omarelsayeed
commited on
Commit
•
a8c9c86
1
Parent(s):
d3ccfd1
Upload model
Browse files- README.md +46 -0
- config.json +697 -0
- tf_model.h5 +3 -0
README.md
ADDED
@@ -0,0 +1,46 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
---
|
2 |
+
tags:
|
3 |
+
- generated_from_keras_callback
|
4 |
+
model-index:
|
5 |
+
- name: Classfier_V0
|
6 |
+
results: []
|
7 |
+
---
|
8 |
+
|
9 |
+
<!-- This model card has been generated automatically according to the information Keras had access to. You should
|
10 |
+
probably proofread and complete it, then remove this comment. -->
|
11 |
+
|
12 |
+
# Classfier_V0
|
13 |
+
|
14 |
+
This model was trained from scratch on an unknown dataset.
|
15 |
+
It achieves the following results on the evaluation set:
|
16 |
+
|
17 |
+
|
18 |
+
## Model description
|
19 |
+
|
20 |
+
More information needed
|
21 |
+
|
22 |
+
## Intended uses & limitations
|
23 |
+
|
24 |
+
More information needed
|
25 |
+
|
26 |
+
## Training and evaluation data
|
27 |
+
|
28 |
+
More information needed
|
29 |
+
|
30 |
+
## Training procedure
|
31 |
+
|
32 |
+
### Training hyperparameters
|
33 |
+
|
34 |
+
The following hyperparameters were used during training:
|
35 |
+
- optimizer: None
|
36 |
+
- training_precision: float32
|
37 |
+
|
38 |
+
### Training results
|
39 |
+
|
40 |
+
|
41 |
+
|
42 |
+
### Framework versions
|
43 |
+
|
44 |
+
- Transformers 4.33.2
|
45 |
+
- TensorFlow 2.13.0
|
46 |
+
- Tokenizers 0.13.3
|
config.json
ADDED
@@ -0,0 +1,697 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"_name_or_path": "/content/Classifier",
|
3 |
+
"architectures": [
|
4 |
+
"BertModel"
|
5 |
+
],
|
6 |
+
"attention_probs_dropout_prob": 0.1,
|
7 |
+
"classifier_dropout": null,
|
8 |
+
"hidden_act": "gelu",
|
9 |
+
"hidden_dropout_prob": 0.1,
|
10 |
+
"hidden_size": 768,
|
11 |
+
"initializer_range": 0.02,
|
12 |
+
"input_layers": [
|
13 |
+
[
|
14 |
+
"agent_txt_inputids",
|
15 |
+
0,
|
16 |
+
0
|
17 |
+
],
|
18 |
+
[
|
19 |
+
"agent_txt_mask",
|
20 |
+
0,
|
21 |
+
0
|
22 |
+
]
|
23 |
+
],
|
24 |
+
"intermediate_size": 3072,
|
25 |
+
"layer_norm_eps": 1e-12,
|
26 |
+
"layers": [
|
27 |
+
{
|
28 |
+
"class_name": "InputLayer",
|
29 |
+
"config": {
|
30 |
+
"batch_input_shape": [
|
31 |
+
null,
|
32 |
+
512
|
33 |
+
],
|
34 |
+
"dtype": "int32",
|
35 |
+
"name": "agent_txt_inputids",
|
36 |
+
"ragged": false,
|
37 |
+
"sparse": false
|
38 |
+
},
|
39 |
+
"inbound_nodes": [],
|
40 |
+
"name": "agent_txt_inputids"
|
41 |
+
},
|
42 |
+
{
|
43 |
+
"class_name": "InputLayer",
|
44 |
+
"config": {
|
45 |
+
"batch_input_shape": [
|
46 |
+
null,
|
47 |
+
512
|
48 |
+
],
|
49 |
+
"dtype": "int32",
|
50 |
+
"name": "agent_txt_mask",
|
51 |
+
"ragged": false,
|
52 |
+
"sparse": false
|
53 |
+
},
|
54 |
+
"inbound_nodes": [],
|
55 |
+
"name": "agent_txt_mask"
|
56 |
