sappho192 commited on
Commit
68c1d27
1 Parent(s): fec1af8

Upload quantized onnx (dynamic,avx2)

Browse files
onnxq/config.json ADDED
@@ -0,0 +1,186 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_name_or_path": "C:\\Users\\tikim\\.cache\\huggingface\\hub\\models--sappho192--aihub-ja-ko-translator\\snapshots\\d6877eb087ddb7fa96848ea09eb17e7901ae055c\\onnx",
3
+ "architectures": [
4
+ "EncoderDecoderModel"
5
+ ],
6
+ "decoder": {
7
+ "_name_or_path": "skt/kogpt2-base-v2",
8
+ "_num_labels": 1,
9
+ "activation_function": "gelu_new",
10
+ "add_cross_attention": true,
11
+ "architectures": [
12
+ "GPT2LMHeadModel"
13
+ ],
14
+ "attn_pdrop": 0.1,
15
+ "author": "Heewon Jeon(madjakarta@gmail.com)",
16
+ "bad_words_ids": null,
17
+ "begin_suppress_tokens": null,
18
+ "bos_token_id": 0,
19
+ "chunk_size_feed_forward": 0,
20
+ "created_date": "2021-04-28",
21
+ "cross_attention_hidden_size": null,
22
+ "decoder_start_token_id": null,
23
+ "diversity_penalty": 0.0,
24
+ "do_sample": false,
25
+ "early_stopping": false,
26
+ "embd_pdrop": 0.1,
27
+ "encoder_no_repeat_ngram_size": 0,
28
+ "eos_token_id": 1,
29
+ "exponential_decay_length_penalty": null,
30
+ "finetuning_task": null,
31
+ "forced_bos_token_id": null,
32
+ "forced_eos_token_id": null,
33
+ "gradient_checkpointing": false,
34
+ "id2label": {
35
+ "0": "LABEL_0"
36
+ },
37
+ "initializer_range": 0.02,
38
+ "is_decoder": true,
39
+ "is_encoder_decoder": false,
40
+ "label2id": {
41
+ "LABEL_0": 0
42
+ },
43
+ "layer_norm_epsilon": 1e-05,
44
+ "length_penalty": 1.0,
45
+ "license": "CC-BY-NC-SA 4.0",
46
+ "max_length": 20,
47
+ "min_length": 0,
48
+ "model_type": "gpt2",
49
+ "n_ctx": 1024,
50
+ "n_embd": 768,
51
+ "n_head": 12,
52
+ "n_inner": null,
53
+ "n_layer": 12,
54
+ "n_positions": 1024,
55
+ "no_repeat_ngram_size": 0,
56
+ "num_beam_groups": 1,
57
+ "num_beams": 1,
58
+ "num_return_sequences": 1,
59
+ "output_attentions": false,
60
+ "output_hidden_states": false,
61
+ "output_scores": false,
62
+ "pad_token_id": 3,
63
+ "prefix": null,
64
+ "problem_type": null,
65
+ "pruned_heads": {},
66
+ "remove_invalid_values": false,
67
+ "reorder_and_upcast_attn": false,
68
+ "repetition_penalty": 1.0,
69
+ "resid_pdrop": 0.1,
70
+ "return_dict": true,
71
+ "return_dict_in_generate": false,
72
+ "scale_attn_by_inverse_layer_idx": false,
73
+ "scale_attn_weights": true,
74
+ "sep_token_id": null,
75
+ "summary_activation": null,
76
+ "summary_first_dropout": 0.1,
77
+ "summary_proj_to_labels": true,
78
+ "summary_type": "cls_index",
79
+ "summary_use_proj": true,
80
+ "suppress_tokens": null,
81
+ "task_specific_params": {
82
+ "text-generation": {
83
+ "do_sample": true,
84
+ "max_length": 50
85
+ }
86
+ },
87
+ "temperature": 1.0,
88
+ "tf_legacy_loss": false,
89
+ "tie_encoder_decoder": false,
90
+ "tie_word_embeddings": true,
91
+ "tokenizer_class": null,
92
+ "top_k": 50,
93
+ "top_p": 1.0,
94
+ "torch_dtype": null,
95
+ "torchscript": false,
96
+ "typical_p": 1.0,
97
+ "use_bfloat16": false,
98
+ "use_cache": true,
99
+ "vocab_size": 51200
100
+ },
101
+ "decoder_start_token_id": 1,
102
+ "encoder": {
103
+ "_name_or_path": "cl-tohoku/bert-base-japanese-v2",
104
+ "add_cross_attention": false,
105
+ "architectures": [
106
+ "BertForMaskedLM"
107
+ ],
108
+ "attention_probs_dropout_prob": 0.1,
109
+ "bad_words_ids": null,
110
+ "begin_suppress_tokens": null,
111
+ "bos_token_id": null,
112
+ "chunk_size_feed_forward": 0,
113
+ "classifier_dropout": null,
114
+ "cross_attention_hidden_size": null,
115
+ "decoder_start_token_id": null,
116
+ "diversity_penalty": 0.0,
117
+ "do_sample": false,
