Add new SentenceTransformer model with an openvino backend
#1
by
vumichien
- opened
- 1_Pooling/config.json +9 -9
- README.md +416 -416
- config.json +25 -24
- config_sentence_transformers.json +9 -9
- modules.json +13 -13
- openvino/openvino_model.bin +3 -0
- openvino/openvino_model.xml +0 -0
- sentence_bert_config.json +3 -3
- special_tokens_map.json +37 -37
- tokenizer_config.json +64 -64
- vocab.txt +0 -0
1_Pooling/config.json
CHANGED
@@ -1,10 +1,10 @@
|
|
1 |
-
{
|
2 |
-
"word_embedding_dimension": 768,
|
3 |
-
"pooling_mode_cls_token": true,
|
4 |
-
"pooling_mode_mean_tokens": false,
|
5 |
-
"pooling_mode_max_tokens": false,
|
6 |
-
"pooling_mode_mean_sqrt_len_tokens": false,
|
7 |
-
"pooling_mode_weightedmean_tokens": false,
|
8 |
-
"pooling_mode_lasttoken": false,
|
9 |
-
"include_prompt": true
|
10 |
}
|
|
|
1 |
+
{
|
2 |
+
"word_embedding_dimension": 768,
|
3 |
+
"pooling_mode_cls_token": true,
|
4 |
+
"pooling_mode_mean_tokens": false,
|
5 |
+
"pooling_mode_max_tokens": false,
|
6 |
+
"pooling_mode_mean_sqrt_len_tokens": false,
|
7 |
+
"pooling_mode_weightedmean_tokens": false,
|
8 |
+
"pooling_mode_lasttoken": false,
|
9 |
+
"include_prompt": true
|
10 |
}
|
README.md
CHANGED
@@ -1,417 +1,417 @@
|
|
1 |
-
---
|
2 |
-
tags:
|
3 |
-
- sentence-transformers
|
4 |
-
- sentence-similarity
|
5 |
-
- feature-extraction
|
6 |
-
- generated_from_trainer
|
7 |
-
- dataset_size:182343
|
8 |
-
- loss:CategoricalContrastiveLoss
|
9 |
-
widget:
|
10 |
-
- source_sentence: 科目:コンクリート。名称:浮き床コンクリート。
|
11 |
-
sentences:
|
12 |
-
- 科目:コンクリート。名称:普通コンクリート。摘要:JIS A5308 FC21 S18粗骨材20。備考:刊-コン 2118EXP.J面コン。
|
13 |
-
- 科目:コンクリート。名称:充填コンクリート(EXP_J内)。
|
14 |
-
- 科目:コンクリート。名称:免震上部コンクリート打設手間。
|
15 |
-
- source_sentence: 科目:コンクリート。名称:基礎部コンクリート打設手間。
|
16 |
-
sentences:
|
17 |
-
- 科目:コンクリート。名称:基礎部高流動コンクリート。摘要:FC36N/mm2 スランプフロー55~65高性能AE減水剤。備考:代価表 0059。
|
18 |
-
- 科目:コンクリート。名称:基礎部コンクリート。摘要:FC24N/mm2 スランプ15。備考:代価表 0040。
|
19 |
-
- 科目:コンクリート。名称:コンクリート(個別)。摘要:F0=18N/mm2 S=18 徳島1。備考:B1-111111 H2906BD 個別嵩上げコンクリート。
|
20 |
-
- source_sentence: 科目:コンクリート。名称:浮き床コンクリート。
|
21 |
-
sentences:
|
22 |
-
- 科目:コンクリート。名称:コンクリートポンプ圧送。摘要:100m3/回以上基本料金別途加算。備考:B0-434226 No.1 市場地上部コン(5F)。
|
23 |
-
- 科目:コンクリート。名称:普通コンクリート。摘要:JIS A5308 FC48 フロー60粗骨材20高性能AE減水剤。備考:刊-コン 4860K免震装置下部コン。
|
24 |
-
- 科目:コンクリート。名称:防水押えコンクリート。
|
25 |
-
- source_sentence: 科目:コンクリート。名称:基礎コンクリート。
|
26 |
-
sentences:
|
27 |
-
- 科目:コンクリート。名称:均しコンクリート。
|
28 |
-
- 科目:コンクリート。名称:普通コンクリート。摘要:FC=24 S18粗骨材地上部。備考:代価表 0060。
|
29 |
-
- 科目:コンクリート。名称:基礎部高流動コンクリート。摘要:FC36N/mm2 スランプフロー55~65高性能AE減水剤。備考:代価表 0059。
|
30 |
-
- source_sentence: 科目:コンクリート。名称:コンクリート打設手間。
|
31 |
-
sentences:
|
32 |
-
- 科目:コンクリート。名称:免震上部コンクリート打設手間。
|
33 |
-
- 科目:コンクリート。名称:普通コンクリート。摘要:JIS A5308 FC=18 S15粗骨材20。備考:B0-114112 H22.11 協議防水保護コンクリート。
|
34 |
-
- 科目:コンクリート。名称:コンクリート打設手間。
|
35 |
-
pipeline_tag: sentence-similarity
|
36 |
-
library_name: sentence-transformers
|
37 |
-
---
|
38 |
-
|
39 |
-
# SentenceTransformer
|
40 |
-
|
41 |
-
This is a [sentence-transformers](https://www.SBERT.net) model trained. It maps sentences & paragraphs to a 768-dimensional dense vector space and can be used for semantic textual similarity, semantic search, paraphrase mining, text classification, clustering, and more.
|
42 |
-
|
43 |
-
## Model Details
|
44 |
-
|
45 |
-
### Model Description
|
46 |
-
- **Model Type:** Sentence Transformer
|
47 |
-
<!-- - **Base model:** [Unknown](https://huggingface.co/unknown) -->
|
48 |
-
- **Maximum Sequence Length:** 512 tokens
|
49 |
-
- **Output Dimensionality:** 768 dimensions
|
50 |
-
- **Similarity Function:** Cosine Similarity
|
51 |
-
<!-- - **Training Dataset:** Unknown -->
|
52 |
-
<!-- - **Language:** Unknown -->
|
53 |
-
<!-- - **License:** Unknown -->
|
54 |
-
|
55 |
-
### Model Sources
|
56 |
-
|
57 |
-
- **Documentation:** [Sentence Transformers Documentation](https://sbert.net)
|
58 |
-
- **Repository:** [Sentence Transformers on GitHub](https://github.com/UKPLab/sentence-transformers)
|
59 |
-
- **Hugging Face:** [Sentence Transformers on Hugging Face](https://huggingface.co/models?library=sentence-transformers)
|
60 |
-
|
61 |
-
### Full Model Architecture
|
62 |
-
|
63 |
-
```
|
64 |
-
SentenceTransformer(
|
65 |
-
(0): Transformer({'max_seq_length': 512, 'do_lower_case': False}) with Transformer model: BertModel
|
66 |
-
(1): Pooling({'word_embedding_dimension': 768, 'pooling_mode_cls_token': True, 'pooling_mode_mean_tokens': False, 'pooling_mode_max_tokens': False, 'pooling_mode_mean_sqrt_len_tokens': False, 'pooling_mode_weightedmean_tokens': False, 'pooling_mode_lasttoken': False, 'include_prompt': True})
|
67 |
-
)
|
68 |
-
```
|
69 |
-
|
70 |
-
## Usage
|
71 |
-
|
72 |
-
### Direct Usage (Sentence Transformers)
|
73 |
-
|
74 |
-
First install the Sentence Transformers library:
|
75 |
-
|
76 |
-
```bash
|
77 |
-
pip install -U sentence-transformers
|
78 |
-
```
|
79 |
-
|
80 |
-
Then you can load this model and run inference.
