--- tags: - sentence-transformers - sentence-similarity - feature-extraction - generated_from_trainer - dataset_size:25580 - loss:OnlineContrastiveLoss base_model: sentence-transformers/paraphrase-multilingual-MiniLM-L12-v2 widget: - source_sentence: ikhtisar arus kas triwulan 1, 2004 (miliar) sentences: - Balita (0-59 Bulan) Menurut Status Gizi, Tahun 1998-2005 - Perbandingan Indeks dan Tingkat Inflasi Desember 2023 Kota-kota di Luar Pulau Jawa dan Sumatera dengan Nasional (2018=100) - Rata-rata Konsumsi dan Pengeluaran Perkapita Seminggu Menurut Komoditi Makanan dan Golongan Pengeluaran per Kapita Seminggu di Provinsi Sulawesi Tengah, 2018-2023 - source_sentence: BaIgaimana gambaran neraca arus dana dUi Indonesia pada kuartal kedua tahun 2015? sentences: - Jumlah Sekolah, Guru, dan Murid Sekolah Menengah Pertama (SMP) di Bawah Kementrian Pendidikan dan Kebudayaan Menurut Provinsi 2011/2012-2015/2016 - Ringkasan Neraca Arus Dana Triwulan III Tahun 2003 (Miliar Rupiah) - Rata-rata Konsumsi dan Pengeluaran Perkapita Seminggu Menurut Komoditi Makanan dan Golongan Pengeluaran per Kapita Seminggu di Provinsi Sulawesi Tenggara, 2018-2023 - source_sentence: Berapa persen pengeluaran orang di kotaa untuk makanan vs non-makanan, per provinsi, 2018? sentences: - Ekspor Tanaman Obat, Aromatik, dan Rempah-Rempah menurut Negara Tujuan Utama, 2012-2023 - Rata-rata Pendapatan Bersih Pekerja Bebas Menurut Provinsi dan Pendidikan Tertinggi yang Ditamatkan (ribu rupiah), 2017 - IHK dan Rata-rata Upah per Bulan Buruh Industri di Bawah Mandor (Supervisor), 1996-2014 (1996=100) - source_sentence: Negara-negara asal impor crude oil dan produk turunannya tahun 2002-2023 sentences: - Persentase Pengeluaran Rata-rata per Kapita Sebulan Menurut Kelompok Barang, Indonesia, 1999, 2002-2023 - Rata-rata Pendapatan Bersih Berusaha Sendiri menurut Provinsi dan Pendidikan yang Ditamatkan (ribu rupiah), 2016 - Perkembangan Beberapa Agregat Pendapatan dan Pendapatan per Kapita Atas Dasar Harga Berlaku, 2010-2016 - source_sentence: Arus dana Q3 2006 sentences: - Posisi Simpanan Berjangka Rupiah pada Bank Umum dan BPR Menurut Golongan Pemilik (miliar rupiah), 2005-2018 - Ringkasan Neraca Arus Dana, Triwulan III, 2006, (Miliar Rupiah) - Rata-Rata Pengeluaran per Kapita Sebulan di Daerah Perkotaan Menurut Kelompok Barang dan Golongan Pengeluaran per Kapita Sebulan, 2000-2012 datasets: - yahyaabd/query-hard-pos-neg-doc-pairs-statictable pipeline_tag: sentence-similarity library_name: sentence-transformers metrics: - cosine_accuracy - cosine_accuracy_threshold - cosine_f1 - cosine_f1_threshold - cosine_precision - cosine_recall - cosine_ap - cosine_mcc model-index: - name: SentenceTransformer based on sentence-transformers/paraphrase-multilingual-MiniLM-L12-v2 results: - task: type: binary-classification name: Binary Classification dataset: name: allstats semantic mini v1 test type: allstats-semantic-mini-v1_test metrics: - type: cosine_accuracy value: 0.8731520350428911 name: Cosine Accuracy - type: cosine_accuracy_threshold value: 0.6012464165687561 name: Cosine Accuracy Threshold - type: cosine_f1 value: 0.7941662080352229 name: Cosine F1 - type: cosine_f1_threshold value: 0.5294472575187683 name: Cosine F1 Threshold - type: cosine_precision value: 0.7762237762237763 name: Cosine Precision - type: cosine_recall value: 0.8129577464788732 name: Cosine Recall - type: cosine_ap value: 0.8789017077404396 name: Cosine Ap - type: cosine_mcc value: 0.6925319076430283 name: Cosine Mcc - task: type: binary-classification name: Binary Classification dataset: name: allstats semantic mini v1 dev type: allstats-semantic-mini-v1_dev metrics: - type: cosine_accuracy