File size: 46,066 Bytes
eb8f16f |
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258 259 260 261 262 263 264 265 266 267 268 269 270 271 272 273 274 275 276 277 278 279 280 281 282 283 284 285 286 287 288 289 290 291 292 293 294 295 296 297 298 299 300 301 302 303 304 305 306 307 308 309 310 311 312 313 314 315 316 317 318 319 320 321 322 323 324 325 326 327 328 329 330 331 332 333 334 335 336 337 338 339 340 341 342 343 344 345 346 347 348 349 350 351 352 353 354 355 356 357 358 359 360 361 362 363 364 365 366 367 368 369 370 371 372 373 374 375 376 377 378 379 380 381 382 383 384 385 386 387 388 389 390 391 392 393 394 395 396 397 398 399 400 401 402 403 404 405 406 407 408 409 410 411 412 413 414 415 416 417 418 419 420 421 422 423 424 425 426 427 428 429 430 431 432 433 434 435 436 437 438 439 440 441 442 443 444 445 446 447 448 449 450 451 452 453 454 455 456 457 458 459 460 461 462 463 464 465 466 467 468 469 470 471 472 473 474 475 476 477 478 479 480 481 482 483 484 485 486 487 488 489 490 491 492 493 494 495 496 497 498 499 500 501 502 503 504 505 506 507 508 509 510 511 512 513 514 515 516 517 518 519 520 521 522 523 524 525 526 527 528 529 530 531 532 533 534 535 536 537 538 539 540 541 542 543 544 545 546 547 548 549 550 551 552 553 554 555 556 557 558 559 560 561 562 563 564 565 566 567 568 569 570 571 572 573 574 575 576 577 578 579 580 581 582 583 584 585 586 587 588 589 590 591 592 593 594 595 596 597 598 599 600 601 602 603 604 605 606 607 608 609 610 611 612 613 614 615 616 617 618 619 620 621 622 623 624 625 626 627 628 629 630 631 632 633 634 635 636 637 638 639 640 641 642 643 644 645 646 647 648 649 650 651 652 653 654 655 656 657 658 659 660 661 662 663 664 665 666 667 668 669 670 671 672 673 674 675 676 677 678 679 680 681 682 683 684 685 686 687 688 689 690 691 692 693 694 695 696 697 698 699 700 701 702 703 704 705 706 707 708 709 710 711 712 713 714 715 716 717 718 719 720 721 722 723 724 725 726 727 728 729 730 731 732 733 734 735 736 737 738 739 740 741 742 743 744 745 746 747 748 749 750 751 752 753 754 755 756 757 758 759 760 761 762 763 764 765 766 767 768 769 770 771 772 773 774 775 776 777 778 779 780 781 782 783 784 785 786 787 788 789 790 791 792 793 794 795 796 797 798 799 800 801 802 803 804 805 806 807 808 809 810 811 812 813 814 815 816 817 818 819 820 821 822 823 824 825 826 827 828 829 830 831 832 833 834 835 836 837 838 839 840 841 842 843 844 845 846 847 848 849 850 851 852 853 854 855 856 857 858 859 860 861 862 863 864 865 866 867 868 869 870 871 872 873 874 875 876 877 878 879 880 881 882 883 884 885 886 887 888 889 890 891 892 893 894 895 896 897 898 899 900 901 902 903 904 905 906 907 908 909 |
---
language:
- en
license: apache-2.0
library_name: sentence-transformers
tags:
- sentence-transformers
- sentence-similarity
- feature-extraction
- generated_from_trainer
- dataset_size:2720
- loss:MatryoshkaLoss
- loss:MultipleNegativesRankingLoss
base_model: keepitreal/vietnamese-sbert
datasets: []
metrics:
- cosine_accuracy@1
- cosine_accuracy@3
- cosine_accuracy@5
- cosine_accuracy@10
- cosine_precision@1
- cosine_precision@3
- cosine_precision@5
- cosine_precision@10
- cosine_recall@1
- cosine_recall@3
- cosine_recall@5
- cosine_recall@10
- cosine_ndcg@10
- cosine_mrr@10
- cosine_map@100
widget:
- source_sentence: 'Tại Bộ phận Một cửa: Bố trí khu vực cung cấp thông tin, thủ tục
hành chính; khu vực tiếp nhận và trả kết quả được chia thành từng quầy tương ứng
với từng lĩnh vực khác nhau; bố trí đủ ghế ngồi chờ, bàn để viết, máy tính có
kết nối mạng dành cho tổ chức, cá nhân đến giao dịch, thực hiện dịch vụ công trực
tuyến; lắp đặt camera theo dõi toàn bộ khu vực làm việc của Bộ phận Một cửa có
kết nối với cơ quan nhà nước cấp trên và trong toàn hệ thống; bố trí khu vực đặt
các trang thiết bị: máy lấy số xếp hàng tự động kết nối tới Hệ thống thông tin
một cửa điện tử; các màn hình cảm ứng phục vụ tổ chức, cá nhân tra cứu thông tin,
thủ tục hành chính, tra cứu kết quả giải quyết thủ tục hành chính; bố trí khu
vực cung cấp dịch vụ quy định tại điểm g Khoản 1 Điều 8 Nghị định này.'
sentences:
- Bộ phận một cửa của UBND xã phải được bố trí những trang thiết bị nào?
- Tổ chức thẩm định kết quả pháp điển hệ thống quy phạm pháp luật tại Bộ Tư pháp
được quy định như thế nào?
- Có được đọc, sử dụng thánh kinh trong trại giam không?