+
},
|
57 |
+
{
|
58 |
+
"class_name": "Custom>TFBertMainLayer",
|
59 |
+
"config": {
|
60 |
+
"config": {
|
61 |
+
"_name_or_path": "/content/drive/MyDrive/ARA-BERT-TF",
|
62 |
+
"add_cross_attention": false,
|
63 |
+
"architectures": [
|
64 |
+
"BertForMaskedLM"
|
65 |
+
],
|
66 |
+
"attention_probs_dropout_prob": 0.1,
|
67 |
+
"bad_words_ids": null,
|
68 |
+
"begin_suppress_tokens": null,
|
69 |
+
"bos_token_id": null,
|
70 |
+
"chunk_size_feed_forward": 0,
|
71 |
+
"classifier_dropout": null,
|
72 |
+
"cross_attention_hidden_size": null,
|
73 |
+
"decoder_start_token_id": null,
|
74 |
+
"diversity_penalty": 0.0,
|
75 |
+
"do_sample": false,
|
76 |
+
"early_stopping": false,
|
77 |
+
"encoder_no_repeat_ngram_size": 0,
|
78 |
+
"eos_token_id": null,
|
79 |
+
"exponential_decay_length_penalty": null,
|
80 |
+
"finetuning_task": null,
|
81 |
+
"forced_bos_token_id": null,
|
82 |
+
"forced_eos_token_id": null,
|
83 |
+
"hidden_act": "gelu",
|
84 |
+
"hidden_dropout_prob": 0.1,
|
85 |
+
"hidden_size": 768,
|
86 |
+
"id2label": {
|
87 |
+
"0": "LABEL_0",
|
88 |
+
"1": "LABEL_1"
|
89 |
+
},
|
90 |
+
"initializer_range": 0.02,
|
91 |
+
"intermediate_size": 3072,
|
92 |
+
"is_decoder": false,
|
93 |
+
"is_encoder_decoder": false,
|
94 |
+
"label2id": {
|
95 |
+
"LABEL_0": 0,
|
96 |
+
"LABEL_1": 1
|
97 |
+
},
|
98 |
+
"layer_norm_eps": 1e-12,
|
99 |
+
"length_penalty": 1.0,
|
100 |
+
"max_length": 20,
|
101 |
+
"max_position_embeddings": 512,
|
102 |
+
"min_length": 0,
|
103 |
+
"model_type": "bert",
|
104 |
+
"no_repeat_ngram_size": 0,
|
105 |
+
"num_attention_heads": 12,
|
106 |
+
"num_beam_groups": 1,
|
107 |
+
"num_beams": 1,
|
108 |
+
"num_hidden_layers": 12,
|
109 |
+
"num_return_sequences": 1,
|
110 |
+
"output_attentions": true,
|
111 |
+
"output_hidden_states": true,
|
112 |
+
"output_scores": false,
|
113 |
+
"pad_token_id": 0,
|
114 |
+
"position_embedding_type": "absolute",
|
115 |
+
"prefix": null,
|
116 |
+
"problem_type": null,
|
117 |
+
"pruned_heads": {},
|
118 |
+
"remove_invalid_values": false,
|
119 |
+
"repetition_penalty": 1.0,
|
120 |
+
"return_dict": true,
|
121 |
+
"return_dict_in_generate": false,
|
122 |
+
"sep_token_id": null,
|
123 |
+
"suppress_tokens": null,
|
124 |
+
"task_specific_params": null,
|
125 |
+
"temperature": 1.0,
|
126 |
+
"tf_legacy_loss": false,
|
127 |
+
"tie_encoder_decoder": false,
|
128 |
+
"tie_word_embeddings": true,
|
129 |
+
"tokenizer_class": null,
|
130 |
+
"top_k": 50,
|
131 |
+
"top_p": 1.0,
|
132 |
+
"torch_dtype": null,
|
133 |
+
"torchscript": false,
|
134 |
+
"transformers_version": "4.33.2",
|
135 |
+
"type_vocab_size": 2,
|
136 |
+
"typical_p": 1.0,
|
137 |
+
"use_bfloat16": false,
|
138 |
+
"use_cache": true,
|
139 |
+
"vocab_size": 64000
|
140 |
+
},
|
141 |
+
"dtype": "float32",
|
142 |
+
"name": "bert",