value: 0.9812009490782989 name: Cosine Accuracy - type: cosine_accuracy_threshold value: 0.7647212743759155 name: Cosine Accuracy Threshold - type: cosine_f1 value: 0.9711565387846541 name: Cosine F1 - type: cosine_f1_threshold value: 0.7647212743759155 name: Cosine F1 Threshold - type: cosine_precision value: 0.965478841870824 name: Cosine Precision - type: cosine_recall value: 0.9769014084507043 name: Cosine Recall - type: cosine_ap value: 0.994325385501373 name: Cosine Ap - type: cosine_mcc value: 0.9572502628719205 name: Cosine Mcc --- # SentenceTransformer based on sentence-transformers/paraphrase-multilingual-MiniLM-L12-v2 This is a [sentence-transformers](https://www.SBERT.net) model finetuned from [sentence-transformers/paraphrase-multilingual-MiniLM-L12-v2](https://huggingface.co/sentence-transformers/paraphrase-multilingual-MiniLM-L12-v2) on the [query-hard-pos-neg-doc-pairs-statictable](https://huggingface.co/datasets/yahyaabd/query-hard-pos-neg-doc-pairs-statictable) dataset. It maps sentences & paragraphs to a 384-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:** [sentence-transformers/paraphrase-multilingual-MiniLM-L12-v2](https://huggingface.co/sentence-transformers/paraphrase-multilingual-MiniLM-L12-v2) - **Maximum Sequence Length:** 128 tokens - **Output Dimensionality:** 384 dimensions - **Similarity Function:** Cosine Similarity - **Training Dataset:** - [query-hard-pos-neg-doc-pairs-statictable](https://huggingface.co/datasets/yahyaabd/query-hard-pos-neg-doc-pairs-statictable) ### 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': 128, 'do_lower_case': False}) with Transformer model: BertModel (1): Pooling({'word_embedding_dimension': 384, '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("yahyaabd/allstats-search-miniLM-v1-2") # Run inference sentences = [ 'Arus dana Q3 2006', 'Ringkasan Neraca Arus Dana, Triwulan III, 2006, (Miliar Rupiah)', 'Rata-Rata Pengeluaran per Kapita Sebulan di Daerah Perkotaan Menurut Kelompok Barang dan Golongan Pengeluaran per Kapita Sebulan, 2000-2012', ] embeddings = model.encode(sentences) print(embeddings.shape) # [3, 384] # Get the similarity scores for the embeddings similarities = model.similarity(embeddings, embeddings) print(similarities.shape) # [3, 3] ``` ## Evaluation ### Metrics #### Binary Classification * Datasets: `allstats-semantic-mini-v1_test` and `allstats-semantic-mini-v1_dev` * Evaluated with [BinaryClassificationEvaluator](https://sbert.net/docs/package_reference/sentence_transformer/evaluation.html#sentence_transformers.evaluation.BinaryClassificationEvaluator) | Metric | allstats-semantic-mini-v1_test | allstats-semantic-mini-v1_dev | |:--------------------------|:-------------------------------|:------------------------------| | cosine_accuracy | 0.8732 | 0.9812 | | cosine_accuracy_threshold | 0.6012 | 0.7647 | | cosine_f1 | 0.7942 | 0.9712 | | cosine_f1_threshold | 0.5294 | 0.7647 | | cosine_precision | 0.7762 | 0.9655 | | cosine_recall | 0.813 | 0.9769 | | **cosine_ap** | **0.8789** | **0.9943** | | cosine_mcc | 0.6925 | 0.9573 | ## Training Details ### Training Dataset #### query-hard-pos-neg-doc-pairs-statictable * Dataset: [query-hard-pos-neg-doc-pairs-statictable](https://huggingface.co/datasets/yahyaabd/query-hard-pos-neg-doc-pairs-statictable) at [7b28b96](https://huggingface.co/datasets/yahyaabd/query-hard-pos-neg-doc-pairs-statictable/tree/7b28b964daa3073a4d012d1ffca46ecd4f26bb5f) * Size: 25,580 training samples * Columns: query, doc, and label * Approximate statistics based on the first 1000 samples: | | query | doc | label | |:--------|:----------------------------------------------------------------------------------|:---------------------------------------------------------------------------------|:------------------------------------------------| | type | string | string | int | | details | | | | * Samples: | query | doc | label | |:-------------------------------------------------------------------------|:----------------------------------------------|:---------------| | Status pekerjaan utama penduduk usia 15+ yang bekerja, 2020 | Jumlah Penghuni Lapas per Kanwil | 0 | | status pekerjaan utama penduduk usia 15+ yang bekerja, 2020 | Jumlah Penghuni Lapas per Kanwil | 0 | | STATUS PEKERJAAN UTAMA PENDUDUK USIA 15+ YANG BEKERJA, 2020 | Jumlah Penghuni Lapas per Kanwil | 0 | * Loss: [OnlineContrastiveLoss](https://sbert.net/docs/package_reference/sentence_transformer/losses.html#onlinecontrastiveloss) ### Evaluation Dataset #### query-hard-pos-neg-doc-pairs-statictable * Dataset: [query-hard-pos-neg-doc-pairs-statictable](https://huggingface.co/datasets/yahyaabd/query-hard-pos-neg-doc-pairs-statictable) at [7b28b96](https://huggingface.co/datasets/yahyaabd/query-hard-pos-neg-doc-pairs-statictable/tree/7b28b964daa3073a4d012d1ffca46ecd4f26bb5f) * Size: 5,479 evaluation samples * Columns: query, doc, and label * Approximate statistics based on the first 1000 samples: | | query | doc | label | |:--------|:----------------------------------------------------------------------------------|:----------------------------------------------------------------------------------|:------------------------------------------------| | type | string | string | int | | details | | | | * Samples: | query | doc | label | |:-----------------------------------------------------------------------------------------|:----------------------------------------------------------------------------------------------------------------------------|:---------------| | Bagaimana perbandingan PNS pria dan wanita di berbagai golongan tahun 2014? | Rata-rata Pendapatan Bersih Berusaha Sendiri Menurut Provinsi dan Lapangan Pekerjaan Utama (ribu rupiah), 2017 | 0 | | bagaimana perbandingan pns pria dan wanita di berbagai golongan tahun 2014? | Rata-rata Pendapatan Bersih Berusaha Sendiri Menurut Provinsi dan Lapangan Pekerjaan Utama (ribu rupiah), 2017 | 0 | | BAGAIMANA PERBANDINGAN PNS PRIA DAN WANITA DI BERBAGAI GOLONGAN TAHUN 2014? | Rata-rata Pendapatan Bersih Berusaha Sendiri Menurut Provinsi dan Lapangan Pekerjaan Utama (ribu rupiah), 2017 | 0 | * Loss: [OnlineContrastiveLoss](https://sbert.net/docs/package_reference/sentence_transformer/losses.html#onlinecontrastiveloss) ### Training Hyperparameters #### Non-Default Hyperparameters - `eval_strategy`: steps - `per_device_train_batch_size`: 64 - `per_device_eval_batch_size`: 64 - `warmup_ratio`: 0.05 - `fp16`: True - `load_best_model_at_end`: True - `eval_on_start`: True #### All Hyperparameters
Click to expand - `overwrite_output_dir`: False - `do_predict`: False - `eval_strategy`: steps - `prediction_loss_only`: True - `per_device_train_batch_size`: 64 - `per_device_eval_batch_size`: 64 - `per_gpu_train_batch_size`: None - `per_gpu_eval_batch_size`: None - `gradient_accumulation_steps`: 1 - `eval_accumulation_steps`: None - `torch_empty_cache_steps`: None - `learning_rate`: 5e-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`: 3 - `max_steps`: -1 - `lr_scheduler_type`: linear - `lr_scheduler_kwargs`: {} - `warmup_ratio`: 0.05 - `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 - `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`: None - `hub_always_push`: False - `gradient_checkpointing`: False - `gradient_checkpointing_kwargs`: None - `include_inputs_for_metrics`: False - `include_for_metrics`: [] - `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 - `eval_on_start`: True - `use_liger_kernel`: False - `eval_use_gather_object`: False - `average_tokens_across_devices`: False - `prompts`: None - `batch_sampler`: batch_sampler - `multi_dataset_batch_sampler`: proportional