- source_sentence: "Cơ sở sản xuất được rút tên khỏi Danh sách cơ sở sản xuất có thuốc\
\ vi phạm chất lượng khi đáp ứng đầy đủ các quy định sau đây: \na) Cơ sở nhập\
\ khẩu thực hiện đầy đủ việc kiểm nghiệm thuốc trước khi đưa ra lưu hành theo\
\ thời hạn quy định tại Khoản 1 Điều này; \nb) Cơ sở sản xuất hoặc cơ sở đăng\
\ ký thuốc có báo cáo theo quy định tại Mẫu số 07 Phụ lục III ban hành kèm theo\
\ Thông tư này, kèm theo bằng chứng thực hiện việc kiểm nghiệm toàn bộ các lô\
\ thuốc nhập khẩu vào Việt Nam trong thời hạn thực hiện quy định tại Khoản 1 Điều\
\ này; \nc) Cơ sở sản xuất không có vi phạm chất lượng thuốc (kể cả thu hồi thuốc\
\ theo hình thức tự nguyện vì lý do chất lượng) trong thời hạn thực hiện quy định\
\ tại Khoản 1 Điều này."
sentences:
- Đi du học 2 năm có bị xóa tên trong sổ đăng ký tạm trú không?
- Điều kiện để cơ sở sản xuất được rút tên khỏi Danh sách cơ sở sản xuất có thuốc
vi phạm chất lượng được quy định như thế nào?
- Thế nào là kinh doanh vận tải khách du lịch bằng xe ô tô?
- source_sentence: '"k) Cơ sở đã được cấp một trong các Giấy chứng nhận: ... Hệ thống
quản lý an toàn thực phẩm ISO 22000, ... hoặc tương đương còn hiệu lực."'
sentences:
- Thế nào là phương thức khớp lệnh tập trung?
- Đã có giấy chứng nhận ISO có phải xin giấy chứng nhận cơ sở đủ điều kiện an toàn
thực phẩm không?
- Hợp đồng tặng cho bất động sản có hiệu lực từ khi nào?
- source_sentence: 'Hồ sơ nhập khẩu thức ăn chăn nuôi để giới thiệu tại hội chợ, triển
lãm bao gồm:
a) Đơn đề nghị nhập khẩu thức ăn chăn nuôi để giới thiệu tại hội chợ, triển lãm
theo Mẫu số 07.TACN Phụ lục I ban hành kèm theo Nghị định này;
b) Văn bản chứng minh về việc tổ chức, tham gia hội chợ, triển lãm tại Việt Nam.'
sentences:
- Quy định về hồ sơ nhập khẩu thức ăn chăn nuôi để giới thiệu tại hội chợ, triển
lãm
- Mức phạt người điều khiển ô tô sử dụng tem kiểm định an toàn kỹ thuật và BVMT
không do cơ quan có thẩm quyền cấp
- Xác định mục tiêu, thời hạn của chương trình quản lý tổng hợp tài nguyên vùng
bờ được quy định như thế nào?
- source_sentence: Trường hợp bảo đảm bằng quyền sử dụng đất, tài sản gắn liền với
đất thì hợp đồng thế chấp vẫn còn hiệu lực, biện pháp thế chấp vẫn còn hiệu lực
đối kháng với người thứ ba khi tài sản gắn liền với đất hoặc quyền sử dụng đất
không phải là tài sản bảo đảm được mua bán, được chuyển nhượng, được chuyển giao
khác về quyền sở hữu hoặc được dùng để bảo đảm thực hiện nghĩa vụ. Hiệu lực của
hợp đồng bảo đảm, hiệu lực đối kháng của biện pháp bảo đảm bằng tài sản được tạo
lập từ quyền bề mặt, quyền hưởng dụng với người thứ ba không thay đổi hoặc không
chấm dứt trong trường hợp quyền sử dụng đất có quyền bề mặt, tài sản là đối tượng
của quyền hưởng dụng được mua bán, được chuyển nhượng, được chuyển giao khác về
quyền sở hữu hoặc được dùng để bảo đảm thực hiện nghĩa vụ.
sentences:
- Thời hạn hủy bỏ quyết định tạm đình chỉ liên quan đến tham nhũng được quy định
như thế nào?
- Bảo đảm bằng quyền sử dụng đất, tài sản gắn liền với đất và tài sản được tạo lập
từ quyền bề mặt, quyền hưởng dụng được quy định như thế nào?
- Phòng Kinh tế thuộc UBND huyện được giao quản lý về thú y có được thu phí, lệ
phí thú y không?