|
81 |
-
```python
|
82 |
-
from sentence_transformers import SentenceTransformer
|
83 |
-
|
84 |
-
# Download from the 🤗 Hub
|
85 |
-
model = SentenceTransformer("Detomo/cl-nagoya-sup-simcse-ja-nss-v1_0_7_8")
|
86 |
-
# Run inference
|
87 |
-
sentences = [
|
88 |
-
'科目:コンクリート。名称:コンクリート打設手間。',
|
89 |
-
'科目:コンクリート。名称:免震上部コンクリート打設手間。',
|
90 |
-
'科目:コンクリート。名称:普通コンクリート。摘要:JIS A5308 FC=18 S15粗骨材20。備考:B0-114112 H22.11 協議防水保護コンクリート。',
|
91 |
-
]
|
92 |
-
embeddings = model.encode(sentences)
|
93 |
-
print(embeddings.shape)
|
94 |
-
# [3, 768]
|
95 |
-
|
96 |
-
# Get the similarity scores for the embeddings
|
97 |
-
similarities = model.similarity(embeddings, embeddings)
|
98 |
-
print(similarities.shape)
|
99 |
-
# [3, 3]
|
100 |
-
```
|
101 |
-
|
102 |
-
<!--
|
103 |
-
### Direct Usage (Transformers)
|
104 |
-
|
105 |
-
<details><summary>Click to see the direct usage in Transformers</summary>
|
106 |
-
|
107 |
-
</details>
|
108 |
-
-->
|
109 |
-
|
110 |
-
<!--
|
111 |
-
### Downstream Usage (Sentence Transformers)
|
112 |
-
|
113 |
-
You can finetune this model on your own dataset.
|
114 |
-
|
115 |
-
<details><summary>Click to expand</summary>
|
116 |
-
|
117 |
-
</details>
|
118 |
-
-->
|
119 |
-
|
120 |
-
<!--
|
121 |
-
### Out-of-Scope Use
|
122 |
-
|
123 |
-
*List how the model may foreseeably be misused and address what users ought not to do with the model.*
|
124 |
-
-->
|
125 |
-
|
126 |
-
<!--
|
127 |
-
## Bias, Risks and Limitations
|
128 |
-
|
129 |
-
*What are the known or foreseeable issues stemming from this model? You could also flag here known failure cases or weaknesses of the model.*
|
130 |
-
-->
|
131 |
-
|
132 |
-
<!--
|
133 |
-
### Recommendations
|
134 |
-
|
135 |
-
*What are recommendations with respect to the foreseeable issues? For example, filtering explicit content.*
|
136 |
-
-->
|
137 |
-
|
138 |
-
## Training Details
|
139 |
-
|
140 |
-
### Training Dataset
|
141 |
-
|
142 |
-
#### Unnamed Dataset
|
143 |
-
|
144 |
-
* Size: 182,343 training samples
|
145 |
-
* Columns: <code>sentence1</code>, <code>sentence2</code>, and <code>label</code>
|
146 |
-
* Approximate statistics based on the first 1000 samples:
|
147 |
-
| | sentence1 | sentence2 | label |
|
148 |
-
|:--------|:-----------------------------------------------------------------------------------|:----------------------------------------------------------------------------------|:------------------------------------------------------------------|
|
149 |
-
| type | string | string | int |
|
150 |
-
| details | <ul><li>min: 11 tokens</li><li>mean: 13.32 tokens</li><li>max: 19 tokens</li></ul> | <ul><li>min: 11 tokens</li><li>mean: 34.8 tokens</li><li>max: 72 tokens</li></ul> | <ul><li>0: ~68.50%</li><li>1: ~4.50%</li><li>2: ~27.00%</li></ul> |
|
151 |
-
* Samples:
|
152 |
-
| sentence1 | sentence2 | label |
|
153 |
-
|:-----------------------------------------|:-------------------------------------------------------------------------------------------------|:---------------|
|
154 |
-
| <code>科目:コンクリート。名称:コンクリートポンプ圧送。</code> | <code>科目:コンクリート。名称:ポンプ圧送。</code> | <code>1</code> |
|
155 |
-
| <code>科目:コンクリート。名称:コンクリートポンプ圧送。</code> | <code>科目:コンクリート。名称:コンクリートポンプ圧送。摘要:100m3/回以上基本料金別途加算。備考:B0-434226 No.1 市場免震層下部コン。</code> | <code>2</code> |
|
156 |
-
| <code>科目:コンクリート。名称:コンクリートポンプ圧送。</code> | <code>科目:コンクリート。名称:コンクリートポンプ圧送。摘要:100m3/回以上基本料金別途加算。備考:B0-434226 No.1 市場基礎部マスコン。</code> | <code>2</code> |
|
157 |
-
* Loss: <code>sentence_transformer_lib.categorical_constrastive_loss.CategoricalContrastiveLoss</code>
|
158 |
-
|
159 |
-
### Training Hyperparameters
|
160 |
-
#### Non-Default Hyperparameters
|
161 |
-
|
162 |
-
- `per_device_train_batch_size`: 256
|
163 |
-
- `per_device_eval_batch_size`: 256
|
164 |
-
- `learning_rate`: 1e-05
|
165 |
-
- `weight_decay`: 0.01
|
166 |
-
- `num_train_epochs`: 10
|
167 |
-
- `warmup_ratio`: 0.2
|
168 |
-
- `fp16`: True
|
169 |
-
|
170 |
-
#### All Hyperparameters
|
171 |
-
<details><summary>Click to expand</summary>
|
172 |
-
|
173 |
-
- `overwrite_output_dir`: False
|
174 |
-
- `do_predict`: False
|
175 |
-
- `eval_strategy`: no
|
176 |
-
- `prediction_loss_only`: True
|
177 |
-
- `per_device_train_batch_size`: 256
|
178 |
-
- `per_device_eval_batch_size`: 256
|
179 |
-
- `per_gpu_train_batch_size`: None
|
180 |
-
- `per_gpu_eval_batch_size`: None
|
181 |
-
- `gradient_accumulation_steps`: 1
|
182 |
-
- `eval_accumulation_steps`: None
|
183 |
-
- `torch_empty_cache_steps`: None
|
184 |
-
- `learning_rate`: 1e-05
|
185 |
-
- `weight_decay`: 0.01
|
186 |
-
- `adam_beta1`: 0.9
|
187 |
-
- `adam_beta2`: 0.999
|
188 |
-
- `adam_epsilon`: 1e-08
|
189 |
-
- `max_grad_norm`: 1.0
|
190 |
-
- `num_train_epochs`: 10
|
191 |
-
- `max_steps`: -1
|
192 |
-
- `lr_scheduler_type`: linear
|
193 |
-
- `lr_scheduler_kwargs`: {}
|
194 |
-
- `warmup_ratio`: 0.2
|
195 |
-
- `warmup_steps`: 0
|
196 |
-
- `log_level`: passive
|
197 |
-
- `log_level_replica`: warning
|
198 |
-
- `log_on_each_node`: True
|
199 |
-
- `logging_nan_inf_filter`: True
|
200 |
-
- `save_safetensors`: True
|
201 |
-
- `save_on_each_node`: False
|
202 |
-
- `save_only_model`: False
|
203 |
-
- `restore_callback_states_from_checkpoint`: False
|
204 |
-
- `no_cuda`: False
|
205 |
-
- `use_cpu`: False
|
206 |
-
- `use_mps_device`: False
|
207 |
-
- `seed`: 42
|
208 |
-
- `data_seed`: None
|
209 |
-
- `jit_mode_eval`: False
|
210 |
-
- `use_ipex`: False
|
211 |
-
- `bf16`: False
|
212 |
-
- `fp16`: True
|
213 |
-
- `fp16_opt_level`: O1
|
214 |
-
- `half_precision_backend`: auto
|
215 |
-
- `bf16_full_eval`: False
|
216 |
-
- `fp16_full_eval`: False
|
