LightChen2333
commited on
Commit
•
8a382ff
1
Parent(s):
6eb29e3
Upload 3 files
Browse files- config.json +19 -9
config.json
CHANGED
@@ -194,7 +194,7 @@
|
|
194 |
"mlp": [
|
195 |
{
|
196 |
"_model_target_": "torch.nn.Linear",
|
197 |
-
"in_features":
|
198 |
"out_features": 256
|
199 |
},
|
200 |
{
|
@@ -204,14 +204,14 @@
|
|
204 |
{
|
205 |
"_model_target_": "torch.nn.Linear",
|
206 |
"in_features": 256,
|
207 |
-
"out_features":
|
208 |
}
|
209 |
],
|
210 |
"mode": "intent",
|
211 |
"multi_threshold": 0.5,
|
212 |
"return_sentence_level": true,
|
213 |
"use_intent": true,
|
214 |
-
"use_multi":
|
215 |
"weight": 0.3
|
216 |
},
|
217 |
"interaction": {
|
@@ -219,12 +219,14 @@
|
|
219 |
"alpha": 0.2,
|
220 |
"dropout_rate": 0.4,
|
221 |
"hidden_dim": 128,
|
222 |
-
"input_dim":
|
223 |
"intent_embedding_dim": 128,
|
|
|
224 |
"num_heads": 4,
|
225 |
"num_layers": 2,
|
226 |
-
"output_dim":
|
227 |
-
"row_normalized": true
|
|
|
228 |
},
|
229 |
"slot_classifier": {
|
230 |
"_model_target_": "model.decoder.classifier.AutoregressiveLSTMClassifier",
|
@@ -232,9 +234,9 @@
|
|
232 |
"dropout_rate": 0.4,
|
233 |
"embedding_dim": 128,
|
234 |
"force_ratio": 0.9,
|
235 |
-
"hidden_dim":
|
236 |
"ignore_index": -100,
|
237 |
-
"input_dim":
|
238 |
"layer_num": 1,
|
239 |
"mode": "slot",
|
240 |
"multi_threshold": 0.5,
|
@@ -254,7 +256,8 @@
|
|
254 |
},
|
255 |
"embedding": {
|
256 |
"dropout_rate": 0.4,
|
257 |
-
"embedding_dim": 128
|
|
|
258 |
},
|
259 |
"encoder_name": "self-attention-lstm",
|
260 |
"lstm": {
|
@@ -272,6 +275,13 @@
|
|
272 |
}
|
273 |
},
|
274 |
"return_dict": false,
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
275 |
"tokenizer_class": "OpenSLUv1",
|
276 |
"torch_dtype": "float32",
|
277 |
"transformers_version": "4.25.1",
|
|
|
194 |
"mlp": [
|
195 |
{
|
196 |
"_model_target_": "torch.nn.Linear",
|
197 |
+
"in_features": 384,
|
198 |
"out_features": 256
|
199 |
},
|
200 |
{
|
|
|
204 |
{
|
205 |
"_model_target_": "torch.nn.Linear",
|
206 |
"in_features": 256,
|
207 |
+
"out_features": 7
|
208 |
}
|
209 |
],
|
210 |
"mode": "intent",
|
211 |
"multi_threshold": 0.5,
|
212 |
"return_sentence_level": true,
|
213 |
"use_intent": true,
|
214 |
+
"use_multi": true,
|
215 |
"weight": 0.3
|
216 |
},
|
217 |
"interaction": {
|
|
|
219 |
"alpha": 0.2,
|
220 |
"dropout_rate": 0.4,
|
221 |
"hidden_dim": 128,
|
222 |
+
"input_dim": 384,
|
223 |
"intent_embedding_dim": 128,
|
224 |
+
"intent_label_num": 7,
|
225 |
"num_heads": 4,
|
226 |
"num_layers": 2,
|
227 |
+
"output_dim": 128,
|
228 |
+
"row_normalized": true,
|
229 |
+
"slot_label_num": 72
|
230 |
},
|
231 |
"slot_classifier": {
|
232 |
"_model_target_": "model.decoder.classifier.AutoregressiveLSTMClassifier",
|
|
|
234 |
"dropout_rate": 0.4,
|
235 |
"embedding_dim": 128,
|
236 |
"force_ratio": 0.9,
|
237 |
+
"hidden_dim": 128,
|
238 |
"ignore_index": -100,
|
239 |
+
"input_dim": 384,
|
240 |
"layer_num": 1,
|
241 |
"mode": "slot",
|
242 |
"multi_threshold": 0.5,
|
|
|
256 |
},
|
257 |
"embedding": {
|
258 |
"dropout_rate": 0.4,
|
259 |
+
"embedding_dim": 128,
|
260 |
+
"vocab_size": 11758
|
261 |
},
|
262 |
"encoder_name": "self-attention-lstm",
|
263 |
"lstm": {
|
|
|
275 |
}
|
276 |
},
|
277 |
"return_dict": false,
|
278 |
+
"tokenizer": {
|
279 |
+
"_align_mode_": "fast",
|
280 |
+
"_padding_side_": "right",
|
281 |
+
"_tokenizer_name_": "word_tokenizer",
|
282 |
+
"add_special_tokens": false,
|
283 |
+
"max_length": 512
|
284 |
+
},
|
285 |
"tokenizer_class": "OpenSLUv1",
|
286 |
"torch_dtype": "float32",
|
287 |
"transformers_version": "4.25.1",
|