pipeline_tag: sentence-similarity
model-index:
- name: Legal vietnamese-sbert
results:
- task:
type: information-retrieval
name: Information Retrieval
dataset:
name: dim 768
type: dim_768
metrics:
- type: cosine_accuracy@1
value: 0.6105610561056105
name: Cosine Accuracy@1
- type: cosine_accuracy@3
value: 0.7557755775577558
name: Cosine Accuracy@3
- type: cosine_accuracy@5
value: 0.7887788778877888
name: Cosine Accuracy@5
- type: cosine_accuracy@10
value: 0.834983498349835
name: Cosine Accuracy@10
- type: cosine_precision@1
value: 0.6105610561056105
name: Cosine Precision@1
- type: cosine_precision@3
value: 0.2519251925192519
name: Cosine Precision@3
- type: cosine_precision@5
value: 0.15775577557755774
name: Cosine Precision@5
- type: cosine_precision@10
value: 0.08349834983498348
name: Cosine Precision@10
- type: cosine_recall@1
value: 0.6105610561056105
name: Cosine Recall@1
- type: cosine_recall@3
value: 0.7557755775577558
name: Cosine Recall@3
- type: cosine_recall@5
value: 0.7887788778877888
name: Cosine Recall@5
- type: cosine_recall@10
value: 0.834983498349835
name: Cosine Recall@10
- type: cosine_ndcg@10
value: 0.7257394196122352
name: Cosine Ndcg@10
- type: cosine_mrr@10
value: 0.6905364345958406
name: Cosine Mrr@10
- type: cosine_map@100
value: 0.6948809217615789
name: Cosine Map@100
- task:
type: information-retrieval
name: Information Retrieval
dataset:
name: dim 512
type: dim_512
metrics:
- type: cosine_accuracy@1
value: 0.6072607260726073
name: Cosine Accuracy@1
- type: cosine_accuracy@3
value: 0.7524752475247525
name: Cosine Accuracy@3
- type: cosine_accuracy@5
value: 0.7854785478547854
name: Cosine Accuracy@5
- type: cosine_accuracy@10
value: 0.8283828382838284
name: Cosine Accuracy@10
- type: cosine_precision@1
value: 0.6072607260726073
name: Cosine Precision@1
- type: cosine_precision@3
value: 0.25082508250825075
name: Cosine Precision@3
- type: cosine_precision@5
value: 0.15709570957095706
name: Cosine Precision@5
- type: cosine_precision@10
value: 0.08283828382838283
name: Cosine Precision@10
- type: cosine_recall@1
value: 0.6072607260726073
name: Cosine Recall@1
- type: cosine_recall@3
value: 0.7524752475247525
name: Cosine Recall@3
- type: cosine_recall@5
value: 0.7854785478547854
name: Cosine Recall@5
- type: cosine_recall@10
value: 0.8283828382838284
name: Cosine Recall@10
- type: cosine_ndcg@10
value: 0.72090653484064
name: Cosine Ndcg@10
- type: cosine_mrr@10
value: 0.6861739745403111
name: Cosine Mrr@10
- type: cosine_map@100
value: 0.6912926659337115
name: Cosine Map@100
- task:
type: information-retrieval
name: Information Retrieval
dataset:
name: dim 256
type: dim_256
metrics:
- type: cosine_accuracy@1
value: 0.5742574257425742
name: Cosine Accuracy@1
- type: cosine_accuracy@3
value: 0.7425742574257426
name: Cosine Accuracy@3
- type: cosine_accuracy@5
value: 0.7788778877887789
name: Cosine Accuracy@5
- type: cosine_accuracy@10
value: 0.8250825082508251
name: Cosine Accuracy@10
- type: cosine_precision@1
value: 0.5742574257425742
name: Cosine Precision@1
- type: cosine_precision@3
value: 0.24752475247524747
name: Cosine Precision@3
- type: cosine_precision@5
value: 0.15577557755775576
name: Cosine Precision@5
- type: cosine_precision@10
value: 0.08250825082508248
name: Cosine Precision@10
- type: cosine_recall@1
value: 0.5742574257425742
name: Cosine Recall@1
- type: cosine_recall@3
value: 0.7425742574257426
name: Cosine Recall@3
- type: cosine_recall@5
value: 0.7788778877887789
name: Cosine Recall@5
- type: cosine_recall@10
value: 0.8250825082508251
name: Cosine Recall@10
- type: cosine_ndcg@10
value: 0.7044686281966969
name: Cosine Ndcg@10
- type: cosine_mrr@10
value: 0.6654683325475405
name: Cosine Mrr@10
- type: cosine_map@100
value: 0.6705497333666119
name: Cosine Map@100
- task:
type: information-retrieval
name: Information Retrieval
dataset:
name: dim 128
type: dim_128
metrics:
- type: cosine_accuracy@1
value: 0.570957095709571
name: Cosine Accuracy@1
- type: cosine_accuracy@3
value: 0.7062706270627063
name: Cosine Accuracy@3
- type: cosine_accuracy@5
value: 0.7557755775577558
name: Cosine Accuracy@5
- type: cosine_accuracy@10
value: 0.8118811881188119
name: Cosine Accuracy@10
- type: cosine_precision@1
value: 0.570957095709571
name: Cosine Precision@1
- type: cosine_precision@3
value: 0.23542354235423535
name: Cosine Precision@3
- type: cosine_precision@5
value: 0.15115511551155114
name: Cosine Precision@5
- type: cosine_precision@10
value: 0.08118811881188116
name: Cosine Precision@10
- type: cosine_recall@1
value: 0.570957095709571
name: Cosine Recall@1
- type: cosine_recall@3
value: 0.7062706270627063
name: Cosine Recall@3
- type: cosine_recall@5
value: 0.7557755775577558
name: Cosine Recall@5
- type: cosine_recall@10
value: 0.8118811881188119
name: Cosine Recall@10
- type: cosine_ndcg@10
value: 0.6904548657504692
name: Cosine Ndcg@10
- type: cosine_mrr@10
value: 0.6517208863743519
name: Cosine Mrr@10
- type: cosine_map@100
value: 0.6569610011044806
name: Cosine Map@100
- task:
type: information-retrieval
name: Information Retrieval
dataset:
name: dim 64
type: dim_64
metrics:
- type: cosine_accuracy@1
value: 0.5346534653465347
name: Cosine Accuracy@1
- type: cosine_accuracy@3
value: 0.7029702970297029
name: Cosine Accuracy@3
- type: cosine_accuracy@5
value: 0.7491749174917491
name: Cosine Accuracy@5
- type: cosine_accuracy@10
value: 0.801980198019802
name: Cosine Accuracy@10
- type: cosine_precision@1
value: 0.5346534653465347
name: Cosine Precision@1
- type: cosine_precision@3
value: 0.23432343234323427
name: Cosine Precision@3
- type: cosine_precision@5
value: 0.1498349834983498
name: Cosine Precision@5
- type: cosine_precision@10
value: 0.08019801980198019
name: Cosine Precision@10
- type: cosine_recall@1
value: 0.5346534653465347
name: Cosine Recall@1
- type: cosine_recall@3
value: 0.7029702970297029
name: Cosine Recall@3
- type: cosine_recall@5
value: 0.7491749174917491
name: Cosine Recall@5
- type: cosine_recall@10
value: 0.801980198019802
name: Cosine Recall@10
- type: cosine_ndcg@10
value: 0.670267766192947
name: Cosine Ndcg@10
- type: cosine_mrr@10
value: 0.628023992875478
name: Cosine Mrr@10
- type: cosine_map@100
value: 0.6331293046972268
name: Cosine Map@100
---
# Legal vietnamese-sbert
This is a [sentence-transformers](https://www.SBERT.net) model finetuned from [keepitreal/vietnamese-sbert](https://huggingface.co/keepitreal/vietnamese-sbert). 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.