|
143 |
+
"trainable": true
|
144 |
+
},
|
145 |
+
"inbound_nodes": [
|
146 |
+
[
|
147 |
+
[
|
148 |
+
"agent_txt_inputids",
|
149 |
+
0,
|
150 |
+
0,
|
151 |
+
{
|
152 |
+
"attention_mask": [
|
153 |
+
"agent_txt_mask",
|
154 |
+
0,
|
155 |
+
0
|
156 |
+
]
|
157 |
+
}
|
158 |
+
]
|
159 |
+
]
|
160 |
+
],
|
161 |
+
"name": "bert"
|
162 |
+
},
|
163 |
+
{
|
164 |
+
"class_name": "Sequential",
|
165 |
+
"config": {
|
166 |
+
"layers": [
|
167 |
+
{
|
168 |
+
"class_name": "InputLayer",
|
169 |
+
"config": {
|
170 |
+
"batch_input_shape": [
|
171 |
+
null,
|
172 |
+
512,
|
173 |
+
768
|
174 |
+
],
|
175 |
+
"dtype": "float32",
|
176 |
+
"name": "encoder_input",
|
177 |
+
"ragged": false,
|
178 |
+
"sparse": false
|
179 |
+
}
|
180 |
+
},
|
181 |
+
{
|
182 |
+
"class_name": "Bidirectional",
|
183 |
+
"config": {
|
184 |
+
"dtype": "float32",
|
185 |
+
"layer": {
|
186 |
+
"class_name": "GRU",
|
187 |
+
"config": {
|
188 |
+
"activation": "tanh",
|
189 |
+
"activity_regularizer": null,
|
190 |
+
"bias_constraint": null,
|
191 |
+
"bias_initializer": {
|
192 |
+
"class_name": "Zeros",
|
193 |
+
"config": {},
|
194 |
+
"module": "keras.initializers",
|
195 |
+
"registered_name": null
|
196 |
+
},
|
197 |
+
"bias_regularizer": null,
|
198 |
+
"dropout": 0.0,
|
199 |
+
"dtype": "float32",
|
200 |
+
"go_backwards": false,
|
201 |
+
"implementation": 2,
|
202 |
+
"kernel_constraint": null,
|
203 |
+
"kernel_initializer": {
|
204 |
+
"class_name": "GlorotUniform",
|
205 |
+
"config": {
|
206 |
+
"seed": null
|
207 |
+
},
|
208 |
+
"module": "keras.initializers",
|
209 |
+
"registered_name": null
|
210 |
+
},
|
211 |
+
"kernel_regularizer": null,
|
212 |
+
"name": "gru_6",
|
213 |
+
"recurrent_activation": "sigmoid",
|
214 |
+
"recurrent_constraint": null,
|
215 |
+
"recurrent_dropout": 0.0,
|
216 |
+
"recurrent_initializer": {
|
217 |
+
"class_name": "Orthogonal",
|
218 |
+
"config": {
|
219 |
+
"gain": 1.0,
|
220 |
+
"seed": null
|
221 |
+
},
|
222 |
+
"module": "keras.initializers",
|
223 |
+
"registered_name": null
|
224 |
+
},
|
225 |
+
"recurrent_regularizer": null,
|
226 |
+
"reset_after": true,
|
227 |
+
"return_sequences": true,
|
228 |
+
"return_state": false,
|
229 |
+
"stateful": false,
|
230 |
+
"time_major": false,
|
231 |
+
"trainable": true,
|
232 |
+
"units": 64,
|
233 |
+
"unroll": false,
|
234 |
+
"use_bias": true
|
235 |
+
},
|
236 |
+
"module": "keras.layers",
|
237 |
+
"registered_name": null
|
238 |
+
},
|
239 |
+
"merge_mode": "concat",
|
240 |
+
"name": "encoder",
|
241 |
+
"trainable": true
|
242 |
+
}
|
243 |
+
},
|
244 |
+
{
|
245 |
+
"class_name": "GlobalMaxPooling1D",
|
246 |
+
"config": {
|
247 |
+
"data_format": "channels_last",
|
248 |
+
"dtype": "float32",
|
249 |
+
"keepdims": false,
|
250 |
+
"name": "global_max_pooling1d_6",
|
251 |
+
