NimaKL commited on
Commit
9dd702b
·
1 Parent(s): 9f28cf3

Upload 2 files

Browse files
Files changed (2) hide show
  1. config.json +259 -0
  2. tf_model.h5 +3 -0
config.json ADDED
@@ -0,0 +1,259 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_name_or_path": "TC32_HF/model",
3
+ "architectures": [
4
+ "BertModel"
5
+ ],
6
+ "attention_probs_dropout_prob": 0.1,
7
+ "backend": "tensorflow",
8
+ "class_name": "Functional",
9
+ "classifier_dropout": null,
10
+ "config": {
11
+ "input_layers": [
12
+ [
13
+ "input_ids",
14
+ 0,
15
+ 0
16
+ ],
17
+ [
18
+ "attention_mask",
19
+ 0,
20
+ 0
21
+ ]
22
+ ],
23
+ "layers": [
24
+ {
25
+ "class_name": "InputLayer",
26
+ "config": {
27
+ "batch_input_shape": [
28
+ null,
29
+ 256
30
+ ],
31
+ "dtype": "int32",
32
+ "name": "input_ids",
33
+ "ragged": false,
34
+ "sparse": false
35
+ },
36
+ "inbound_nodes": [],
37
+ "name": "input_ids"
38
+ },
39
+ {
40
+ "class_name": "InputLayer",
41
+ "config": {
42
+ "batch_input_shape": [
43
+ null,
44
+ 256
45
+ ],
46
+ "dtype": "int32",
47
+ "name": "attention_mask",
48
+ "ragged": false,
49
+ "sparse": false
50
+ },
51
+ "inbound_nodes": [],
52
+ "name": "attention_mask"
53
+ },
54
+ {
55
+ "class_name": "Custom>TFBertMainLayer",
56
+ "config": {
57
+ "config": {
58
+ "_name_or_path": "dbmdz/bert-base-turkish-128k-cased",
59
+ "add_cross_attention": false,
60
+ "architectures": null,
61
+ "attention_probs_dropout_prob": 0.1,
62
+ "bad_words_ids": null,
63
+ "begin_suppress_tokens": null,
64
+ "bos_token_id": null,
65
+ "chunk_size_feed_forward": 0,
66
+ "classifier_dropout": null,
67
+ "cross_attention_hidden_size": null,
68
+ "decoder_start_token_id": null,
69
+ "diversity_penalty": 0.0,
70
+ "do_sample": false,
71
+ "early_stopping": false,
72
+ "encoder_no_repeat_ngram_size": 0,
73
+ "eos_token_id": null,
74
+ "exponential_decay_length_penalty": null,
75
+ "finetuning_task": null,
76
+ "forced_bos_token_id": null,
77
+ "forced_eos_token_id": null,
78
+ "hidden_act": "gelu",
79
+ "hidden_dropout_prob": 0.1,
80
+ "hidden_size": 768,
81
+ "id2label": {
82
+ "0": "LABEL_0",
83
+ "1": "LABEL_1"
84
+ },
85
+ "initializer_range": 0.02,
86
+ "intermediate_size": 3072,
87
+ "is_decoder": false,
88
+ "is_encoder_decoder": false,
89
+ "label2id": {
90
+ "LABEL_0": 0,
91
+ "LABEL_1": 1
92
+ },
93
+ "layer_norm_eps": 1e-12,
94
+ "length_penalty": 1.0,
95
+ "max_length": 20,
96
+ "max_position_embeddings": 512,
97
+ "min_length": 0,
98
+ "model_type": "bert",
99
+ "no_repeat_ngram_size": 0,
100
+ "num_attention_heads": 12,
101
+ "num_beam_groups": 1,
102
+ "num_beams": 1,
103
+ "num_hidden_layers": 12,
104
+ "num_return_sequences": 1,
105
+ "output_attentions": false,
106
+ "output_hidden_states": false,
107
+ "output_scores": false,
108
+ "pad_token_id": 0,
109
+ "position_embedding_type": "absolute",
110
+ "prefix": null,
111
+ "problem_type": null,
112
+ "pruned_heads": {},
113
+ "remove_invalid_values": false,
114
+ "repetition_penalty": 1.0,
115
+ "return_dict": true,
116
+ "return_dict_in_generate": false,
117
+ "sep_token_id": null,
118
+ "suppress_tokens": null,