## Model Details
### Model Description
- **Model Type:** Sentence Transformer
- **Base model:** [keepitreal/vietnamese-sbert](https://huggingface.co/keepitreal/vietnamese-sbert) <!-- at revision a9467ef2ef47caa6448edeabfd8e5e5ce0fa2a23 -->
- **Maximum Sequence Length:** 256 tokens
- **Output Dimensionality:** 768 tokens
- **Similarity Function:** Cosine Similarity
<!-- - **Training Dataset:** Unknown -->
- **Language:** en
- **License:** apache-2.0
### Model Sources
- **Documentation:** [Sentence Transformers Documentation](https://sbert.net)
- **Repository:** [Sentence Transformers on GitHub](https://github.com/UKPLab/sentence-transformers)
- **Hugging Face:** [Sentence Transformers on Hugging Face](https://huggingface.co/models?library=sentence-transformers)
### Full Model Architecture
```
SentenceTransformer(
(0): Transformer({'max_seq_length': 256, 'do_lower_case': False}) with Transformer model: RobertaModel
(1): Pooling({'word_embedding_dimension': 768, 'pooling_mode_cls_token': False, 'pooling_mode_mean_tokens': True, 'pooling_mode_max_tokens': False, 'pooling_mode_mean_sqrt_len_tokens': False, 'pooling_mode_weightedmean_tokens': False, 'pooling_mode_lasttoken': False, 'include_prompt': True})
)
```
## Usage
### Direct Usage (Sentence Transformers)
First install the Sentence Transformers library:
```bash
pip install -U sentence-transformers
```
Then you can load this model and run inference.
```python
from sentence_transformers import SentenceTransformer
# Download from the 🤗 Hub
model = SentenceTransformer("ictumuk/legal-vietnamese-sbert")
# Run inference
sentences = [
'Trường hợp bảo đảm bằng quyền sử dụng đất, tài sản gắn liền với đất thì hợp đồng thế chấp vẫn còn hiệu lực, biện pháp thế chấp vẫn còn hiệu lực đối kháng với người thứ ba khi tài sản gắn liền với đất hoặc quyền sử dụng đất không phải là tài sản bảo đảm được mua bán, được chuyển nhượng, được chuyển giao khác về quyền sở hữu hoặc được dùng để bảo đảm thực hiện nghĩa vụ. Hiệu lực của hợp đồng bảo đảm, hiệu lực đối kháng của biện pháp bảo đảm bằng tài sản được tạo lập từ quyền bề mặt, quyền hưởng dụng với người thứ ba không thay đổi hoặc không chấm dứt trong trường hợp quyền sử dụng đất có quyền bề mặt, tài sản là đối tượng của quyền hưởng dụng được mua bán, được chuyển nhượng, được chuyển giao khác về quyền sở hữu hoặc được dùng để bảo đảm thực hiện nghĩa vụ.',
'Bảo đảm bằng quyền sử dụng đất, tài sản gắn liền với đất và tài sản được tạo lập từ quyền bề mặt, quyền hưởng dụng được quy định như thế nào?',
'Thời hạn hủy bỏ quyết định tạm đình chỉ liên quan đến tham nhũng được quy định như thế nào?',
]
embeddings = model.encode(sentences)
print(embeddings.shape)
# [3, 768]
# Get the similarity scores for the embeddings
similarities = model.similarity(embeddings, embeddings)
print(similarities.shape)
# [3, 3]
```
<!--
### Direct Usage (Transformers)
<details><summary>Click to see the direct usage in Transformers</summary>
</details>
-->
<!--
### Downstream Usage (Sentence Transformers)
You can finetune this model on your own dataset.