"trainable": true
|
252 |
+
}
|
253 |
+
},
|
254 |
+
{
|
255 |
+
"class_name": "Dense",
|
256 |
+
"config": {
|
257 |
+
"activation": "linear",
|
258 |
+
"activity_regularizer": null,
|
259 |
+
"bias_constraint": null,
|
260 |
+
"bias_initializer": {
|
261 |
+
"class_name": "Zeros",
|
262 |
+
"config": {},
|
263 |
+
"module": "keras.initializers",
|
264 |
+
"registered_name": null
|
265 |
+
},
|
266 |
+
"bias_regularizer": null,
|
267 |
+
"dtype": "float32",
|
268 |
+
"kernel_constraint": null,
|
269 |
+
"kernel_initializer": {
|
270 |
+
"class_name": "HeNormal",
|
271 |
+
"config": {
|
272 |
+
"seed": null
|
273 |
+
},
|
274 |
+
"module": "keras.initializers",
|
275 |
+
"registered_name": null
|
276 |
+
},
|
277 |
+
"kernel_regularizer": null,
|
278 |
+
"name": "dense_6",
|
279 |
+
"trainable": true,
|
280 |
+
"units": 512,
|
281 |
+
"use_bias": true
|
282 |
+
}
|
283 |
+
},
|
284 |
+
{
|
285 |
+
"class_name": "BatchNormalization",
|
286 |
+
"config": {
|
287 |
+
"axis": [
|
288 |
+
1
|
289 |
+
],
|
290 |
+
"beta_constraint": null,
|
291 |
+
"beta_initializer": {
|
292 |
+
"class_name": "Zeros",
|
293 |
+
"config": {},
|
294 |
+
"module": "keras.initializers",
|
295 |
+
"registered_name": null
|
296 |
+
},
|
297 |
+
"beta_regularizer": null,
|
298 |
+
"center": true,
|
299 |
+
"dtype": "float32",
|
300 |
+
"epsilon": 0.001,
|
301 |
+
"gamma_constraint": null,
|
302 |
+
"gamma_initializer": {
|
303 |
+
"class_name": "Ones",
|
304 |
+
"config": {},
|
305 |
+
"module": "keras.initializers",
|
306 |
+
"registered_name": null
|
307 |
+
},
|
308 |
+
"gamma_regularizer": null,
|
309 |
+
"momentum": 0.99,
|
310 |
+
"moving_mean_initializer": {
|
311 |
+
"class_name": "Zeros",
|
312 |
+
"config": {},
|
313 |
+
"module": "keras.initializers",
|
314 |
+
"registered_name": null
|
315 |
+
},
|
316 |
+
"moving_variance_initializer": {
|
317 |
+
"class_name": "Ones",
|
318 |
+
"config": {},
|
319 |
+
"module": "keras.initializers",
|
320 |
+
"registered_name": null
|
321 |
+
},
|
322 |
+
"name": "batch_norm",
|
323 |
+
"scale": true,
|
324 |
+
"trainable": true
|
325 |
+
}
|
326 |
+
},
|
327 |
+
{
|
328 |
+
"class_name": "Activation",
|
329 |
+
"config": {
|
330 |
+
"activation": "elu",
|
331 |
+
"dtype": "float32",
|
332 |
+
"name": "activation_18",
|
333 |
+
"trainable": true
|
334 |
+
}
|
335 |
+
},
|
336 |
+
{
|
337 |
+
"class_name": "Dropout",
|
338 |
+
"config": {
|
339 |
+
"dtype": "float32",
|
340 |
+
"name": "dropout",
|
341 |
+
"noise_shape": null,
|
342 |
+
"rate": 0.2,
|
343 |
+
"seed": null,
|
344 |
+
"trainable": true
|
345 |
+
}
|
346 |
+
}
|
347 |
+
],
|
348 |
+
"name": "sequential_6"
|
349 |
+
},
|
350 |
+
"inbound_nodes": [
|
351 |
+
[
|
352 |
+
[
|
353 |
+
"bert",
|
354 |
+
0,
|
355 |
+
23,
|
356 |
+
{}
|
357 |
+
]
|
358 |
+
]
|
359 |
+
],
|
360 |
+
"name": "sequential_6"