### Training Logs | Epoch | Step | Training Loss | Validation Loss | allstats-semantic-mini-v1_test_cosine_ap | allstats-semantic-mini-v1_dev_cosine_ap | |:-------:|:--------:|:-------------:|:---------------:|:----------------------------------------:|:---------------------------------------:| | -1 | -1 | - | - | 0.8789 | - | | 0 | 0 | - | 2.3102 | - | 0.8789 | | 0.05 | 20 | 1.7642 | 1.2458 | - | 0.9167 | | 0.1 | 40 | 0.9637 | 0.6870 | - | 0.9636 | | 0.15 | 60 | 0.4319 | 0.4890 | - | 0.9758 | | 0.2 | 80 | 0.3251 | 0.4944 | - | 0.9773 | | 0.25 | 100 | 0.2665 | 0.3988 | - | 0.9832 | | 0.3 | 120 | 0.1938 | 0.3795 | - | 0.9839 | | 0.35 | 140 | 0.1495 | 0.2839 | - | 0.9855 | | 0.4 | 160 | 0.0681 | 0.3011 | - | 0.9879 | | 0.45 | 180 | 0.1775 | 0.3116 | - | 0.9877 | | 0.5 | 200 | 0.0829 | 0.2536 | - | 0.9902 | | 0.55 | 220 | 0.2332 | 0.2887 | - | 0.9880 | | 0.6 | 240 | 0.1171 | 0.2862 | - | 0.9883 | | 0.65 | 260 | 0.1059 | 0.2467 | - | 0.9884 | | 0.7 | 280 | 0.1089 | 0.2240 | - | 0.9895 | | 0.75 | 300 | 0.0445 | 0.1772 | - | 0.9916 | | 0.8 | 320 | 0.0633 | 0.2392 | - | 0.9894 | | 0.85 | 340 | 0.0506 | 0.2440 | - | 0.9893 | | 0.9 | 360 | 0.1086 | 0.1926 | - | 0.9935 | | 0.95 | 380 | 0.064 | 0.2984 | - | 0.9900 | | 1.0 | 400 | 0.0478 | 0.2764 | - | 0.9897 | | 1.05 | 420 | 0.0508 | 0.2393 | - | 0.9911 | | 1.1 | 440 | 0.0266 | 0.2295 | - | 0.9912 | | 1.15 | 460 | 0.0236 | 0.2477 | - | 0.9899 | | 1.2 | 480 | 0.0142 | 0.2077 | - | 0.9919 | | 1.25 | 500 | 0.0128 | 0.1972 | - | 0.9921 | | 1.3 | 520 | 0.0205 | 0.2116 | - | 0.9912 | | 1.35 | 540 | 0.0447 | 0.2425 | - | 0.9904 | | 1.4 | 560 | 0.0 | 0.1999 | - | 0.9919 | | 1.45 | 580 | 0.0284 | 0.1989 | - | 0.9920 | | 1.5 | 600 | 0.0222 | 0.1789 | - | 0.9921 | | 1.55 | 620 | 0.0066 | 0.1957 | - | 0.9917 | | 1.6 | 640 | 0.0187 | 0.1993 | - | 0.9918 | | 1.65 | 660 | 0.0489 | 0.1901 | - | 0.9924 | | 1.7 | 680 | 0.0236 | 0.1556 | - | 0.9933 | | 1.75 | 700 | 0.0186 | 0.1597 | - | 0.9935 | | 1.8 | 720 | 0.0475 | 0.1813 | - | 0.9926 | | 1.85 | 740 | 0.0215 | 0.1689 | - | 0.9937 | | 1.9 | 760 | 0.0066 | 0.1746 | - | 0.9935 | | 1.95 | 780 | 0.0158 | 0.1808 | - | 0.9932 | | 2.0 | 800 | 0.0412 | 0.1799 | - | 0.9924 | | 2.05 | 820 | 0.0 | 0.1809 | - | 0.9923 | | 2.1 | 840 | 0.0072 | 0.1519 | - | 0.9936 | | 2.15 | 860 | 0.032 | 0.1538 | - | 0.9937 | | 2.2 | 880 | 0.0 | 0.1605 | - | 0.9934 | | 2.25 | 900 | 0.016 | 0.1812 | - | 0.9927 | | 2.3 | 920 | 0.0216 | 0.1550 | - | 0.9937 | | 2.35 | 940 | 0.0124 | 0.1533 | - | 0.9941 | | 2.4 | 960 | 0.0087 | 0.1499 | - | 0.9944 | | 2.45 | 980 | 0.0 | 0.1493 | - | 0.9945 | | **2.5** | **1000** | **0.0063** | **0.1483** | **-** | **0.9943** | | 2.55 | 1020 | 0.0 | 0.1505 | - | 0.9942 | | 2.6 | 1040 | 0.0 | 0.1508 | - | 0.9942 | | 2.65 | 1060 | 0.0 | 0.1508 | - | 0.9942 | | 2.7 | 1080 | 0.0 | 0.1508 | - | 0.9942 | | 2.75 | 1100 | 0.0191 | 0.1546 | - | 0.9942 | | 2.8 | 1120 | 0.0073 | 0.1566 | - | 0.9943 | | 2.85 | 1140 | 0.0095 | 0.1529 | - | 0.9943 | | 2.9 | 1160 | 0.0065 | 0.1512 | - | 0.9943 | | 2.95 | 1180 | 0.0 | 0.1508 | - | 0.9943 | | 3.0 | 1200 | 0.0 | 0.1508 | - | 0.9943 | * The bold row denotes the saved checkpoint. ### Framework Versions - Python: 3.10.12 - Sentence Transformers: 3.4.0 - Transformers: 4.48.1 - PyTorch: 2.5.1+cu124 - Accelerate: 1.3.0 - Datasets: 3.2.0 - Tokenizers: 0.21.0 ## 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", } ```