<details><summary>Click to expand</summary>
</details>
-->
<!--
### Out-of-Scope Use
*List how the model may foreseeably be misused and address what users ought not to do with the model.*
-->
## Evaluation
### Metrics
#### Information Retrieval
* Dataset: `dim_768`
* Evaluated with [<code>InformationRetrievalEvaluator</code>](https://sbert.net/docs/package_reference/sentence_transformer/evaluation.html#sentence_transformers.evaluation.InformationRetrievalEvaluator)
| Metric | Value |
|:--------------------|:-----------|
| cosine_accuracy@1 | 0.6106 |
| cosine_accuracy@3 | 0.7558 |
| cosine_accuracy@5 | 0.7888 |
| cosine_accuracy@10 | 0.835 |
| cosine_precision@1 | 0.6106 |
| cosine_precision@3 | 0.2519 |
| cosine_precision@5 | 0.1578 |
| cosine_precision@10 | 0.0835 |
| cosine_recall@1 | 0.6106 |
| cosine_recall@3 | 0.7558 |
| cosine_recall@5 | 0.7888 |
| cosine_recall@10 | 0.835 |
| cosine_ndcg@10 | 0.7257 |
| cosine_mrr@10 | 0.6905 |
| **cosine_map@100** | **0.6949** |
#### Information Retrieval
* Dataset: `dim_512`
* Evaluated with [<code>InformationRetrievalEvaluator</code>](https://sbert.net/docs/package_reference/sentence_transformer/evaluation.html#sentence_transformers.evaluation.InformationRetrievalEvaluator)
| Metric | Value |
|:--------------------|:-----------|
| cosine_accuracy@1 | 0.6073 |
| cosine_accuracy@3 | 0.7525 |
| cosine_accuracy@5 | 0.7855 |
| cosine_accuracy@10 | 0.8284 |
| cosine_precision@1 | 0.6073 |
| cosine_precision@3 | 0.2508 |
| cosine_precision@5 | 0.1571 |
| cosine_precision@10 | 0.0828 |
| cosine_recall@1 | 0.6073 |
| cosine_recall@3 | 0.7525 |
| cosine_recall@5 | 0.7855 |
| cosine_recall@10 | 0.8284 |
| cosine_ndcg@10 | 0.7209 |
| cosine_mrr@10 | 0.6862 |
| **cosine_map@100** | **0.6913** |
#### Information Retrieval
* Dataset: `dim_256`
* Evaluated with [<code>InformationRetrievalEvaluator</code>](https://sbert.net/docs/package_reference/sentence_transformer/evaluation.html#sentence_transformers.evaluation.InformationRetrievalEvaluator)
| Metric | Value |
|:--------------------|:-----------|
| cosine_accuracy@1 | 0.5743 |
| cosine_accuracy@3 | 0.7426 |
| cosine_accuracy@5 | 0.7789 |
| cosine_accuracy@10 | 0.8251 |
| cosine_precision@1 | 0.5743 |
| cosine_precision@3 | 0.2475 |
| cosine_precision@5 | 0.1558 |
| cosine_precision@10 | 0.0825 |
| cosine_recall@1 | 0.5743 |
| cosine_recall@3 | 0.7426 |
| cosine_recall@5 | 0.7789 |
| cosine_recall@10 | 0.8251 |
| cosine_ndcg@10 | 0.7045 |
| cosine_mrr@10 | 0.6655 |
| **cosine_map@100** | **0.6705** |
#### Information Retrieval
* Dataset: `dim_128`
* Evaluated with [<code>InformationRetrievalEvaluator</code>](https://sbert.net/docs/package_reference/sentence_transformer/evaluation.html#sentence_transformers.evaluation.InformationRetrievalEvaluator)
| Metric | Value |
|:--------------------|:----------|
| cosine_accuracy@1 | 0.571 |
| cosine_accuracy@3 | 0.7063 |
| cosine_accuracy@5 | 0.7558 |
| cosine_accuracy@10 | 0.8119 |
| cosine_precision@1 | 0.571 |
| cosine_precision@3 | 0.2354 |
| cosine_precision@5 | 0.1512 |
| cosine_precision@10 | 0.0812 |
| cosine_recall@1 | 0.571 |
| cosine_recall@3 | 0.7063 |
| cosine_recall@5 | 0.7558 |
| cosine_recall@10 | 0.8119 |
| cosine_ndcg@10 | 0.6905 |
| cosine_mrr@10 | 0.6517 |
| **cosine_map@100** | **0.657** |
#### Information Retrieval
* Dataset: `dim_64`
* Evaluated with [<code>InformationRetrievalEvaluator</code>](https://sbert.net/docs/package_reference/sentence_transformer/evaluation.html#sentence_transformers.evaluation.InformationRetrievalEvaluator)
| Metric | Value |
|:--------------------|:-----------|
| cosine_accuracy@1 | 0.5347 |
| cosine_accuracy@3 | 0.703 |
| cosine_accuracy@5 | 0.7492 |
| cosine_accuracy@10 | 0.802 |
| cosine_precision@1 | 0.5347 |
| cosine_precision@3 | 0.2343 |
| cosine_precision@5 | 0.1498 |
| cosine_precision@10 | 0.0802 |
| cosine_recall@1 | 0.5347 |
| cosine_recall@3 | 0.703 |
| cosine_recall@5 | 0.7492 |
| cosine_recall@10 | 0.802 |
| cosine_ndcg@10 | 0.6703 |
| cosine_mrr@10 | 0.628 |
| **cosine_map@100** | **0.6331** |
<!--
## Bias, Risks and Limitations
*What are the known or foreseeable issues stemming from this model? You could also flag here known failure cases or weaknesses of the model.*
-->
<!--
### Recommendations
*What are recommendations with respect to the foreseeable issues? For example, filtering explicit content.*
-->
## Training Details
### Training Dataset
#### Unnamed Dataset
* Size: 2,720 training samples