217 |
-
- `tf32`: None
|
218 |
-
- `local_rank`: 0
|
219 |
-
- `ddp_backend`: None
|
220 |
-
- `tpu_num_cores`: None
|
221 |
-
- `tpu_metrics_debug`: False
|
222 |
-
- `debug`: []
|
223 |
-
- `dataloader_drop_last`: False
|
224 |
-
- `dataloader_num_workers`: 0
|
225 |
-
- `dataloader_prefetch_factor`: None
|
226 |
-
- `past_index`: -1
|
227 |
-
- `disable_tqdm`: False
|
228 |
-
- `remove_unused_columns`: True
|
229 |
-
- `label_names`: None
|
230 |
-
- `load_best_model_at_end`: False
|
231 |
-
- `ignore_data_skip`: False
|
232 |
-
- `fsdp`: []
|
233 |
-
- `fsdp_min_num_params`: 0
|
234 |
-
- `fsdp_config`: {'min_num_params': 0, 'xla': False, 'xla_fsdp_v2': False, 'xla_fsdp_grad_ckpt': False}
|
235 |
-
- `tp_size`: 0
|
236 |
-
- `fsdp_transformer_layer_cls_to_wrap`: None
|
237 |
-
- `accelerator_config`: {'split_batches': False, 'dispatch_batches': None, 'even_batches': True, 'use_seedable_sampler': True, 'non_blocking': False, 'gradient_accumulation_kwargs': None}
|
238 |
-
- `deepspeed`: None
|
239 |
-
- `label_smoothing_factor`: 0.0
|
240 |
-
- `optim`: adamw_torch
|
241 |
-
- `optim_args`: None
|
242 |
-
- `adafactor`: False
|
243 |
-
- `group_by_length`: False
|
244 |
-
- `length_column_name`: length
|
245 |
-
- `ddp_find_unused_parameters`: None
|
246 |
-
- `ddp_bucket_cap_mb`: None
|
247 |
-
- `ddp_broadcast_buffers`: False
|
248 |
-
- `dataloader_pin_memory`: True
|
249 |
-
- `dataloader_persistent_workers`: False
|
250 |
-
- `skip_memory_metrics`: True
|
251 |
-
- `use_legacy_prediction_loop`: False
|
252 |
-
- `push_to_hub`: False
|
253 |
-
- `resume_from_checkpoint`: None
|
254 |
-
- `hub_model_id`: None
|
255 |
-
- `hub_strategy`: every_save
|
256 |
-
- `hub_private_repo`: None
|
257 |
-
- `hub_always_push`: False
|
258 |
-
- `gradient_checkpointing`: False
|
259 |
-
- `gradient_checkpointing_kwargs`: None
|
260 |
-
- `include_inputs_for_metrics`: False
|
261 |
-
- `include_for_metrics`: []
|
262 |
-
- `eval_do_concat_batches`: True
|
263 |
-
- `fp16_backend`: auto
|
264 |
-
- `push_to_hub_model_id`: None
|
265 |
-
- `push_to_hub_organization`: None
|
266 |
-
- `mp_parameters`:
|
267 |
-
- `auto_find_batch_size`: False
|
268 |
-
- `full_determinism`: False
|
269 |
-
- `torchdynamo`: None
|
270 |
-
- `ray_scope`: last
|
271 |
-
- `ddp_timeout`: 1800
|
272 |
-
- `torch_compile`: False
|
273 |
-
- `torch_compile_backend`: None
|
274 |
-
- `torch_compile_mode`: None
|
275 |
-
- `include_tokens_per_second`: False
|
276 |
-
- `include_num_input_tokens_seen`: False
|
277 |
-
- `neftune_noise_alpha`: None
|
278 |
-
- `optim_target_modules`: None
|
279 |
-
- `batch_eval_metrics`: False
|
280 |
-
- `eval_on_start`: False
|
281 |
-
- `use_liger_kernel`: False
|
282 |
-
- `eval_use_gather_object`: False
|
283 |
-
- `average_tokens_across_devices`: False
|
284 |
-
- `prompts`: None
|
285 |
-
- `batch_sampler`: batch_sampler
|
286 |
-
- `multi_dataset_batch_sampler`: proportional
|
287 |
-
|
288 |
-
</details>
|
289 |
-
|
290 |
-
### Training Logs
|
291 |
-
| Epoch | Step | Training Loss |
|
292 |
-
|:------:|:----:|:-------------:|
|
293 |
-
| 0.0701 | 50 | 0.2825 |
|
294 |
-
| 0.1403 | 100 | 0.1467 |
|
295 |
-
| 0.2104 | 150 | 0.0947 |
|
296 |
-
| 0.2805 | 200 | 0.0839 |
|
297 |
-
| 0.3506 | 250 | 0.0769 |
|
298 |
-
| 0.4208 | 300 | 0.0684 |
|
299 |
-
| 0.4909 | 350 | 0.0625 |
|
300 |
-
| 0.5610 | 400 | 0.0582 |
|
301 |
-
| 0.6311 | 450 | 0.0579 |
|
302 |
-
| 0.7013 | 500 | 0.0514 |
|
303 |
-
| 0.7714 | 550 | 0.0514 |
|
304 |
-
| 0.8415 | 600 | 0.0448 |
|
305 |
-
| 0.9116 | 650 | 0.0436 |
|
306 |
-
| 0.9818 | 700 | 0.0422 |
|
307 |
-
| 1.0519 | 750 | 0.0371 |
|
308 |
-
| 1.1220 | 800 | 0.0377 |
|
309 |
-
| 1.1921 | 850 | 0.0353 |
|
310 |
-
| 1.2623 | 900 | 0.0354 |
|
311 |
-
| 1.3324 | 950 | 0.0325 |
|
312 |
-
| 1.4025 | 1000 | 0.0328 |
|
313 |
-
| 1.4727 | 1050 | 0.0302 |
|
314 |
-
| 1.5428 | 1100 | 0.0259 |
|
315 |
-
| 1.6129 | 1150 | 0.0267 |
|
316 |
-
| 1.6830 | 1200 | 0.0274 |
|
317 |
-
| 1.7532 | 1250 | 0.0262 |
|
318 |
-
| 1.8233 | 1300 | 0.0234 |
|
319 |
-
| 1.8934 | 1350 | 0.0244 |
|
320 |
-
| 1.9635 | 1400 | 0.0238 |
|
321 |
-
| 2.0337 | 1450 | 0.02 |
|
322 |
-
| 2.1038 | 1500 | 0.0187 |
|
323 |
-
| 2.1739 | 1550 | 0.0185 |
|
324 |
-
| 2.2440 | 1600 | 0.0178 |
|
325 |
-
| 2.3142 | 1650 | 0.016 |
|
326 |
-
| 2.3843 | 1700 | 0.0169 |
|
327 |
-
| 2.4544 | 1750 | 0.0171 |
|
328 |
-
| 2.5245 | 1800 | 0.0146 |
|
329 |
-
| 2.5947 | 1850 | 0.0145 |
|
330 |
-
| 2.6648 | 1900 | 0.0146 |
|
331 |
-
| 2.7349 | 1950 | 0.0139 |
|
332 |
-
| 2.8050 | 2000 | 0.0119 |
|
333 |
-
| 2.8752 | 2050 | 0.0131 |
|
334 |
-
| 2.9453 | 2100 | 0.0124 |
|
335 |
-
| 3.0154 | 2150 | 0.011 |
|
336 |
-
| 3.0856 | 2200 | 0.0109 |
|
337 |
-
| 3.1557 | 2250 | 0.0103 |
|
338 |
-
| 3.2258 | 2300 | 0.0102 |
|
339 |
-
| 3.2959 | 2350 | 0.0089 |
|
340 |
-
| 3.3661 | 2400 | 0.0083 |
|
341 |
-
| 3.4362 | 2450 | 0.0095 |
|
342 |
-
| 3.5063 | 2500 | 0.0085 |
|
343 |
-
| 3.5764 | 2550 | 0.009 |
|
344 |
-
| 3.6466 | 2600 | 0.0083 |
|
345 |
-
| 3.7167 | 2650 | 0.0093 |
|
346 |
-
| 3.7868 | 2700 | 0.0084 |
|
347 |
-
| 3.8569 | 2750 | 0.0084 |
|
348 |
-
| 3.9271 | 2800 | 0.0088 |
|
349 |
-
| 3.9972 | 2850 | 0.0086 |
|
350 |
-
| 4.0673 | 2900 | 0.0057 |
|
351 |
-
| 4.1374 | 2950 | 0.0078 |
|
352 |
-
| 4.2076 | 3000 | 0.0062 |
|
353 |
-
| 4.2777 | 3050 | 0.0066 |
|
354 |
-
| 4.3478 | 3100 | 0.006 |
|
355 |
-
| 4.4180 | 3150 | 0.0078 |
|
356 |
-
| 4.4881 | 3200 | 0.0056 |
|
357 |
-
| 4.5582 | 3250 | 0.0064 |
|
358 |
-
| 4.6283 | 3300 | 0.0063 |
|
359 |
-
| 4.6985 | 3350 | 0.0058 |
|
360 |
-
| 4.7686 | 3400 | 0.005 |
|
361 |
-
| 4.8387 | 3450 | 0.0057 |
|
