  "Unsqueeze",
80
+ "Transpose",
81
+ "Sqrt",
82
+ "Sigmoid",
83
+ "Mul",
84
+ "Pow",
85
+ "Equal",
86
+ "Less",
87
+ "Expand",
88
+ "Sub",
89
+ "Cast"
90
  ],
91
  "weight_type": "QInt8"
92
  },
93
+ "decoder_model": {
94
  "op_types": [
 
 
 
 
 
95
  "Shape",
 
 
 
96
  "Div",
97
+ "Reshape",
98
+ "Concat",
 
 
 
99
  "Squeeze",
100
+ "Constant",
101
+ "Range",
102
  "ConstantOfShape",
103
  "MatMul",
104
+ "ReduceMean",
105
  "Where",
106
+ "Add",
107
+ "Gather",
 
108
  "Softmax",
109
+ "Slice",
110
  "Unsqueeze",
111
+ "Transpose",
112
+ "Sqrt",
113
+ "Sigmoid",
114
+ "Mul",
115
+ "Pow",
116
+ "Equal",
117
+ "Less",
118
+ "Expand",
119
+ "Sub",
120
+ "Cast"
121
  ],
122
  "weight_type": "QInt8"
123
  }
tokenizer.json CHANGED
The diff for this file is too large to render. See raw diff