* Columns: <code>positive</code> and <code>anchor</code>
* Approximate statistics based on the first 1000 samples:
| | positive | anchor |
|:--------|:-----------------------------------------------------------------------------------|:---------------------------------------------------------------------------------|
| type | string | string |
| details | <ul><li>min: 5 tokens</li><li>mean: 75.95 tokens</li><li>max: 256 tokens</li></ul> | <ul><li>min: 8 tokens</li><li>mean: 24.4 tokens</li><li>max: 56 tokens</li></ul> |
* Samples:
| positive | anchor |
|:--------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------|:-----------------------------------------------------------------------------------------------------------------------|
| <code>Mỗi năm có thể được rút ngắn thời gian thử thách 01 lần từ 03 tháng đến 02 năm.</code> | <code>01 lần rút ngắn thời gian thử thách được bao nhiêu ngày?</code> |
| <code>Đánh giá tác động của biến đổi khí hậu đến hoạt động kinh tế - xã hội và các vấn đề liên ngành, liên vùng, liên lĩnh vực, bao gồm các nội dung sau: <br>a) Đánh giá các tác động tích cực, tiêu cực, ngắn hạn, dài hạn của biến đổi khí hậu đến hoạt động kinh tế - xã hội liên quan đến phạm vi của chiến lược, quy hoạch, kế hoạch; <br>b) Đánh giá tác động tích cực, tiêu cực, ngắn hạn, dài hạn của biến đổi khí hậu đến các vấn đề liên ngành, liên vùng, liên lĩnh vực.</code> | <code>Nội dung đánh giá tác động của biến đổi khí hậu đến hoạt động kinh tế - xã hội được quy định như thế nào?</code> |
| <code>Không có thông tin liên quan tới trách nhiệm của Ngân hàng nhà nước trong hai đoạn văn trên. Hai đoạn văn trên chỉ đề cập tới việc thực hiện xác nhận đăng ký hoặc đăng ký thay đổi khoản vay nước ngoài được Chính phủ bảo lãnh và cập nhật thông tin tín dụng Quốc gia Việt Nam.</code> | <code>Trách nhiệm của Ngân hàng nhà nước liên quan tới bảo lãnh Chính phủ được quy định như thế nào?</code> |
* Loss: [<code>MatryoshkaLoss</code>](https://sbert.net/docs/package_reference/sentence_transformer/losses.html#matryoshkaloss) with these parameters:
```json
{
"loss": "MultipleNegativesRankingLoss",
"matryoshka_dims": [
768,
512,
256,
128,
64
],
"matryoshka_weights": [
1,
1,
1,
1,
1
],
"n_dims_per_step": -1
}
```
### Evaluation Dataset
#### Unnamed Dataset
* Size: 303 evaluation samples
* Columns: <code>positive</code> and <code>anchor</code>
* Approximate statistics based on the first 1000 samples:
| | positive | anchor |
|:--------|:-----------------------------------------------------------------------------------|:----------------------------------------------------------------------------------|
| type | string | string |
| details | <ul><li>min: 4 tokens</li><li>mean: 77.76 tokens</li><li>max: 256 tokens</li></ul> | <ul><li>min: 6 tokens</li><li>mean: 24.18 tokens</li><li>max: 73 tokens</li></ul> |
* Samples:
| positive | anchor |
|:----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------|:------------------------------------------------------------------------------------------------------------|
| <code>Căn cứ lập hồ sơ mời thầu dự án đầu tư có sử dụng đất bao gồm:<br><br>a) Danh mục dự án đầu tư có sử dụng đất cần lựa chọn nhà đầu tư được công bố theo quy định;<br>b) Kế hoạch lựa chọn nhà đầu tư được duyệt;<br>c) Quy định hiện hành của pháp luật về đất đai, nhà ở, kinh doanh bất động sản, đầu tư, xây dựng và quy định của pháp luật có liên quan.</code> | <code>Lập hồ sơ mời thầu dự án đầu tư có sử dụng đất dựa trên những căn cứ nào?</code> |
| <code>Phạt tiền từ 3.000.000 đồng đến 5.000.000 đồng đối với hành vi phá hoại cơ sở vật chất - kỹ thuật, kho bảo quản hàng dự trữ quốc gia.</code> | <code>Mức phạt đối với hành vi phá hoại kho bảo quản hàng dự trữ quốc gia được quy định như thế nào?</code> |
| <code>Hồ sơ đăng ký thành viên bù trừ bao gồm:<br><br>a) Giấy đăng ký thành viên bù trừ theo Mẫu số 48 Phụ lục ban hành kèm theo Nghị định này;<br>b) Giấy chứng nhận đủ điều kiện cung cấp dịch vụ bù trừ, thanh toán giao dịch chứng khoán do Ủy ban Chứng khoán Nhà nước cấp;<br>c) Bản thuyết minh về hạ tầng công nghệ thông tin, nhân sự, quy trình nghiệp vụ.</code> | <code>Hồ sơ về đăng ký thành viên bù trừ chứng khoán Việt Nam được quy định như thế nào?</code> |
* Loss: [<code>MatryoshkaLoss</code>](https://sbert.net/docs/package_reference/sentence_transformer/losses.html#matryoshkaloss) with these parameters:
```json
{
"loss": "MultipleNegativesRankingLoss",
"matryoshka_dims": [
768,
512,
256,
128,
64