118
+ "early_stopping": false,
119
+ "encoder_no_repeat_ngram_size": 0,
120
+ "eos_token_id": null,
121
+ "exponential_decay_length_penalty": null,
122
+ "finetuning_task": null,
123
+ "forced_bos_token_id": null,
124
+ "forced_eos_token_id": null,
125
+ "hidden_act": "gelu",
126
+ "hidden_dropout_prob": 0.1,
127
+ "hidden_size": 768,
128
+ "id2label": {
129
+ "0": "LABEL_0",
130
+ "1": "LABEL_1"
131
+ },
132
+ "initializer_range": 0.02,
133
+ "intermediate_size": 3072,
134
+ "is_decoder": false,
135
+ "is_encoder_decoder": false,
136
+ "label2id": {
137
+ "LABEL_0": 0,
138
+ "LABEL_1": 1
139
+ },
140
+ "layer_norm_eps": 1e-12,
141
+ "length_penalty": 1.0,
142
+ "max_length": 20,
143
+ "max_position_embeddings": 512,
144
+ "min_length": 0,
145
+ "model_type": "bert",
146
+ "no_repeat_ngram_size": 0,
147
+ "num_attention_heads": 12,
148
+ "num_beam_groups": 1,
149
+ "num_beams": 1,
150
+ "num_hidden_layers": 12,
151
+ "num_return_sequences": 1,
152
+ "output_attentions": false,
153
+ "output_hidden_states": false,
154
+ "output_scores": false,
155
+ "pad_token_id": 0,
156
+ "position_embedding_type": "absolute",
157
+ "prefix": null,
158
+ "problem_type": null,
159
+ "pruned_heads": {},
160
+ "remove_invalid_values": false,
161
+ "repetition_penalty": 1.0,
162
+ "return_dict": true,
163
+ "return_dict_in_generate": false,
164
+ "sep_token_id": null,
165
+ "suppress_tokens": null,
166
+ "task_specific_params": null,
167
+ "temperature": 1.0,
168
+ "tf_legacy_loss": false,
169
+ "tie_encoder_decoder": false,
170
+ "tie_word_embeddings": true,
171
+ "tokenizer_class": "BertJapaneseTokenizer",
172
+ "top_k": 50,
173
+ "top_p": 1.0,
174
+ "torch_dtype": null,
175
+ "torchscript": false,
176
+ "type_vocab_size": 2,
177
+ "typical_p": 1.0,
178
+ "use_bfloat16": false,
179
+ "use_cache": true,
180
+ "vocab_size": 32768
181
+ },
182
+ "is_encoder_decoder": true,
183
+ "model_type": "encoder-decoder",
184
+ "pad_token_id": 1,
185
+ "transformers_version": "4.36.2"
186
+ }
onnxq/decoder_model.onnx ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:dccd55e39cf23ad083f9c2349973143de25533c8dcdd9be936ae47cbfc4c386a
3
+ size 157181120
onnxq/decoder_model_merged.onnx ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:89483106eccf5fc934e82dee77f198a2fc6616725da2ebf67462d31337f35a31
3
+ size 159578955
onnxq/decoder_with_past_model.onnx ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:155e32a6effbc50c9c4acbc8eabd897cac0c96f32f4946760debcdde48d5730c
3
+ size 157172983
onnxq/encoder_model.onnx ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:576e648583182827867a24302dbbeef1aa8002bd92a6efd09e10ee6f2d273fd1
3
+ size 111923966
onnxq/ort_config.json ADDED
@@ -0,0 +1,35 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "one_external_file": true,
3
+ "opset": null,
4
+ "optimization": {},
5
+ "optimum_version": "1.16.1",
6
+ "quantization": {
7
+ "activations_dtype": "QUInt8",
8
+ "activations_symmetric": false,
9
+ "format": "QOperator",
10
+ "is_static": false,
11
+ "mode": "IntegerOps",
12
+ "nodes_to_exclude": [],
13
+ "nodes_to_quantize": [],
14
+ "operators_to_quantize": [
15
+ "Conv",
16
+ "MatMul",
17
+ "Attention",
18
+ "LSTM",
19
+ "Gather",
20
+ "Transpose",
21
+ "EmbedLayerNormalization"
22
+ ],
23
+ "per_channel": true,
24
+ "qdq_add_pair_to_weight": false,
25
+ "qdq_dedicated_pair": false,
26
+ "qdq_op_type_per_channel_support_to_axis": {
27
+ "MatMul": 1
28
+ },
29
+ "reduce_range": false,
30
+ "weights_dtype": "QUInt8",
31
+ "weights_symmetric": true
32
+ },
33
+ "transformers_version": "4.36.2",
34
+ "use_external_data_format": false
35
+ }