362 |
-
| 4.9088 | 3500 | 0.0059 |
|
363 |
-
| 4.9790 | 3550 | 0.0063 |
|
364 |
-
| 5.0491 | 3600 | 0.0046 |
|
365 |
-
| 5.1192 | 3650 | 0.0041 |
|
366 |
-
| 5.1893 | 3700 | 0.005 |
|
367 |
-
| 5.2595 | 3750 | 0.0043 |
|
368 |
-
| 5.3296 | 3800 | 0.0046 |
|
369 |
-
| 5.3997 | 3850 | 0.0041 |
|
370 |
-
| 5.4698 | 3900 | 0.006 |
|
371 |
-
| 5.5400 | 3950 | 0.0052 |
|
372 |
-
| 5.6101 | 4000 | 0.0043 |
|
373 |
-
|
374 |
-
|
375 |
-
### Framework Versions
|
376 |
-
- Python: 3.11.12
|
377 |
-
- Sentence Transformers: 4.1.0
|
378 |
-
- Transformers: 4.51.3
|
379 |
-
- PyTorch: 2.6.0+cu124
|
380 |
-
- Accelerate: 1.6.0
|
381 |
-
- Datasets: 2.14.4
|
382 |
-
- Tokenizers: 0.21.1
|
383 |
-
|
384 |
-
## Citation
|
385 |
-
|
386 |
-
### BibTeX
|
387 |
-
|
388 |
-
#### Sentence Transformers
|
389 |
-
```bibtex
|
390 |
-
@inproceedings{reimers-2019-sentence-bert,
|
391 |
-
title = "Sentence-BERT: Sentence Embeddings using Siamese BERT-Networks",
|
392 |
-
author = "Reimers, Nils and Gurevych, Iryna",
|
393 |
-
booktitle = "Proceedings of the 2019 Conference on Empirical Methods in Natural Language Processing",
|
394 |
-
month = "11",
|
395 |
-
year = "2019",
|
396 |
-
publisher = "Association for Computational Linguistics",
|
397 |
-
url = "https://arxiv.org/abs/1908.10084",
|
398 |
-
}
|
399 |
-
```
|
400 |
-
|
401 |
-
<!--
|
402 |
-
## Glossary
|
403 |
-
|
404 |
-
*Clearly define terms in order to be accessible across audiences.*
|
405 |
-
-->
|
406 |
-
|
407 |
-
<!--
|
408 |
-
## Model Card Authors
|
409 |
-
|
410 |
-
*Lists the people who create the model card, providing recognition and accountability for the detailed work that goes into its construction.*
|
411 |
-
-->
|
412 |
-
|
413 |
-
<!--
|
414 |
-
## Model Card Contact
|
415 |
-
|
416 |
-
*Provides a way for people who have updates to the Model Card, suggestions, or questions, to contact the Model Card authors.*
|
417 |
-->
|
|
|
1 |
+
---
|
2 |
+
tags:
|
3 |
+
- sentence-transformers
|
4 |
+
- sentence-similarity
|
5 |
+
- feature-extraction
|
6 |
+
- generated_from_trainer
|
7 |
+
- dataset_size:182343
|
8 |
+
- loss:CategoricalContrastiveLoss
|
9 |
+
widget:
|
10 |
+
- source_sentence: 科目:コンクリート。名称:浮き床コンクリート。
|
11 |
+
sentences:
|
12 |
+
- 科目:コンクリート。名称:普通コンクリート。摘要:JIS A5308 FC21 S18粗骨材20。備考:刊-コン 2118EXP.J面コン。
|
13 |
+
- 科目:コンクリート。名称:充填コンクリート(EXP_J内)。
|
14 |
+
- 科目:コンクリート。名称:免震上部コンクリート打設手間。
|
15 |
+
- source_sentence: 科目:コンクリート。名称:基礎部コンクリート打設手間。
|
16 |
+
sentences:
|
17 |
+
- 科目:コンクリート。名称:基礎部高流動コンクリート。摘要:FC36N/mm2 スランプフロー55~65高性能AE減水剤。備考:代価表 0059。
|
18 |
+
- 科目:コンクリート。名称:基礎部コンクリート。摘要:FC24N/mm2 スランプ15。備考:代価表 0040。
|
19 |
+
- 科目:コンクリート。名称:コンクリート(個別)。摘要:F0=18N/mm2 S=18 徳島1。備考:B1-111111 H2906BD 個別嵩上げコンクリート。
|
20 |
+
- source_sentence: 科目:コンクリート。名称:浮き床コンクリート。
|
21 |
+
sentences:
|
22 |
+
- 科目:コンクリート。名称:コンクリートポンプ圧送。摘要:100m3/回以上基本料金別途加算。備考:B0-434226 No.1 市場地上部コン(5F)。
|
23 |
+
- 科目:コンクリート。名称:普通コンクリート。摘要:JIS A5308 FC48 フロー60粗骨材20高性能AE減水剤。備考:刊-コン 4860K免震装置下部コン。
|
24 |
+
- 科目:コンクリート。名称:防水押えコンクリート。
|
25 |
+
- source_sentence: 科目:コンクリート。名称:基礎コンクリート。
|
26 |
+
sentences:
|
27 |
+
- 科目:コンクリート。名称:均しコンクリート。
|
28 |
+
- 科目:コンクリート。名称:普通コンクリート。摘要:FC=24 S18粗骨材地上部。備考:代価表 0060。
|
29 |
+
- 科目:コンクリート。名称:基礎部高流動コンクリート。摘要:FC36N/mm2 スランプフロー55~65高性能AE減水剤。備考:代価表 0059。
|
30 |
+
- source_sentence: 科目:コンクリート。名称:コンクリート打設手間。
|
31 |
+
sentences:
|
32 |
+
- 科目:コンクリート。名称:免震上部コンクリート打設手間。
|
33 |
+
- 科目:コンクリート。名称:普通コンクリート。摘要:JIS A5308 FC=18 S15粗骨材20。備考:B0-114112 H22.11 協議防水保護コンクリート。
|
34 |
+
- 科目:コンクリート。名称:コンクリート打設手間。
|
35 |
+
pipeline_tag: sentence-similarity
|
36 |
+
library_name: sentence-transformers
|
37 |
+
---
|
38 |
+
|
39 |
+
# SentenceTransformer
|
40 |
+
|
41 |
+
This is a [sentence-transformers](https://www.SBERT.net) model trained. It maps sentences & paragraphs to a 768-dimensional dense vector space and can be used for semantic textual similarity, semantic search, paraphrase mining, text classification, clustering, and more.
|
42 |
+
|
43 |
+
## Model Details
|
44 |
+
|
45 |
+
### Model Description
|
46 |
+
- **Model Type:** Sentence Transformer
|
47 |
+
<!-- - **Base model:** [Unknown](https://huggingface.co/unknown) -->
|
48 |
+
- **Maximum Sequence Length:** 512 tokens
|
49 |
+
- **Output Dimensionality:** 768 dimensions
|
50 |
+
- **Similarity Function:** Cosine Similarity
|
51 |
+
<!-- - **Training Dataset:** Unknown -->
|
52 |
+
<!-- - **Language:** Unknown -->
|
53 |
+
<!-- - **License:** Unknown -->
|
54 |
+
|
55 |
+
### Model Sources
|
56 |
+
|
57 |
+
- **Documentation:** [Sentence Transformers Documentation](https://sbert.net)
|
58 |
+
- **Repository:** [Sentence Transformers on GitHub](https://github.com/UKPLab/sentence-transformers)
|
59 |
+
- **Hugging Face:** [Sentence Transformers on Hugging Face](https://huggingface.co/models?library=sentence-transformers)
|
60 |
+
|
61 |
+
### Full Model Architecture
|
62 |
+
|
63 |
+
```
|
64 |
+
SentenceTransformer(
|
65 |
+
(0): Transformer({'max_seq_length': 512, 'do_lower_case': False}) with Transformer model: BertModel
|
66 |
+
(1): Pooling({'word_embedding_dimension': 768, 'pooling_mode_cls_token': True, 'pooling_mode_mean_tokens': False, 'pooling_mode_max_tokens': False, 'pooling_mode_mean_sqrt_len_tokens': False, 'pooling_mode_weightedmean_tokens': False, 'pooling_mode_lasttoken': False, 'include_prompt': True})
|
67 |
+
)
|
68 |
+
```
|
69 |
+
|
70 |
+
## Usage
|
71 |
+
|
72 |
+
### Direct Usage (Sentence Transformers)
|
73 |
+
|
74 |
+
First install the Sentence Transformers library:
|
75 |
+
|
76 |
+
```bash
|
77 |
+
pip install -U sentence-transformers
|
78 |
+
```
|
79 |
+
|
80 |
+
Then you can load this model and run inference.