],
"matryoshka_weights": [
1,
1,
1,
1,
1
],
"n_dims_per_step": -1
}
```
### Training Hyperparameters
#### Non-Default Hyperparameters
- `eval_strategy`: epoch
- `per_device_train_batch_size`: 16
- `per_device_eval_batch_size`: 16
- `gradient_accumulation_steps`: 16
- `learning_rate`: 2e-05
- `num_train_epochs`: 25
- `lr_scheduler_type`: cosine
- `warmup_ratio`: 0.1
- `fp16`: True
- `load_best_model_at_end`: True
- `optim`: adamw_torch_fused
- `batch_sampler`: no_duplicates
#### All Hyperparameters
<details><summary>Click to expand</summary>
- `overwrite_output_dir`: False
- `do_predict`: False
- `eval_strategy`: epoch
- `prediction_loss_only`: True
- `per_device_train_batch_size`: 16
- `per_device_eval_batch_size`: 16
- `per_gpu_train_batch_size`: None
- `per_gpu_eval_batch_size`: None
- `gradient_accumulation_steps`: 16
- `eval_accumulation_steps`: None
- `learning_rate`: 2e-05
- `weight_decay`: 0.0
- `adam_beta1`: 0.9
- `adam_beta2`: 0.999
- `adam_epsilon`: 1e-08
- `max_grad_norm`: 1.0
- `num_train_epochs`: 25
- `max_steps`: -1
- `lr_scheduler_type`: cosine
- `lr_scheduler_kwargs`: {}
- `warmup_ratio`: 0.1
- `warmup_steps`: 0
- `log_level`: passive
- `log_level_replica`: warning
- `log_on_each_node`: True
- `logging_nan_inf_filter`: True
- `save_safetensors`: True
- `save_on_each_node`: False
- `save_only_model`: False
- `restore_callback_states_from_checkpoint`: False
- `no_cuda`: False
- `use_cpu`: False
- `use_mps_device`: False
- `seed`: 42
- `data_seed`: None
- `jit_mode_eval`: False
- `use_ipex`: False
- `bf16`: False
- `fp16`: True
- `fp16_opt_level`: O1
- `half_precision_backend`: auto
- `bf16_full_eval`: False
- `fp16_full_eval`: False
- `tf32`: None
- `local_rank`: 0
- `ddp_backend`: None
- `tpu_num_cores`: None
- `tpu_metrics_debug`: False
- `debug`: []
- `dataloader_drop_last`: False
- `dataloader_num_workers`: 0
- `dataloader_prefetch_factor`: None
- `past_index`: -1
- `disable_tqdm`: False
- `remove_unused_columns`: True
- `label_names`: None
- `load_best_model_at_end`: True
- `ignore_data_skip`: False
- `fsdp`: []
- `fsdp_min_num_params`: 0
- `fsdp_config`: {'min_num_params': 0, 'xla': False, 'xla_fsdp_v2': False, 'xla_fsdp_grad_ckpt': False}
- `fsdp_transformer_layer_cls_to_wrap`: None
- `accelerator_config`: {'split_batches': False, 'dispatch_batches': None, 'even_batches': True, 'use_seedable_sampler': True, 'non_blocking': False, 'gradient_accumulation_kwargs': None}
- `deepspeed`: None
- `label_smoothing_factor`: 0.0
- `optim`: adamw_torch_fused
- `optim_args`: None
- `adafactor`: False
- `group_by_length`: False
- `length_column_name`: length
- `ddp_find_unused_parameters`: None
- `ddp_bucket_cap_mb`: None
- `ddp_broadcast_buffers`: False
- `dataloader_pin_memory`: True
- `dataloader_persistent_workers`: False
- `skip_memory_metrics`: True
- `use_legacy_prediction_loop`: False
- `push_to_hub`: False
- `resume_from_checkpoint`: None
- `hub_model_id`: None
- `hub_strategy`: every_save
- `hub_private_repo`: False
- `hub_always_push`: False
- `gradient_checkpointing`: False
- `gradient_checkpointing_kwargs`: None
- `include_inputs_for_metrics`: False
- `eval_do_concat_batches`: True
- `fp16_backend`: auto
- `push_to_hub_model_id`: None
- `push_to_hub_organization`: None
- `mp_parameters`:
- `auto_find_batch_size`: False
- `full_determinism`: False
- `torchdynamo`: None
- `ray_scope`: last
- `ddp_timeout`: 1800
- `torch_compile`: False
- `torch_compile_backend`: None
- `torch_compile_mode`: None
- `dispatch_batches`: None
- `split_batches`: None
- `include_tokens_per_second`: False
- `include_num_input_tokens_seen`: False
- `neftune_noise_alpha`: None
- `optim_target_modules`: None
- `batch_eval_metrics`: False
- `batch_sampler`: no_duplicates
- `multi_dataset_batch_sampler`: proportional
</details>
### Training Logs
| Epoch | Step | Training Loss | loss | dim_128_cosine_map@100 | dim_256_cosine_map@100 | dim_512_cosine_map@100 | dim_64_cosine_map@100 | dim_768_cosine_map@100 |
|:-----------:|:-------:|:-------------:|:----------:|:----------------------:|:----------------------:|:----------------------:|:---------------------:|:----------------------:|
| 0.9412 | 10 | 3.6588 | 2.3368 | 0.5313 | 0.5648 | 0.5747 | 0.4508 | 0.5856 |
| 1.8824 | 20 | 2.6403 | - | - | - | - | - | - |
| 1.9765 | 21 | - | 1.7613 | 0.5788 | 0.5934 | 0.6062 | 0.5009 | 0.6134 |
| 2.8235 | 30 | 1.7794 | - | - | - | - | - | - |
| 2.9176 | 31 | - | 1.5005 | 0.5999 | 0.6218 | 0.6274 | 0.5328 | 0.6299 |