|
361 |
+
},
|
362 |
+
{
|
363 |
+
"class_name": "Dense",
|
364 |
+
"config": {
|
365 |
+
"activation": "linear",
|
366 |
+
"activity_regularizer": null,
|
367 |
+
"bias_constraint": null,
|
368 |
+
"bias_initializer": {
|
369 |
+
"class_name": "Zeros",
|
370 |
+
"config": {},
|
371 |
+
"module": "keras.initializers",
|
372 |
+
"registered_name": null
|
373 |
+
},
|
374 |
+
"bias_regularizer": null,
|
375 |
+
"dtype": "float32",
|
376 |
+
"kernel_constraint": null,
|
377 |
+
"kernel_initializer": {
|
378 |
+
"class_name": "HeNormal",
|
379 |
+
"config": {
|
380 |
+
"seed": null
|
381 |
+
},
|
382 |
+
"module": "keras.initializers",
|
383 |
+
"registered_name": null
|
384 |
+
},
|
385 |
+
"kernel_regularizer": null,
|
386 |
+
"name": "fully_connected_1",
|
387 |
+
"trainable": true,
|
388 |
+
"units": 128,
|
389 |
+
"use_bias": true
|
390 |
+
},
|
391 |
+
"inbound_nodes": [
|
392 |
+
[
|
393 |
+
[
|
394 |
+
"sequential_6",
|
395 |
+
1,
|
396 |
+
0,
|
397 |
+
{}
|
398 |
+
]
|
399 |
+
]
|
400 |
+
],
|
401 |
+
"name": "fully_connected_1"
|
402 |
+
},
|
403 |
+
{
|
404 |
+
"class_name": "BatchNormalization",
|
405 |
+
"config": {
|
406 |
+
"axis": [
|
407 |
+
1
|
408 |
+
],
|
409 |
+
"beta_constraint": null,
|
410 |
+
"beta_initializer": {
|
411 |
+
"class_name": "Zeros",
|
412 |
+
"config": {},
|
413 |
+
"module": "keras.initializers",
|
414 |
+
"registered_name": null
|
415 |
+
},
|
416 |
+
"beta_regularizer": null,
|
417 |
+
"center": true,
|
418 |
+
"dtype": "float32",
|
419 |
+
"epsilon": 0.001,
|
420 |
+
"gamma_constraint": null,
|
421 |
+
"gamma_initializer": {
|
422 |
+
"class_name": "Ones",
|
423 |
+
"config": {},
|
424 |
+
"module": "keras.initializers",
|
425 |
+
"registered_name": null
|
426 |
+
},
|
427 |
+
"gamma_regularizer": null,
|
428 |
+
"momentum": 0.99,
|
429 |
+
"moving_mean_initializer": {
|
430 |
+
"class_name": "Zeros",
|
431 |
+
"config": {},
|
432 |
+
"module": "keras.initializers",
|
433 |
+
"registered_name": null
|
434 |
+
},
|
435 |
+
"moving_variance_initializer": {
|
436 |
+
"class_name": "Ones",
|
437 |
+
"config": {},
|
438 |
+
"module": "keras.initializers",
|
439 |
+
"registered_name": null
|
440 |
+
},
|
441 |
+
"name": "batch_norm_1",
|
442 |
+
"scale": true,
|
443 |
+
"trainable": true
|
444 |
+
},
|
445 |
+
"inbound_nodes": [
|
446 |
+
[
|
447 |
+
[
|
448 |
+
"fully_connected_1",
|
449 |
+
0,
|
450 |
+
0,
|
451 |
+
{}
|
452 |
+
]
|
453 |
+
]
|
454 |
+
],
|
455 |
+
"name": "batch_norm_1"
|
456 |
+
},
|
457 |
+
{
|
458 |
+
"class_name": "Activation",
|
459 |
+
"config": {
|
460 |
+
"activation": "elu",
|
461 |
+
"dtype": "float32",
|
462 |
+
"name": "activation_19",
|
463 |
+
"trainable": true
|
464 |
+
},
|
465 |
+
"inbound_nodes": [
|
466 |
+
[
|
467 |
+
[
|
468 |
+
"batch_norm_1",
|
469 |
+
0,
|
470 |
+
0,
|
471 |
+
{}