|
81 |
+
```python
|
82 |
+
from sentence_transformers import SentenceTransformer
|
83 |
+
|
84 |
+
# Download from the 🤗 Hub
|
85 |
+
model = SentenceTransformer("Detomo/cl-nagoya-sup-simcse-ja-nss-v1_0_7_8")
|
86 |
+
# Run inference
|
87 |
+
sentences = [
|
88 |
+
'科目:コンクリート。名称:コンクリート打設手間。',
|
89 |
+
'科目:コンクリート。名称:免震上部コンクリート打設手間。',
|
90 |
+
'科目:コンクリート。名称:普通コンクリート。摘要:JIS A5308 FC=18 S15粗骨材20。備考:B0-114112 H22.11 協議防水保護コンクリート。',
|
91 |
+
]
|
92 |
+
embeddings = model.encode(sentences)
|
93 |
+
print(embeddings.shape)
|
94 |
+
# [3, 768]
|
95 |
+
|
96 |
+
# Get the similarity scores for the embeddings
|
97 |
+
similarities = model.similarity(embeddings, embeddings)
|
98 |
+
print(similarities.shape)
|
99 |
+
# [3, 3]
|
100 |
+
```
|
101 |
+
|
102 |
+
<!--
|
103 |
+
### Direct Usage (Transformers)
|
104 |
+
|
105 |
+
<details><summary>Click to see the direct usage in Transformers</summary>
|
106 |
+
|
107 |
+
</details>
|
108 |
+
-->
|
109 |
+
|
110 |
+
<!--
|
111 |
+
### Downstream Usage (Sentence Transformers)
|
112 |
+
|
113 |
+
You can finetune this model on your own dataset.
|
114 |
+
|
115 |
+
<details><summary>Click to expand</summary>
|
116 |
+
|
117 |
+
</details>
|
118 |
+
-->
|
119 |
+
|
120 |
+
<!--
|
121 |
+
### Out-of-Scope Use
|
122 |
+
|
123 |
+
*List how the model may foreseeably be misused and address what users ought not to do with the model.*
|
124 |
+
-->
|
125 |
+
|
126 |
+
<!--
|
127 |
+
## Bias, Risks and Limitations
|
128 |
+
|
129 |
+
*What are the known or foreseeable issues stemming from this model? You could also flag here known failure cases or weaknesses of the model.*
|
130 |
+
-->
|
131 |
+
|
132 |
+
<!--
|
133 |
+
### Recommendations
|
134 |
+
|
135 |
+
*What are recommendations with respect to the foreseeable issues? For example, filtering explicit content.*
|
136 |
+
-->
|
137 |
+
|
138 |
+
## Training Details
|
139 |
+
|
140 |
+
### Training Dataset
|
141 |
+
|
142 |
+
#### Unnamed Dataset
|
143 |
+
|
144 |
+
* Size: 182,343 training samples
|
145 |
+
* Columns: <code>sentence1</code>, <code>sentence2</code>, and <code>label</code>
|
146 |
+
* Approximate statistics based on the first 1000 samples:
|
147 |
+
| | sentence1 | sentence2 | label |
|
148 |
+
|:--------|:-----------------------------------------------------------------------------------|:----------------------------------------------------------------------------------|:------------------------------------------------------------------|
|
149 |
+
| type | string | string | int |
|
150 |
+
| details | <ul><li>min: 11 tokens</li><li>mean: 13.32 tokens</li><li>max: 19 tokens</li></ul> | <ul><li>min: 11 tokens</li><li>mean: 34.8 tokens</li><li>max: 72 tokens</li></ul> | <ul><li>0: ~68.50%</li><li>1: ~4.50%</li><li>2: ~27.00%</li></ul> |
|
151 |
+
* Samples:
|
152 |
+
| sentence1 | sentence2 | label |
|
153 |
+
|:-----------------------------------------|:-------------------------------------------------------------------------------------------------|:---------------|
|
154 |
+
| <code>科目:コンクリート。名称:コンクリートポンプ圧送。</code> | <code>科目:コンクリート。名称:ポンプ圧送。</code> | <code>1</code> |
|
155 |
+
| <code>科目:コンクリート。名称:コンクリートポンプ圧送。</code> | <code>科目:コンクリート。名称:コンクリートポンプ圧送。摘要:100m3/回以上基本料金別途加算。備考:B0-434226 No.1 市場免震層下部コン。</code> | <code>2</code> |
|
156 |
+
| <code>科目:コンクリート。名称:コンクリートポンプ圧送。</code> | <code>科目:コンクリート。名称:コンクリートポンプ圧送。摘要:100m3/回以上基本料金別途加算。備考:B0-434226 No.1 市場基礎部マスコン。</code> | <code>2</code> |
|
157 |
+
* Loss: <code>sentence_transformer_lib.categorical_constrastive_loss.CategoricalContrastiveLoss</code>
|
158 |
+
|
159 |
+
### Training Hyperparameters
|
160 |
+
#### Non-Default Hyperparameters
|
161 |
+
|
162 |
+
- `per_device_train_batch_size`: 256
|
163 |
+
- `per_device_eval_batch_size`: 256
|
164 |
+
- `learning_rate`: 1e-05
|
165 |
+
- `weight_decay`: 0.01
|
166 |
+
- `num_train_epochs`: 10
|
167 |
+
- `warmup_ratio`: 0.2
|
168 |
+
- `fp16`: True
|
169 |
+
|
170 |
+
#### All Hyperparameters
|
171 |
+
<details><summary>Click to expand</summary>
|
172 |
+
|
173 |
+
- `overwrite_output_dir`: False
|
174 |
+
- `do_predict`: False
|
175 |
+
- `eval_strategy`: no
|
176 |
+
- `prediction_loss_only`: True
|
177 |
+
- `per_device_train_batch_size`: 256
|
178 |
+
- `per_device_eval_batch_size`: 256
|
179 |
+
- `per_gpu_train_batch_size`: None
|
180 |
+
- `per_gpu_eval_batch_size`: None
|
181 |
+
- `gradient_accumulation_steps`: 1
|
182 |
+
- `eval_accumulation_steps`: None
|
183 |
+
- `torch_empty_cache_steps`: None
|
184 |
+
- `learning_rate`: 1e-05
|
185 |
+
- `weight_decay`: 0.01
|
186 |
+
- `adam_beta1`: 0.9
|
187 |
+
- `adam_beta2`: 0.999
|
188 |
+
- `adam_epsilon`: 1e-08
|
189 |
+
- `max_grad_norm`: 1.0
|
190 |
+
- `num_train_epochs`: 10
|
191 |
+
- `max_steps`: -1
|
192 |
+
- `lr_scheduler_type`: linear
|
193 |
+
- `lr_scheduler_kwargs`: {}
|
194 |
+
- `warmup_ratio`: 0.2
|
195 |
+
- `warmup_steps`: 0
|
196 |
+
- `log_level`: passive
|
197 |
+
- `log_level_replica`: warning
|
198 |
+
- `log_on_each_node`: True
|
199 |
+
- `logging_nan_inf_filter`: True
|
200 |
+
- `save_safetensors`: True
|
201 |
+
- `save_on_each_node`: False
|
202 |
+
- `save_only_model`: False
|
203 |
+
