yahyaabd/allstats-search-miniLM-v1-4

SentenceTransformer based on sentence-transformers/paraphrase-multilingual-MiniLM-L12-v2

This is a sentence-transformers model finetuned from sentence-transformers/paraphrase-multilingual-MiniLM-L12-v2 on the 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
• Maximum Sequence Length: 128 tokens
• Output Dimensionality: 384 dimensions
• Similarity Function: Cosine Similarity
• Training Dataset:
  • query-hard-pos-neg-doc-pairs-statictable

Model Sources

• Documentation: Sentence Transformers Documentation
• Repository: Sentence Transformers on GitHub
• Hugging Face: Sentence Transformers on Hugging Face

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:

pip install -U sentence-transformers

Then you can load this model and run inference.

from sentence_transformers import SentenceTransformer

# Download from the 🤗 Hub
model = SentenceTransformer("yahyaabd/allstats-search-miniLM-v1-4")
# 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

Metric	allstats-semantic-mini-v1_test	allstats-semantic-mini-v1_dev
cosine_accuracy	0.9739	0.9695
cosine_accuracy_threshold	0.7544	0.7802
cosine_f1	0.9602	0.9532
cosine_f1_threshold	0.754	0.7692
cosine_precision	0.9498	0.9437
cosine_recall	0.9707	0.9628
cosine_ap	0.9915	0.9911
cosine_mcc	0.9409	0.9305

Training Details

Training Dataset

query-hard-pos-neg-doc-pairs-statictable

• Dataset: query-hard-pos-neg-doc-pairs-statictable at 7b28b96
• 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	min: 7 tokens mean: 20.14 tokens max: 55 tokens	min: 5 tokens mean: 24.9 tokens max: 47 tokens	0: ~70.80% 1: ~29.20%

• 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

Evaluation Dataset

query-hard-pos-neg-doc-pairs-statictable

• Dataset: query-hard-pos-neg-doc-pairs-statictable at 7b28b96
• 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	min: 7 tokens mean: 20.78 tokens max: 52 tokens	min: 4 tokens mean: 26.28 tokens max: 43 tokens	0: ~71.50% 1: ~28.50%

• 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

Training Hyperparameters

Non-Default Hyperparameters

• eval_strategy: steps
• per_device_train_batch_size: 32
• per_device_eval_batch_size: 32
• num_train_epochs: 2
• warmup_ratio: 0.1
• 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: 32
• per_device_eval_batch_size: 32
• 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: 2
• max_steps: -1
• lr_scheduler_type: linear
• 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
• 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	-	1.0484	-	0.8789
0.025	20	0.9076	0.7143	-	0.8976
0.05	40	0.4666	0.4744	-	0.9234
0.075	60	0.4514	0.3208	-	0.9542
0.1	80	0.3153	0.2520	-	0.9666
0.125	100	0.1726	0.2074	-	0.9725
0.15	120	0.1056	0.1860	-	0.9750
0.175	140	0.1414	0.2540	-	0.9674
0.2	160	0.1091	0.2077	-	0.9747
0.225	180	0.108	0.2333	-	0.9690
0.25	200	0.1672	0.1618	-	0.9771
0.275	220	0.1086	0.1804	-	0.9775
0.3	240	0.083	0.1805	-	0.9760
0.325	260	0.083	0.1674	-	0.9709
0.35	280	0.1197	0.1735	-	0.9734
0.375	300	0.0811	0.1272	-	0.9805
0.4	320	0.049	0.1491	-	0.9791
0.425	340	0.0373	0.1651	-	0.9721
0.45	360	0.1116	0.1742	-	0.9756
0.475	380	0.0665	0.1175	-	0.9837
0.5	400	0.0698	0.1165	-	0.9841
0.525	420	0.1316	0.1353	-	0.9817
0.55	440	0.0753	0.1276	-	0.9824
0.575	460	0.0411	0.1353	-	0.9801
0.6	480	0.0099	0.1292	-	0.9811
0.625	500	0.0473	0.1118	-	0.9836
0.65	520	0.0201	0.1083	-	0.9836
0.675	540	0.0519	0.1089	-	0.9856
0.7	560	0.0652	0.1003	-	0.9875
0.725	580	0.0594	0.1201	-	0.9872
0.75	600	0.0536	0.0896	-	0.9893
0.775	620	0.0479	0.0855	-	0.9874
0.8	640	0.0301	0.0948	-	0.9876
0.825	660	0.014	0.0993	-	0.9883
0.85	680	0.0199	0.0930	-	0.9884
0.875	700	0.0375	0.0765	-	0.9918
0.9	720	0.0	0.0805	-	0.9916
0.925	740	0.0243	0.0816	-	0.9916
0.95	760	0.0209	0.0935	-	0.9896
0.975	780	0.02	0.0831	-	0.9897
1.0	800	0.0376	0.0849	-	0.9890
1.025	820	0.0113	0.0960	-	0.9883
1.05	840	0.01	0.1131	-	0.9868
1.075	860	0.0294	0.1069	-	0.9861
1.1	880	0.0367	0.0921	-	0.9899
1.125	900	0.0	0.0910	-	0.9898
1.15	920	0.0163	0.1122	-	0.9871
1.175	940	0.0072	0.1204	-	0.9852
1.2	960	0.0175	0.1047	-	0.9872
1.225	980	0.0065	0.0992	-	0.9882
1.25	1000	0.0104	0.0932	-	0.9890
1.275	1020	0.0281	0.0866	-	0.9897
1.3	1040	0.0169	0.0874	-	0.9899
1.325	1060	0.0069	0.0910	-	0.9904
1.35	1080	0.0	0.0983	-	0.9898
1.375	1100	0.0	0.0985	-	0.9897
1.4	1120	0.0146	0.0919	-	0.9904
1.425	1140	0.0075	0.0852	-	0.9908
1.45	1160	0.014	0.0845	-	0.9908
1.475	1180	0.0065	0.0816	-	0.9907
1.5	1200	0.0	0.0811	-	0.9907
1.525	1220	0.0103	0.0785	-	0.9910
1.55	1240	0.013	0.0721	-	0.9915
1.575	1260	0.0066	0.0793	-	0.9910
1.6	1280	0.0	0.0810	-	0.9909
1.625	1300	0.0239	0.0803	-	0.9912
1.65	1320	0.0155	0.0816	-	0.9908
1.675	1340	0.009	0.0859	-	0.9904
1.7	1360	0.0065	0.0855	-	0.9900
1.725	1380	0.0	0.0866	-	0.9899
1.75	1400	0.0127	0.0865	-	0.9907
1.775	1420	0.0064	0.0819	-	0.9909
1.8	1440	0.0	0.0828	-	0.9910
1.825	1460	0.0081	0.0818	-	0.9912
1.85	1480	0.0068	0.0875	-	0.9909
1.875	1500	0.0	0.0886	-	0.9909
1.9	1520	0.011	0.0846	-	0.9911
1.925	1540	0.0	0.0843	-	0.9911
1.95	1560	0.0	0.0843	-	0.9911
1.975	1580	0.0	0.0843	-	0.9911
2.0	1600	0.0162	0.0850	-	0.9911
-1	-1	-	-	0.9915	-

• 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

@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",
}