|
472 |
+
]
|
473 |
+
]
|
474 |
+
],
|
475 |
+
"name": "activation_19"
|
476 |
+
},
|
477 |
+
{
|
478 |
+
"class_name": "Dropout",
|
479 |
+
"config": {
|
480 |
+
"dtype": "float32",
|
481 |
+
"name": "dropout_1",
|
482 |
+
"noise_shape": null,
|
483 |
+
"rate": 0.2,
|
484 |
+
"seed": null,
|
485 |
+
"trainable": true
|
486 |
+
},
|
487 |
+
"inbound_nodes": [
|
488 |
+
[
|
489 |
+
[
|
490 |
+
"activation_19",
|
491 |
+
0,
|
492 |
+
0,
|
493 |
+
{}
|
494 |
+
]
|
495 |
+
]
|
496 |
+
],
|
497 |
+
"name": "dropout_1"
|
498 |
+
},
|
499 |
+
{
|
500 |
+
"class_name": "Dense",
|
501 |
+
"config": {
|
502 |
+
"activation": "linear",
|
503 |
+
"activity_regularizer": null,
|
504 |
+
"bias_constraint": null,
|
505 |
+
"bias_initializer": {
|
506 |
+
"class_name": "Zeros",
|
507 |
+
"config": {},
|
508 |
+
"module": "keras.initializers",
|
509 |
+
"registered_name": null
|
510 |
+
},
|
511 |
+
"bias_regularizer": null,
|
512 |
+
"dtype": "float32",
|
513 |
+
"kernel_constraint": null,
|
514 |
+
"kernel_initializer": {
|
515 |
+
"class_name": "HeNormal",
|
516 |
+
"config": {
|
517 |
+
"seed": null
|
518 |
+
},
|
519 |
+
"module": "keras.initializers",
|
520 |
+
"registered_name": null
|
521 |
+
},
|
522 |
+
"kernel_regularizer": null,
|
523 |
+
"name": "fully_connected_1.1",
|
524 |
+
"trainable": true,
|
525 |
+
"units": 64,
|
526 |
+
"use_bias": true
|
527 |
+
},
|
528 |
+
"inbound_nodes": [
|
529 |
+
[
|
530 |
+
[
|
531 |
+
"dropout_1",
|
532 |
+
0,
|
533 |
+
0,
|
534 |
+
{}
|
535 |
+
]
|
536 |
+
]
|
537 |
+
],
|
538 |
+
"name": "fully_connected_1.1"
|
539 |
+
},
|
540 |
+
{
|
541 |
+
"class_name": "BatchNormalization",
|
542 |
+
"config": {
|
543 |
+
"axis": [
|
544 |
+
1
|
545 |
+
],
|
546 |
+
"beta_constraint": null,
|
547 |
+
"beta_initializer": {
|
548 |
+
"class_name": "Zeros",
|
549 |
+
"config": {},
|
550 |
+
"module": "keras.initializers",
|
551 |
+
"registered_name": null
|
552 |
+
},
|
553 |
+
"beta_regularizer": null,
|
554 |
+
"center": true,
|
555 |
+
"dtype": "float32",
|
556 |
+
"epsilon": 0.001,
|
557 |
+
"gamma_constraint": null,
|
558 |
+
"gamma_initializer": {
|
559 |
+
"class_name": "Ones",
|
560 |
+
"config": {},
|
561 |
+
"module": "keras.initializers",
|
562 |
+
"registered_name": null
|
563 |
+
},
|
564 |
+
"gamma_regularizer": null,
|
565 |
+
"momentum": 0.99,
|
566 |
+
"moving_mean_initializer": {
|
567 |
+
"class_name": "Zeros",
|
568 |
+
"config": {},
|
569 |
+
"module": "keras.initializers",
|
570 |
+
"registered_name": null
|
571 |
+
},
|
572 |
+
"moving_variance_initializer": {
|
573 |
+
"class_name": "Ones",
|
574 |
+
"config": {},
|
575 |
+
"module": "keras.initializers",
|
576 |
+
"registered_name": null
|
577 |
+
},
|
578 |
+
"name": "batch_norm_1.1",
|
579 |
+
"scale": true,
|
580 |
+
"trainable": true
|
581 |
+
},
|
582 |
+