119
+ "task_specific_params": null,
120
+ "temperature": 1.0,
121
+ "tf_legacy_loss": false,
122
+ "tie_encoder_decoder": false,
123
+ "tie_word_embeddings": true,
124
+ "tokenizer_class": null,
125
+ "top_k": 50,
126
+ "top_p": 1.0,
127
+ "torch_dtype": null,
128
+ "torchscript": false,
129
+ "transformers_version": "4.23.1",
130
+ "type_vocab_size": 2,
131
+ "typical_p": 1.0,
132
+ "use_bfloat16": false,
133
+ "use_cache": true,
134
+ "vocab_size": 128000
135
+ },
136
+ "dtype": "float32",
137
+ "name": "bert",
138
+ "trainable": true
139
+ },
140
+ "inbound_nodes": [
141
+ [
142
+ [
143
+ "input_ids",
144
+ 0,
145
+ 0,
146
+ {
147
+ "attention_mask": [
148
+ "attention_mask",
149
+ 0,
150
+ 0
151
+ ]
152
+ }
153
+ ]
154
+ ]
155
+ ],
156
+ "name": "bert"
157
+ },
158
+ {
159
+ "class_name": "Dense",
160
+ "config": {
161
+ "activation": "relu",
162
+ "activity_regularizer": null,
163
+ "bias_constraint": null,
164
+ "bias_initializer": {
165
+ "class_name": "Zeros",
166
+ "config": {}
167
+ },
168
+ "bias_regularizer": null,
169
+ "dtype": "float32",
170
+ "kernel_constraint": null,
171
+ "kernel_initializer": {
172
+ "class_name": "GlorotUniform",
173
+ "config": {
174
+ "seed": null
175
+ }
176
+ },
177
+ "kernel_regularizer": null,
178
+ "name": "intermediate_layer",
179
+ "trainable": true,
180
+ "units": 512,
181
+ "use_bias": true
182
+ },
183
+ "inbound_nodes": [
184
+ [
185
+ [
186
+ "bert",
187
+ 0,
188
+ 1,
189
+ {}
190
+ ]
191
+ ]
192
+ ],
193
+ "name": "intermediate_layer"
194
+ },
195
+ {
196
+ "class_name": "Dense",
197
+ "config": {
198
+ "activation": "softmax",
199
+ "activity_regularizer": null,
200
+ "bias_constraint": null,
201
+ "bias_initializer": {
202
+ "class_name": "Zeros",
203
+ "config": {}
204
+ },
205
+ "bias_regularizer": null,
206
+ "dtype": "float32",
207
+ "kernel_constraint": null,
208
+ "kernel_initializer": {
209
+ "class_name": "GlorotUniform",
210
+ "config": {
211
+ "seed": null
212
+ }
213
+ },
214
+ "kernel_regularizer": null,
215
+ "name": "output_layer",
216
+ "trainable": true,
217
+ "units": 32,
218
+ "use_bias": true
219
+ },
220
+ "inbound_nodes": [
221
+ [
222
+ [
223
+ "intermediate_layer",
224
+ 0,
225
+ 0,
226
+ {}
227
+ ]
228
+ ]
229
+ ],
230
+ "name": "output_layer"
231
+ }
232
+ ],
233
+ "name": "model",
234
+ "output_layers": [
235
+ [
236
+ "output_layer",
237
+ 0,
238
+ 0
239
+ ]
240
+ ]
241
+ },
242
+ "hidden_act": "gelu",
243
+ "hidden_dropout_prob": 0.1,
244
+ "hidden_size": 768,
245
+ "initializer_range": 0.02,
246
+ "intermediate_size": 3072,
247
+ "keras_version": "2.10.0",
248
+ "layer_norm_eps": 1e-12,
249
+ "max_position_embeddings": 512,
250
+ "model_type": "bert",
251
+ "num_attention_heads": 12,
252
+ "num_hidden_layers": 12,
253
+ "pad_token_id": 0,
254
+ "position_embedding_type": "absolute",
255
+ "transformers_version": "4.23.1",
256
+ "type_vocab_size": 2,
257
+ "use_cache": true,
258
+ "vocab_size": 30522
259
+ }
tf_model.h5 ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:a1de378e7a10dbcd29540e4addf503fe46af60c50c3f1ad82b59ea9b3de52fee
3
+ size 438196200