- `restore_callback_states_from_checkpoint`: False
|
204 |
+
- `no_cuda`: False
|
205 |
+
- `use_cpu`: False
|
206 |
+
- `use_mps_device`: False
|
207 |
+
- `seed`: 42
|
208 |
+
- `data_seed`: None
|
209 |
+
- `jit_mode_eval`: False
|
210 |
+
- `use_ipex`: False
|
211 |
+
- `bf16`: False
|
212 |
+
- `fp16`: True
|
213 |
+
- `fp16_opt_level`: O1
|
214 |
+
- `half_precision_backend`: auto
|
215 |
+
- `bf16_full_eval`: False
|
216 |
+
- `fp16_full_eval`: False
|
217 |
+
- `tf32`: None
|
218 |
+
- `local_rank`: 0
|
219 |
+
- `ddp_backend`: None
|
220 |
+
- `tpu_num_cores`: None
|
221 |
+
- `tpu_metrics_debug`: False
|
222 |
+
- `debug`: []
|
223 |
+
- `dataloader_drop_last`: False
|
224 |
+
- `dataloader_num_workers`: 0
|
225 |
+
- `dataloader_prefetch_factor`: None
|
226 |
+
- `past_index`: -1
|
227 |
+
- `disable_tqdm`: False
|
228 |
+
- `remove_unused_columns`: True
|
229 |
+
- `label_names`: None
|
230 |
+
- `load_best_model_at_end`: False
|
231 |
+
- `ignore_data_skip`: False
|
232 |
+
- `fsdp`: []
|
233 |
+
- `fsdp_min_num_params`: 0
|
234 |
+
- `fsdp_config`: {'min_num_params': 0, 'xla': False, 'xla_fsdp_v2': False, 'xla_fsdp_grad_ckpt': False}
|
235 |
+
- `tp_size`: 0
|
236 |
+
- `fsdp_transformer_layer_cls_to_wrap`: None
|
237 |
+
- `accelerator_config`: {'split_batches': False, 'dispatch_batches': None, 'even_batches': True, 'use_seedable_sampler': True, 'non_blocking': False, 'gradient_accumulation_kwargs': None}
|
238 |
+
- `deepspeed`: None
|
239 |
+
- `label_smoothing_factor`: 0.0
|
240 |
+
- `optim`: adamw_torch
|
241 |
+
- `optim_args`: None
|
242 |
+
- `adafactor`: False
|
243 |
+
- `group_by_length`: False
|
244 |
+
- `length_column_name`: length
|
245 |
+
- `ddp_find_unused_parameters`: None
|
246 |
+
- `ddp_bucket_cap_mb`: None
|
247 |
+
- `ddp_broadcast_buffers`: False
|
248 |
+
- `dataloader_pin_memory`: True
|
249 |
+
- `dataloader_persistent_workers`: False
|
250 |
+
- `skip_memory_metrics`: True
|
251 |
+
- `use_legacy_prediction_loop`: False
|
252 |
+
- `push_to_hub`: False
|
253 |
+
- `resume_from_checkpoint`: None
|
254 |
+
- `hub_model_id`: None
|
255 |
+
- `hub_strategy`: every_save
|
256 |
+
- `hub_private_repo`: None
|
257 |
+
- `hub_always_push`: False
|
258 |
+
- `gradient_checkpointing`: False
|
259 |
+
- `gradient_checkpointing_kwargs`: None
|
260 |
+
- `include_inputs_for_metrics`: False
|
261 |
+
- `include_for_metrics`: []
|
262 |
+
- `eval_do_concat_batches`: True
|
263 |
+
- `fp16_backend`: auto
|
264 |
+
- `push_to_hub_model_id`: None
|
265 |
+
- `push_to_hub_organization`: None
|
266 |
+
- `mp_parameters`:
|
267 |
+
- `auto_find_batch_size`: False
|
268 |
+
- `full_determinism`: False
|
269 |
+
- `torchdynamo`: None
|
270 |
+
- `ray_scope`: last
|
271 |
+
- `ddp_timeout`: 1800
|
272 |
+
- `torch_compile`: False
|
273 |
+
- `torch_compile_backend`: None
|
274 |
+
- `torch_compile_mode`: None
|
275 |
+
- `include_tokens_per_second`: False
|
276 |
+
- `include_num_input_tokens_seen`: False
|
277 |
+
- `neftune_noise_alpha`: None
|
278 |
+
- `optim_target_modules`: None
|
279 |
+
- `batch_eval_metrics`: False
|
280 |
+
- `eval_on_start`: False
|
281 |
+
- `use_liger_kernel`: False
|
282 |
+
- `eval_use_gather_object`: False
|
283 |
+
- `average_tokens_across_devices`: False
|
284 |
+
- `prompts`: None
|
285 |
+
- `batch_sampler`: batch_sampler
|
286 |
+
- `multi_dataset_batch_sampler`: proportional
|
287 |
+
|
288 |
+
</details>
|
289 |
+
|
290 |
+
### Training Logs
|
291 |
+
| Epoch | Step | Training Loss |
|
292 |
+
|:------:|:----:|:-------------:|
|
293 |
+
| 0.0701 | 50 | 0.2825 |
|
294 |
+
| 0.1403 | 100 | 0.1467 |
|
295 |
+
| 0.2104 | 150 | 0.0947 |
|
296 |
+
| 0.2805 | 200 | 0.0839 |
|
297 |
+
| 0.3506 | 250 | 0.0769 |
|
298 |
+
| 0.4208 | 300 | 0.0684 |
|
299 |
+
| 0.4909 | 350 | 0.0625 |
|
300 |
+
| 0.5610 | 400 | 0.0582 |
|
301 |
+
| 0.6311 | 450 | 0.0579 |
|
302 |
+
| 0.7013 | 500 | 0.0514 |
|
303 |
+
| 0.7714 | 550 | 0.0514 |
|
304 |
+
| 0.8415 | 600 | 0.0448 |
|
305 |
+
| 0.9116 | 650 | 0.0436 |
|
306 |
+
| 0.9818 | 700 | 0.0422 |
|
307 |
+
| 1.0519 | 750 | 0.0371 |
|
308 |
+
| 1.1220 | 800 | 0.0377 |
|
309 |
+
| 1.1921 | 850 | 0.0353 |
|
310 |
+
| 1.2623 | 900 | 0.0354 |
|
311 |
+
| 1.3324 | 950 | 0.0325 |
|
312 |
+
| 1.4025 | 1000 | 0.0328 |
|
313 |
+
| 1.4727 | 1050 | 0.0302 |
|
314 |
+
| 1.5428 | 1100 | 0.0259 |
|
315 |
+
| 1.6129 | 1150 | 0.0267 |
|
316 |
+
| 1.6830 | 1200 | 0.0274 |
|
317 |
+
| 1.7532 | 1250 | 0.0262 |
|
318 |
+
| 1.8233 | 1300 | 0.0234 |
|
319 |
+
| 1.8934 | 1350 | 0.0244 |
|
320 |
+
| 1.9635 | 1400 | 0.0238 |
|
321 |
+
| 2.0337 | 1450 | 0.02 |
|
322 |
+
| 2.1038 | 1500 | 0.0187 |
|
323 |
+
| 2.1739 | 1550 | 0.0185 |
|
324 |
+
| 2.2440 | 1600 | 0.0178 |
|
325 |
+
| 2.3142 | 1650 | 0.016 |
|
326 |
+
| 2.3843 | 1700 | 0.0169 |
|
327 |
+
| 2.4544 | 1750 | 0.0171 |
|
328 |
+
| 2.5245 | 1800 | 0.0146 |
|
329 |
+
| 2.5947 | 1850 | 0.0145 |
|
330 |
+
| 2.6648 | 1900 | 0.0146 |
|
331 |
+
| 2.7349 | 1950 | 0.0139 |
|
332 |
+
| 2.8050 | 2000 | 0.0119 |
|
333 |
+
| 2.8752 | 2050 | 0.0131 |
|
334 |
+
| 2.9453 | 2100 | 0.0124 |
|
335 |
+
| 3.0154 | 2150 | 0.011 |
|
336 |
+
| 3.0856 | 2200 | 0.0109 |
|
337 |
+
| 3.1557 | 2250 | 0.0103 |
