Xenova HF staff commited on
Commit
247a083
1 Parent(s): 7c833c7

Upload folder using huggingface_hub

Browse files
config.json CHANGED
@@ -31,7 +31,7 @@
31
  "summary_proj_to_labels": true,
32
  "summary_type": "cls_index",
33
  "summary_use_proj": true,
34
- "transformers_version": "4.32.0.dev0",
35
  "use_cache": true,
36
  "validate_runner_input": true,
37
  "vocab_size": 49152
 
31
  "summary_proj_to_labels": true,
32
  "summary_type": "cls_index",
33
  "summary_use_proj": true,
34
+ "transformers_version": "4.33.0.dev0",
35
  "use_cache": true,
36
  "validate_runner_input": true,
37
  "vocab_size": 49152
generation_config.json CHANGED
@@ -2,5 +2,5 @@
2
  "_from_model_config": true,
3
  "bos_token_id": 0,
4
  "eos_token_id": 0,
5
- "transformers_version": "4.32.0.dev0"
6
  }
 
2
  "_from_model_config": true,
3
  "bos_token_id": 0,
4
  "eos_token_id": 0,
5
+ "transformers_version": "4.33.0.dev0"
6
  }
onnx/decoder_model.onnx CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:086035920a278bf46877997575337c46cad5ffc746367a188da646df349f0cda
3
- size 875006955
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:994ccba5ea418052a3e1839b1b3c8b4c39342c8537fadf2dacec898daa64ca67
3
+ size 724012077
onnx/decoder_model_merged.onnx CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:e852014acc486bff04ac66ea20bb8e7a8b4b755d608efc404352439f586cbe20
3
- size 942448642
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:1b0f375b96dd3a76b25f4d5377eda6376faae560682bed258510f0bbd8289470
3
+ size 791453859
onnx/decoder_model_merged_quantized.onnx CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:45cf095deb2100ff660c231e8bff39af026c8848ad10ffaf57e89749c1e1c3e5
3
- size 338500911
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:be7061281a908c61b62bd0bc9dea7a707a72d55a374deab3ccfaec173acc2aa2
3
+ size 299954775
onnx/decoder_model_quantized.onnx CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:4910505d4c3186dd789d1d3dfc3e834eca200549185f43beb09bb3bd3e5edd49
3
- size 270851872
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:6889a8fdc9daf18b7aa3c6124684eaffd0fb2fba6ee0103064ad719d79d7bf0a
3
+ size 232306103
onnx/decoder_with_past_model.onnx CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:94f2798827a11f474b4e8a98f02182b66982c6f934fc959d2b2b74d016afbbd0
3
- size 875014307
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:a326b8671e0a683b78557edf7ae89d86172a09451931d1744839ce45c6c42fc9
3
+ size 724019429
onnx/decoder_with_past_model_quantized.onnx CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:bfc393be238b4a5bd3337c1165a8b17cd63a907a847a6b38b539f8cfe39c6be3
3
- size 270860829
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:66061e6a38529a523db921f29bebea1086e4753e60daff773ecfe8fb48ad0b10
3
+ size 232315060
quantize_config.json CHANGED
@@ -1,98 +1,98 @@
1
  {
2
- "per_channel": true,
3
- "reduce_range": true,
4
  "per_model_config": {
5
- "decoder_with_past_model": {
6
  "op_types": [
7
- "CumSum",
8
- "ConstantOfShape",
9
- "Cast",
10
- "Gather",
11
- "ReduceMean",
12
- "Reshape",
13
- "Concat",
14
- "Div",
15
- "Squeeze",
16
  "Unsqueeze",
17
  "Equal",
18
- "Pow",
 
 
19
  "Constant",
20
- "MatMul",
21
- "Sub",
22
- "And",
23
- "Slice",
24
  "Tanh",
25
- "Add",
 
 
 
 
 
26
  "Split",
 
 
27
  "Mul",
28
- "Shape",
29
- "Softmax",
30
- "Sqrt",
31
- "Where",
32
- "Transpose"
33
  ],
34
  "weight_type": "QInt8"
35
  },
36
- "decoder_model": {
37
  "op_types": [
38
- "CumSum",
39
- "ConstantOfShape",
40
- "Cast",
41
- "Gather",
42
- "ReduceMean",
43
- "Reshape",
44
- "Concat",
45
- "Div",
46
- "Squeeze",
47
  "Unsqueeze",
48
  "Equal",
49
- "Pow",
 