| 3.7647 | 40 | 1.1983 | - | - | - | - | - | - |
| 3.9529 | 42 | - | 1.2607 | 0.6128 | 0.6384 | 0.6492 | 0.5776 | 0.6461 |
| 4.7059 | 50 | 0.8626 | - | - | - | - | - | - |
| 4.9882 | 53 | - | 1.1057 | 0.6176 | 0.6543 | 0.6542 | 0.5907 | 0.6591 |
| 5.6471 | 60 | 0.6381 | - | - | - | - | - | - |
| 5.9294 | 63 | - | 1.0423 | 0.6338 | 0.6678 | 0.6704 | 0.6151 | 0.6755 |
| 6.5882 | 70 | 0.4149 | - | - | - | - | - | - |
| 6.9647 | 74 | - | 0.8898 | 0.6350 | 0.6714 | 0.6828 | 0.6183 | 0.6739 |
| 7.5294 | 80 | 0.3352 | - | - | - | - | - | - |
| 8.0 | 85 | - | 0.9243 | 0.6468 | 0.6679 | 0.6845 | 0.6244 | 0.6743 |
| 8.4706 | 90 | 0.2603 | - | - | - | - | - | - |
| 8.9412 | 95 | - | 0.9162 | 0.6461 | 0.6758 | 0.6900 | 0.6247 | 0.6839 |
| 9.4118 | 100 | 0.2124 | - | - | - | - | - | - |
| 9.9765 | 106 | - | 0.8410 | 0.6428 | 0.6623 | 0.6839 | 0.6247 | 0.6810 |
| 10.3529 | 110 | 0.1555 | - | - | - | - | - | - |
| 10.9176 | 116 | - | 0.8458 | 0.6493 | 0.6660 | 0.6889 | 0.6271 | 0.6884 |
| 11.2941 | 120 | 0.1293 | - | - | - | - | - | - |
| 11.9529 | 127 | - | 0.8563 | 0.6477 | 0.6650 | 0.6850 | 0.6306 | 0.6898 |
| 12.2353 | 130 | 0.1059 | - | - | - | - | - | - |
| 12.9882 | 138 | - | 0.8598 | 0.6494 | 0.6703 | 0.6819 | 0.6217 | 0.6901 |
| 13.1765 | 140 | 0.1071 | - | - | - | - | - | - |
| 13.9294 | 148 | - | 0.8248 | 0.6519 | 0.6627 | 0.6786 | 0.6278 | 0.6923 |
| 14.1176 | 150 | 0.0876 | - | - | - | - | - | - |
| 14.9647 | 159 | - | 0.8213 | 0.6506 | 0.6693 | 0.6877 | 0.6342 | 0.6928 |
| 15.0588 | 160 | 0.0835 | - | - | - | - | - | - |
| 16.0 | 170 | 0.0685 | 0.8266 | 0.6515 | 0.6707 | 0.6835 | 0.6301 | 0.6953 |
| 16.9412 | 180 | 0.057 | 0.8119 | 0.6499 | 0.6714 | 0.6895 | 0.6305 | 0.6976 |
| 17.8824 | 190 | 0.0544 | - | - | - | - | - | - |
| 17.9765 | 191 | - | 0.8226 | 0.6538 | 0.6671 | 0.6883 | 0.6347 | 0.7000 |
| 18.8235 | 200 | 0.0586 | - | - | - | - | - | - |
| **18.9176** | **201** | **-** | **0.8241** | **0.6575** | **0.6716** | **0.6912** | **0.6324** | **0.6984** |
| 19.7647 | 210 | 0.0472 | - | - | - | - | - | - |
| 19.9529 | 212 | - | 0.8270 | 0.6553 | 0.6709 | 0.6913 | 0.6323 | 0.6948 |
| 20.7059 | 220 | 0.0487 | - | - | - | - | - | - |
| 20.9882 | 223 | - | 0.8276 | 0.6538 | 0.6727 | 0.6928 | 0.6325 | 0.6949 |
| 21.6471 | 230 | 0.046 | - | - | - | - | - | - |
| 21.9294 | 233 | - | 0.8278 | 0.6569 | 0.6705 | 0.6929 | 0.6314 | 0.6949 |
| 22.5882 | 240 | 0.0427 | - | - | - | - | - | - |
| 22.9647 | 244 | - | 0.8277 | 0.6569 | 0.6705 | 0.6929 | 0.6331 | 0.6949 |
| 23.5294 | 250 | 0.0466 | 0.8277 | 0.6570 | 0.6705 | 0.6913 | 0.6331 | 0.6949 |
* The bold row denotes the saved checkpoint.
### Framework Versions
- Python: 3.10.13
- Sentence Transformers: 3.0.1
- Transformers: 4.41.2
- PyTorch: 2.1.2
- Accelerate: 0.30.1
- Datasets: 2.19.2
- Tokenizers: 0.19.1
## Citation
### BibTeX
#### Sentence Transformers
```bibtex
@inproceedings{reimers-2019-sentence-bert,
title = "Sentence-BERT: Sentence Embeddings using Siamese BERT-Networks",
author = "Reimers, Nils and Gurevych, Iryna",
booktitle = "Proceedings of the 2019 Conference on Empirical Methods in Natural Language Processing",
month = "11",
year = "2019",
publisher = "Association for Computational Linguistics",
url = "https://arxiv.org/abs/1908.10084",
}
```
#### MatryoshkaLoss
```bibtex
@misc{kusupati2024matryoshka,
title={Matryoshka Representation Learning},
author={Aditya Kusupati and Gantavya Bhatt and Aniket Rege and Matthew Wallingford and Aditya Sinha and Vivek Ramanujan and William Howard-Snyder and Kaifeng Chen and Sham Kakade and Prateek Jain and Ali Farhadi},
year={2024},
eprint={2205.13147},
archivePrefix={arXiv},
primaryClass={cs.LG}
}
```
#### MultipleNegativesRankingLoss
```bibtex
@misc{henderson2017efficient,
title={Efficient Natural Language Response Suggestion for Smart Reply},
author={Matthew Henderson and Rami Al-Rfou and Brian Strope and Yun-hsuan Sung and Laszlo Lukacs and Ruiqi Guo and Sanjiv Kumar and Balint Miklos and Ray Kurzweil},
year={2017},
eprint={1705.00652},
archivePrefix={arXiv},
primaryClass={cs.CL}
}
```
<!--
## Glossary
*Clearly define terms in order to be accessible across audiences.*
-->
<!--
## Model Card Authors
*Lists the people who create the model card, providing recognition and accountability for the detailed work that goes into its construction.*
-->
<!--
## Model Card Contact
*Provides a way for people who have updates to the Model Card, suggestions, or questions, to contact the Model Card authors.*
--> |