yahyaabd/allstats-search-miniLM-v1-7

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-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-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-v1-7")
# Run inference
sentences = [
    'Aliran dana Rupiah: Q1 2008',
    'IHK dan Rata-rata Upah per Bulan Buruh Industri di Bawah Mandor (Supervisor), 2012-2014 (2012=100)',
    'Ringkasan Neraca Arus Dana, 2012 (Miliar Rupiah)',
]
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	0.9679	0.9678
cosine_accuracy_threshold	0.7482	0.7902
cosine_f1	0.9678	0.9674
cosine_f1_threshold	0.7444	0.7875
cosine_precision	0.9596	0.9617
cosine_recall	0.9762	0.9731
cosine_ap	0.9922	0.993
cosine_mcc	0.9359	0.9357

Training Details

Training Dataset

query-pos-neg-doc-pairs-statictable

• Dataset: query-pos-neg-doc-pairs-statictable at a31b58d
• Size: 110,773 training samples
• Columns: query, doc, and label
• Approximate statistics based on the first 1000 samples:

  	query	doc	label
  type	string	string	int
  details	min: 9 tokens mean: 21.22 tokens max: 50 tokens	min: 6 tokens mean: 28.24 tokens max: 50 tokens	0: ~43.90% 1: ~56.10%

• Samples:

  query	doc	label
  Data orang yang naik/turun kapal, di pelabuhan yang dikelola maupun tidak, sekitar 2015	Tabel Input-Output Indonesia Transaksi Total Atas Dasar Harga Dasar (185 Produk), 2016 (Juta Rupiah)	0
  data orang yang naik/turun kapal, di pelabuhan yang dikelola maupun tidak, sekitar 2015	Tabel Input-Output Indonesia Transaksi Total Atas Dasar Harga Dasar (185 Produk), 2016 (Juta Rupiah)	0
  DATA ORANG YANG NAIK/TURUN KAPAL, DI PELABUHAN YANG DIKELOLA MAUPUN TIDAK, SEKITAR 2015	Tabel Input-Output Indonesia Transaksi Total Atas Dasar Harga Dasar (185 Produk), 2016 (Juta Rupiah)	0

• Loss: ContrastiveLoss with these parameters:

  {
      "distance_metric": "SiameseDistanceMetric.COSINE_DISTANCE",
      "margin": 0.5,
      "size_average": true
  }
  

Evaluation Dataset

query-pos-neg-doc-pairs-statictable

• Dataset: query-pos-neg-doc-pairs-statictable at a31b58d
• Size: 23,763 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: 20.75 tokens max: 57 tokens	min: 6 tokens mean: 27.44 tokens max: 43 tokens	0: ~50.20% 1: ~49.80%

• Samples:

  query	doc	label
  Cek penghasilan bulanan (gaji bersih) buruh/pegawai, per provinsi dan jenis pekerjaannya, 2019	Rata-rata Pendapatan Bersih Berusaha Sendiri Menurut Provinsi dan Lapangan Pekerjaan Utama, 2021	1
  cek penghasilan bulanan (gaji bersih) buruh/pegawai, per provinsi dan jenis pekerjaannya, 2019	Rata-rata Pendapatan Bersih Berusaha Sendiri Menurut Provinsi dan Lapangan Pekerjaan Utama, 2021	1
  CEK PENGHASILAN BULANAN (GAJI BERSIH) BURUH/PEGAWAI, PER PROVINSI DAN JENIS PEKERJAANNYA, 2019	Rata-rata Pendapatan Bersih Berusaha Sendiri Menurut Provinsi dan Lapangan Pekerjaan Utama, 2021	1

• Loss: ContrastiveLoss with these parameters:

  {
      "distance_metric": "SiameseDistanceMetric.COSINE_DISTANCE",
      "margin": 0.5,
      "size_average": true
  }
  

Training Hyperparameters

Non-Default Hyperparameters

• eval_strategy: steps
• per_device_train_batch_size: 64
• per_device_eval_batch_size: 64
• num_train_epochs: 1
• warmup_ratio: 0.2
• 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: 1
• max_steps: -1
• lr_scheduler_type: linear
• lr_scheduler_kwargs: {}
• warmup_ratio: 0.2
• 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	-	-	0.8699	-
0	0	-	0.0489	-	0.8658
0.0578	100	0.0222	0.0101	-	0.9458
0.1155	200	0.0087	0.0073	-	0.9631
0.1733	300	0.007	0.0059	-	0.9710
0.2311	400	0.0056	0.0049	-	0.9828
0.2889	500	0.0045	0.0044	-	0.9837
0.3466	600	0.0042	0.0041	-	0.9862
0.4044	700	0.0038	0.0038	-	0.9888
0.4622	800	0.0037	0.0037	-	0.9890
0.5199	900	0.0029	0.0036	-	0.9889
0.5777	1000	0.0031	0.0034	-	0.9907
0.6355	1100	0.0029	0.0033	-	0.9923
0.6932	1200	0.0025	0.0034	-	0.9922
0.7510	1300	0.0025	0.0033	-	0.9929
0.8088	1400	0.0024	0.0033	-	0.9928
0.8666	1500	0.0022	0.0033	-	0.9926
0.9243	1600	0.0023	0.0033	-	0.9929
0.9821	1700	0.0022	0.0032	-	0.993
-1	-1	-	-	0.9922	-

• 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",
}

ContrastiveLoss

@inproceedings{hadsell2006dimensionality,
    author={Hadsell, R. and Chopra, S. and LeCun, Y.},
    booktitle={2006 IEEE Computer Society Conference on Computer Vision and Pattern Recognition (CVPR'06)},
    title={Dimensionality Reduction by Learning an Invariant Mapping},
    year={2006},
    volume={2},
    number={},
    pages={1735-1742},
    doi={10.1109/CVPR.2006.100}
}