 
50
  "Constant",
51
- "MatMul",
52
- "Sub",
53
- "And",
54
- "Slice",
55
  "Tanh",
56
- "Add",
 
 
 
 
 
57
  "Split",
 
 
 
58
  "Mul",
59
- "Shape",
60
- "Softmax",
61
- "Sqrt",
62
- "Where",
63
- "Transpose"
64
  ],
65
  "weight_type": "QInt8"
66
  },
67
- "decoder_model_merged": {
68
  "op_types": [
69
- "CumSum",
70
- "ConstantOfShape",
71
- "Cast",
72
- "Gather",
73
- "ReduceMean",
74
- "Reshape",
75
- "Concat",
76
- "Div",
77
- "Squeeze",
78
- "If",
79
  "Unsqueeze",
80
  "Equal",
81
- "Pow",
 
 
82
  "Constant",
83
- "MatMul",
84
- "Sub",
85
- "And",
86
- "Slice",
87
  "Tanh",
88
- "Add",
 
 
 
 
 
89
  "Split",
 
 
90
  "Mul",
91
- "Shape",
92
- "Softmax",
93
- "Sqrt",
94
- "Where",
95
- "Transpose"
96
  ],
97
  "weight_type": "QInt8"
98
  }
 
1
  {
2
+ "per_channel": false,
3
+ "reduce_range": false,
4
  "per_model_config": {
5
+ "decoder_model": {
6
  "op_types": [
7
+ "Add",
8
+ "Shape",
9
+ "Where",
10
+ "Sub",
 
 
 
 
 
11
  "Unsqueeze",
12
  "Equal",
13
+ "Concat",
14
+ "Reshape",
15
+ "Softmax",
16
  "Constant",
 
 
 
 
17
  "Tanh",
18
+ "Gather",
19
+ "CumSum",
20
+ "Squeeze",
21
+ "Sqrt",
22
+ "Cast",
23
+ "Slice",
24
  "Split",
25
+ "Transpose",
26
+ "ConstantOfShape",
27
  "Mul",
28
+ "Div",
29
+ "MatMul",
30
+ "And",
31
+ "Pow",
32
+ "ReduceMean"
33
  ],
34
  "weight_type": "QInt8"
35
  },
36
+ "decoder_model_merged": {
37
  "op_types": [
38
+ "Add",
39
+ "Shape",
40
+ "Where",
41
+ "Sub",
 
 
 
 
 
42
  "Unsqueeze",
43
  "Equal",
44
+ "Concat",
45
+ "Reshape",
46
+ "Softmax",
47
  "Constant",
 
 
 
 
48
  "Tanh",
49
+ "Gather",
50
+ "CumSum",
51
+ "Squeeze",
52
+ "Sqrt",
53
+ "Cast",
54
+ "Slice",
55
  "Split",
56
+ "Transpose",
57
+ "ConstantOfShape",
58
+ "If",
59
  "Mul",
60
+ "Div",
61
+ "MatMul",
62
+ "And",
63
+ "Pow",
64
+ "ReduceMean"
65
  ],
66
  "weight_type": "QInt8"
67
  },
68
+ "decoder_with_past_model": {
69
  "op_types": [
70
+ "Add",
71
+ "Shape",
72
+ "Where",
73
+ "Sub",
 
 
 
 
 
 
74
  "Unsqueeze",
75
  "Equal",
76
+ "Concat",
77
+ "Reshape",
78
+ "Softmax",
79
  "Constant",
 
 
 
 
80
  "Tanh",
81
+ "Gather",
82
+ "Squeeze",
83
+ "CumSum",
84
+ "Sqrt",
85
+ "Slice",
86
+ "Cast",
87
  "Split",
88
+ "Transpose",
89
+ "ConstantOfShape",
90
  "Mul",
91
+ "Div",
92
+ "MatMul",
93
+ "And",
94
+ "Pow",
95
+ "ReduceMean"
96
  ],
97
  "weight_type": "QInt8"
98
  }