"inbound_nodes": [
|
583 |
+
[
|
584 |
+
[
|
585 |
+
"fully_connected_1.1",
|
586 |
+
0,
|
587 |
+
0,
|
588 |
+
{}
|
589 |
+
]
|
590 |
+
]
|
591 |
+
],
|
592 |
+
"name": "batch_norm_1.1"
|
593 |
+
},
|
594 |
+
{
|
595 |
+
"class_name": "Activation",
|
596 |
+
"config": {
|
597 |
+
"activation": "elu",
|
598 |
+
"dtype": "float32",
|
599 |
+
"name": "activation_20",
|
600 |
+
"trainable": true
|
601 |
+
},
|
602 |
+
"inbound_nodes": [
|
603 |
+
[
|
604 |
+
[
|
605 |
+
"batch_norm_1.1",
|
606 |
+
0,
|
607 |
+
0,
|
608 |
+
{}
|
609 |
+
]
|
610 |
+
]
|
611 |
+
],
|
612 |
+
"name": "activation_20"
|
613 |
+
},
|
614 |
+
{
|
615 |
+
"class_name": "Dropout",
|
616 |
+
"config": {
|
617 |
+
"dtype": "float32",
|
618 |
+
"name": "dropout_1.1",
|
619 |
+
"noise_shape": null,
|
620 |
+
"rate": 0.2,
|
621 |
+
"seed": null,
|
622 |
+
"trainable": true
|
623 |
+
},
|
624 |
+
"inbound_nodes": [
|
625 |
+
[
|
626 |
+
[
|
627 |
+
"activation_20",
|
628 |
+
0,
|
629 |
+
0,
|
630 |
+
{}
|
631 |
+
]
|
632 |
+
]
|
633 |
+
],
|
634 |
+
"name": "dropout_1.1"
|
635 |
+
},
|
636 |
+
{
|
637 |
+
"class_name": "Dense",
|
638 |
+
"config": {
|
639 |
+
"activation": "softmax",
|
640 |
+
"activity_regularizer": null,
|
641 |
+
"bias_constraint": null,
|
642 |
+
"bias_initializer": {
|
643 |
+
"class_name": "Zeros",
|
644 |
+
"config": {},
|
645 |
+
"module": "keras.initializers",
|
646 |
+
"registered_name": null
|
647 |
+
},
|
648 |
+
"bias_regularizer": null,
|
649 |
+
"dtype": "float32",
|
650 |
+
"kernel_constraint": null,
|
651 |
+
"kernel_initializer": {
|
652 |
+
"class_name": "GlorotUniform",
|
653 |
+
"config": {
|
654 |
+
"seed": null
|
655 |
+
},
|
656 |
+
"module": "keras.initializers",
|
657 |
+
"registered_name": null
|
658 |
+
},
|
659 |
+
"kernel_regularizer": null,
|
660 |
+
"name": "output_layer",
|
661 |
+
"trainable": true,
|
662 |
+
"units": 49,
|
663 |
+
"use_bias": true
|
664 |
+
},
|
665 |
+
"inbound_nodes": [
|
666 |
+
[
|
667 |
+
[
|
668 |
+
"dropout_1.1",
|
669 |
+
0,
|
670 |
+
0,
|
671 |
+
{}
|
672 |
+
]
|
673 |
+
]
|
674 |
+
],
|
675 |
+
"name": "output_layer"
|
676 |
+
}
|
677 |
+
],
|
678 |
+
"max_position_embeddings": 512,
|
679 |
+
"model_type": "bert",
|
680 |
+
"name": "model_6",
|
681 |
+
"num_attention_heads": 12,
|
682 |
+
"num_hidden_layers": 12,
|
683 |
+
"output_layers": [
|
684 |
+
[
|
685 |
+
"output_layer",
|
686 |
+
0,
|
687 |
+
0
|
688 |
+
]
|
689 |
+
],
|
690 |
+
"pad_token_id": 0,
|
691 |
+
"position_embedding_type": "absolute",
|
692 |
+
"trainable": true,
|
693 |
+
"transformers_version": "4.33.2",
|
694 |
+
"type_vocab_size": 2,
|
695 |
+
"use_cache": true,
|
696 |
+
"vocab_size": 30522
|
697 |
+
}
|
tf_model.h5
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:00c10fa75af8d1b6cc2dd78782f9f5176e1139dd12bffd6b3fb62fe8b241d293
|
3 |
+
size 438196200
|