|
338 |
+
| 3.2258 | 2300 | 0.0102 |
|
339 |
+
| 3.2959 | 2350 | 0.0089 |
|
340 |
+
| 3.3661 | 2400 | 0.0083 |
|
341 |
+
| 3.4362 | 2450 | 0.0095 |
|
342 |
+
| 3.5063 | 2500 | 0.0085 |
|
343 |
+
| 3.5764 | 2550 | 0.009 |
|
344 |
+
| 3.6466 | 2600 | 0.0083 |
|
345 |
+
| 3.7167 | 2650 | 0.0093 |
|
346 |
+
| 3.7868 | 2700 | 0.0084 |
|
347 |
+
| 3.8569 | 2750 | 0.0084 |
|
348 |
+
| 3.9271 | 2800 | 0.0088 |
|
349 |
+
| 3.9972 | 2850 | 0.0086 |
|
350 |
+
| 4.0673 | 2900 | 0.0057 |
|
351 |
+
| 4.1374 | 2950 | 0.0078 |
|
352 |
+
| 4.2076 | 3000 | 0.0062 |
|
353 |
+
| 4.2777 | 3050 | 0.0066 |
|
354 |
+
| 4.3478 | 3100 | 0.006 |
|
355 |
+
| 4.4180 | 3150 | 0.0078 |
|
356 |
+
| 4.4881 | 3200 | 0.0056 |
|
357 |
+
| 4.5582 | 3250 | 0.0064 |
|
358 |
+
| 4.6283 | 3300 | 0.0063 |
|
359 |
+
| 4.6985 | 3350 | 0.0058 |
|
360 |
+
| 4.7686 | 3400 | 0.005 |
|
361 |
+
| 4.8387 | 3450 | 0.0057 |
|
362 |
+
| 4.9088 | 3500 | 0.0059 |
|
363 |
+
| 4.9790 | 3550 | 0.0063 |
|
364 |
+
| 5.0491 | 3600 | 0.0046 |
|
365 |
+
| 5.1192 | 3650 | 0.0041 |
|
366 |
+
| 5.1893 | 3700 | 0.005 |
|
367 |
+
| 5.2595 | 3750 | 0.0043 |
|
368 |
+
| 5.3296 | 3800 | 0.0046 |
|
369 |
+
| 5.3997 | 3850 | 0.0041 |
|
370 |
+
| 5.4698 | 3900 | 0.006 |
|
371 |
+
| 5.5400 | 3950 | 0.0052 |
|
372 |
+
| 5.6101 | 4000 | 0.0043 |
|
373 |
+
|
374 |
+
|
375 |
+
### Framework Versions
|
376 |
+
- Python: 3.11.12
|
377 |
+
- Sentence Transformers: 4.1.0
|
378 |
+
- Transformers: 4.51.3
|
379 |
+
- PyTorch: 2.6.0+cu124
|
380 |
+
- Accelerate: 1.6.0
|
381 |
+
- Datasets: 2.14.4
|
382 |
+
- Tokenizers: 0.21.1
|
383 |
+
|
384 |
+
## Citation
|
385 |
+
|
386 |
+
### BibTeX
|
387 |
+
|
388 |
+
#### Sentence Transformers
|
389 |
+
```bibtex
|
390 |
+
@inproceedings{reimers-2019-sentence-bert,
|
391 |
+
title = "Sentence-BERT: Sentence Embeddings using Siamese BERT-Networks",
|
392 |
+
author = "Reimers, Nils and Gurevych, Iryna",
|
393 |
+
booktitle = "Proceedings of the 2019 Conference on Empirical Methods in Natural Language Processing",
|
394 |
+
month = "11",
|
395 |
+
year = "2019",
|
396 |
+
publisher = "Association for Computational Linguistics",
|
397 |
+
url = "https://arxiv.org/abs/1908.10084",
|
398 |
+
}
|
399 |
+
```
|
400 |
+
|
401 |
+
<!--
|
402 |
+
## Glossary
|
403 |
+
|
404 |
+
*Clearly define terms in order to be accessible across audiences.*
|
405 |
+
-->
|
406 |
+
|
407 |
+
<!--
|
408 |
+
## Model Card Authors
|
409 |
+
|
410 |
+
*Lists the people who create the model card, providing recognition and accountability for the detailed work that goes into its construction.*
|
411 |
+
-->
|
412 |
+
|
413 |
+
<!--
|
414 |
+
## Model Card Contact
|
415 |
+
|
416 |
+
*Provides a way for people who have updates to the Model Card, suggestions, or questions, to contact the Model Card authors.*
|
417 |
-->
|
config.json
CHANGED
@@ -1,24 +1,25 @@
|
|
1 |
-
{
|
2 |
-
"
|
3 |
-
|
4 |
-
|
5 |
-
|
6 |
-
"
|
7 |
-
"
|
8 |
-
"
|
9 |
-
"
|
10 |
-
"
|
11 |
-
"
|
12 |
-
"
|
13 |
-
"
|
14 |
-
"
|
15 |
-
"
|
16 |
-
"
|
17 |
-
"
|
18 |
-
"
|
19 |
-
"
|
20 |
-
"
|
21 |
-
"
|
22 |
-
"
|
23 |
-
"
|
24 |
-
|
|
|
|
1 |
+
{
|
2 |
+
"_name_or_path": "Detomo/cl-nagoya-sup-simcse-ja-nss-v1_0_7_8",
|
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 |
+
"intermediate_size": 3072,
|
13 |
+
"layer_norm_eps": 1e-12,
|
14 |
+
"max_position_embeddings": 512,
|
15 |
+
"model_type": "bert",
|
16 |
+
"num_attention_heads": 12,
|
17 |
+
"num_hidden_layers": 12,
|
18 |
+
"pad_token_id": 0,
|
19 |
+
"position_embedding_type": "absolute",
|
20 |
+
"torch_dtype": "float32",
|
21 |
+
"transformers_version": "4.48.3",
|
22 |
+
"type_vocab_size": 2,
|
23 |
+
"use_cache": true,
|
24 |
+
"vocab_size": 32768
|
25 |
+
}
|
config_sentence_transformers.json
CHANGED
@@ -1,10 +1,10 @@
|
|
1 |
-
{
|
2 |
-
"__version__": {
|
3 |
-
"sentence_transformers": "
|
4 |
-
"transformers": "4.
|
5 |
-
"pytorch": "2.6.0+
|
6 |
-
},
|
7 |
-
"prompts": {},
|
8 |
-
"default_prompt_name": null,
|
9 |
-
"similarity_fn_name": "cosine"
|
10 |
}
|
|
|
1 |
+
{
|
2 |
+
"__version__": {
|
3 |
+
"sentence_transformers": "3.3.1",
|
4 |
+
"transformers": "4.48.3",
|
5 |
+
"pytorch": "2.6.0+cu126"
|
6 |
+
},
|
7 |
+
"prompts": {},
|
8 |
+
"default_prompt_name": null,
|
9 |
+
"similarity_fn_name": "cosine"
|
10 |
}
|
modules.json
CHANGED
@@ -1,14 +1,14 @@
|
|
1 |
-
[
|
2 |
-
{
|
3 |
-
"idx": 0,
|
4 |
-
"name": "0",
|
5 |
-
"path": "",
|
6 |
-
"type": "sentence_transformers.models.Transformer"
|
7 |
-
},
|
8 |
-
{
|
9 |
-
"idx": 1,
|
10 |
-
"name": "1",
|
11 |
-
"path": "1_Pooling",
|
12 |
-
"type": "sentence_transformers.models.Pooling"
|
13 |
-
}
|
14 |
]
|
|
|
1 |
+
[
|
2 |
+
{
|
3 |
+
"idx": 0,
|
4 |
+
"name": "0",
|
5 |
+
"path": "",
|
6 |
+
"type": "sentence_transformers.models.Transformer"
|
7 |
+
},
|
8 |
+
{
|
9 |
+
"idx": 1,
|
10 |
+
"name": "1",
|
11 |
+
"path": "1_Pooling",
|
12 |
+
"type": "sentence_transformers.models.Pooling"
|
13 |
+
}
|
14 |
]
|
openvino/openvino_model.bin
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:714c47a7a0601faeed7d530dc7a3f3b6d027fe77b8c868a929d0620ce9e52268
|
3 |
+
size 442470560
|
openvino/openvino_model.xml
ADDED
The diff for this file is too large to render.
See raw diff
|
|
sentence_bert_config.json
CHANGED
@@ -1,4 +1,4 @@
|
|
1 |
-
{
|
2 |
-
"max_seq_length": 512,
|
3 |
-
"do_lower_case": false
|
4 |
}
|
|
|
1 |
+
{
|
2 |
+
"max_seq_length": 512,
|
3 |
+
"do_lower_case": false
|
4 |
}
|
special_tokens_map.json
CHANGED
@@ -1,37 +1,37 @@
|
|
1 |
-
{
|
2 |
-
"cls_token": {
|
3 |
-
"content": "[CLS]",
|
4 |
-
"lstrip": false,
|
5 |
-
"normalized": false,
|
6 |
-
"rstrip": false,
|
7 |
-
"single_word": false
|
8 |
-
},
|
9 |
-
"mask_token": {
|
10 |
-
"content": "[MASK]",
|
11 |
-
"lstrip": false,
|
12 |
-
"normalized": false,
|
13 |
-
"rstrip": false,
|
14 |
-
"single_word": false
|
15 |
-
},
|
16 |
-
"pad_token": {
|
17 |
-
"content": "[PAD]",
|
18 |
-
"lstrip": false,
|
19 |
-
"normalized": false,
|
20 |
-
"rstrip": false,
|
21 |
-
"single_word": false
|
22 |
-
},
|
23 |
-
"sep_token": {
|
24 |
-
"content": "[SEP]",
|
25 |
-
"lstrip": false,
|
26 |
-
"normalized": false,
|
27 |
-
"rstrip": false,
|
28 |
-
"single_word": false
|
29 |
-
},
|
30 |
-
"unk_token": {
|
31 |
-
"content": "[UNK]",
|
32 |
-
"lstrip": false,
|
33 |
-
"normalized": false,
|
34 |
-
"rstrip": false,
|
35 |
-
"single_word": false
|
36 |
-
}
|
37 |
-
}
|
|
|
1 |
+
{
|
2 |
+
"cls_token": {
|
3 |
+
"content": "[CLS]",
|
4 |
+
"lstrip": false,
|
5 |
+
"normalized": false,
|
6 |
+
"rstrip": false,
|
7 |
+
"single_word": false
|
8 |
+
},
|
9 |
+
"mask_token": {
|
10 |
+
"content": "[MASK]",
|
11 |
+
"lstrip": false,
|
12 |
+
"normalized": false,
|
13 |
+
"rstrip": false,
|
14 |
+
"single_word": false
|
15 |
+
},
|
16 |
+
"pad_token": {
|
17 |
+
"content": "[PAD]",
|
18 |
+
"lstrip": false,
|
19 |
+
"normalized": false,
|
20 |
+
"rstrip": false,
|
21 |
+
"single_word": false
|
22 |
+
},
|
23 |
+
"sep_token": {
|
24 |
+
"content": "[SEP]",
|
25 |
+
"lstrip": false,
|
26 |
+
"normalized": false,
|
27 |
+
"rstrip": false,
|
28 |
+
"single_word": false
|
29 |
+
},
|
30 |
+
"unk_token": {
|
31 |
+
"content": "[UNK]",
|
32 |
+
"lstrip": false,
|
33 |
+
"normalized": false,
|
34 |
+
"rstrip": false,
|
35 |
+
"single_word": false
|
36 |
+
}
|
37 |
+
}
|
tokenizer_config.json
CHANGED
@@ -1,64 +1,64 @@
|
|
1 |
-
{
|
2 |
-
"added_tokens_decoder": {
|
3 |
-
"0": {
|
4 |
-
"content": "[PAD]",
|
5 |
-
"lstrip": false,
|
6 |
-
"normalized": false,
|
7 |
-
"rstrip": false,
|
8 |
-
"single_word": false,
|
9 |
-
"special": true
|
10 |
-
},
|
11 |
-
"1": {
|
12 |
-
"content": "[UNK]",
|
13 |
-
"lstrip": false,
|
14 |
-
"normalized": false,
|
15 |
-
"rstrip": false,
|
16 |
-
"single_word": false,
|
17 |
-
"special": true
|
18 |
-
},
|
19 |
-
"2": {
|
20 |
-
"content": "[CLS]",
|
21 |
-
"lstrip": false,
|
22 |
-
"normalized": false,
|
23 |
-
"rstrip": false,
|
24 |
-
"single_word": false,
|
25 |
-
"special": true
|
26 |
-
},
|
27 |
-
"3": {
|
28 |
-
"content": "[SEP]",
|
29 |
-
"lstrip": false,
|
30 |
-
"normalized": false,
|
31 |
-
"rstrip": false,
|
32 |
-
"single_word": false,
|
33 |
-
"special": true
|
34 |
-
},
|
35 |
-
"4": {
|
36 |
-
"content": "[MASK]",
|
37 |
-
"lstrip": false,
|
38 |
-
"normalized": false,
|
39 |
-
"rstrip": false,
|
40 |
-
"single_word": false,
|
41 |
-
"special": true
|
42 |
-
}
|
43 |
-
},
|
44 |
-
"clean_up_tokenization_spaces": true,
|
45 |
-
"cls_token": "[CLS]",
|
46 |
-
"do_lower_case": false,
|
47 |
-
"do_subword_tokenize": true,
|
48 |
-
"do_word_tokenize": true,
|
49 |
-
"extra_special_tokens": {},
|
50 |
-
"jumanpp_kwargs": null,
|
51 |
-
"mask_token": "[MASK]",
|
52 |
-
"mecab_kwargs": {
|
53 |
-
"mecab_dic": "unidic_lite"
|
54 |
-
},
|
55 |
-
"model_max_length": 512,
|
56 |
-
"never_split": null,
|
57 |
-
"pad_token": "[PAD]",
|
58 |
-
"sep_token": "[SEP]",
|
59 |
-
"subword_tokenizer_type": "wordpiece",
|
60 |
-
"sudachi_kwargs": null,
|
61 |
-
"tokenizer_class": "BertJapaneseTokenizer",
|
62 |
-
"unk_token": "[UNK]",
|
63 |
-
"word_tokenizer_type": "mecab"
|
64 |
-
}
|
|
|
1 |
+
{
|
2 |
+
"added_tokens_decoder": {
|
3 |
+
"0": {
|
4 |
+
"content": "[PAD]",
|
5 |
+
"lstrip": false,
|
6 |
+
"normalized": false,
|
7 |
+
"rstrip": false,
|
8 |
+
"single_word": false,
|
9 |
+
"special": true
|
10 |
+
},
|
11 |
+
"1": {
|
12 |
+
"content": "[UNK]",
|
13 |
+
"lstrip": false,
|
14 |
+
"normalized": false,
|
15 |
+
"rstrip": false,
|
16 |
+
"single_word": false,
|
17 |
+
"special": true
|
18 |
+
},
|
19 |
+
"2": {
|
20 |
+
"content": "[CLS]",
|
21 |
+
"lstrip": false,
|
22 |
+
"normalized": false,
|
23 |
+
"rstrip": false,
|
24 |
+
"single_word": false,
|
25 |
+
"special": true
|
26 |
+
},
|
27 |
+
"3": {
|
28 |
+
"content": "[SEP]",
|
29 |
+
"lstrip": false,
|
30 |
+
"normalized": false,
|
31 |
+
"rstrip": false,
|
32 |
+
"single_word": false,
|
33 |
+
"special": true
|
34 |
+
},
|
35 |
+
"4": {
|
36 |
+
"content": "[MASK]",
|
37 |
+
"lstrip": false,
|
38 |
+
"normalized": false,
|
39 |
+
"rstrip": false,
|
40 |
+
"single_word": false,
|
41 |
+
"special": true
|
42 |
+
}
|
43 |
+
},
|
44 |
+
"clean_up_tokenization_spaces": true,
|
45 |
+
"cls_token": "[CLS]",
|
46 |
+
"do_lower_case": false,
|
47 |
+
"do_subword_tokenize": true,
|
48 |
+
"do_word_tokenize": true,
|
49 |
+
"extra_special_tokens": {},
|
50 |
+
"jumanpp_kwargs": null,
|
51 |
+
"mask_token": "[MASK]",
|
52 |
+
"mecab_kwargs": {
|
53 |
+
"mecab_dic": "unidic_lite"
|
54 |
+
},
|
55 |
+
"model_max_length": 512,
|
56 |
+
"never_split": null,
|
57 |
+
"pad_token": "[PAD]",
|
58 |
+
"sep_token": "[SEP]",
|
59 |
+
"subword_tokenizer_type": "wordpiece",
|
60 |
+
"sudachi_kwargs": null,
|
61 |
+
"tokenizer_class": "BertJapaneseTokenizer",
|
62 |
+
"unk_token": "[UNK]",
|
63 |
+
"word_tokenizer_type": "mecab"
|
64 |
+
}
|
vocab.txt
CHANGED
The diff for this file is too large to render.
See raw diff
|
|