yahyaabd/allstats-search-mini-v1-1-mnrl-v2

SentenceTransformer based on yahyaabd/allstats-search-mini-v1-1-mnrl

This is a sentence-transformers model finetuned from yahyaabd/allstats-search-mini-v1-1-mnrl on the bps-pub-cosine-pairs 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: yahyaabd/allstats-search-mini-v1-1-mnrl
• Maximum Sequence Length: 128 tokens
• Output Dimensionality: 384 dimensions
• Similarity Function: Cosine Similarity
• Training Dataset:
  • bps-pub-cosine-pairs

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-mini-v1-1-mnrl-v2")
# Run inference
sentences = [
    'q-4068',
    'Berapa persentase rumah tangga dengan akses sanitasi layak?',
    '43a5856225b1ff1cb95e319a',
]
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

Semantic Similarity

• Datasets: sts-dev and sts-test
• Evaluated with EmbeddingSimilarityEvaluator

Metric	sts-dev	sts-test
pearson_cosine	0.9259	0.9299
spearman_cosine	0.8465	0.8497

Training Details

Training Dataset

bps-pub-cosine-pairs

• Dataset: bps-pub-cosine-pairs at d58662e
• Size: 8,082 training samples
• Columns: query_id, query, corpus_id, title, and score
• Approximate statistics based on the first 1000 samples:

  	query_id	query	corpus_id	title	score
  type	string	string	string	string	float
  details	min: 4 tokens mean: 5.21 tokens max: 6 tokens	min: 4 tokens mean: 11.04 tokens max: 30 tokens	min: 4 tokens mean: 17.4 tokens max: 23 tokens	min: 5 tokens mean: 13.02 tokens max: 43 tokens	min: 0.1 mean: 0.55 max: 0.9

• Samples:

  query_id	query	corpus_id	title	score
  q-1599	Nilai Tukar Nelayan	0b0da8fc2b6af9329a6d9cfe	Statistik Hotel dan Akomodasi Lainnya di Indonesia 2013	0.1
  q-3595	Berapa angka statistik pertambangan non migas Indonesia periode 2012?	3c83610c3e2e5242177e2b11	Statistik Pertambangan Non Minyak dan Gas Bumi 2011-2015	0.9
  q-9112	Bagaimana situasi angkatan kerja Indonesia di bulan Februari 2021?	b547a5642aeb04d071cb83d4	Keadaan Angkatan Kerja di Indonesia Februari 2021	0.9

• Loss: CosineSimilarityLoss with these parameters:

  {
      "loss_fct": "torch.nn.modules.loss.MSELoss"
  }
  

Evaluation Dataset

bps-pub-cosine-pairs

• Dataset: bps-pub-cosine-pairs at d58662e
• Size: 1,010 evaluation samples
• Columns: query_id, query, corpus_id, title, and score
• Approximate statistics based on the first 1000 samples:

  	query_id	query	corpus_id	title	score
  type	string	string	string	string	float
  details	min: 4 tokens mean: 5.22 tokens max: 6 tokens	min: 4 tokens mean: 11.19 tokens max: 31 tokens	min: 7 tokens mean: 17.25 tokens max: 23 tokens	min: 5 tokens mean: 13.24 tokens max: 44 tokens	min: 0.1 mean: 0.56 max: 0.9

• Samples:

  query_id	query	corpus_id	title	score
  q-1273	Sosek Desember 2021	b7890a143bc751d1d84dcf4a	Laporan Bulanan Data Sosial Ekonomi Desember 2021	0.9
  q-4882	Ekspor Indonesia menurut SITC 2019-2020	9f3d9054c2f29bc478d56cd1	Statistik Perdagangan Luar Negeri Indonesia Ekspor Menurut Kode SITC, 2019-2020	0.9
  q-7141	Pengeluaran konsumsi penduduk Indonesia Maret 2018	4194e924ca33f087b68ab2de	Pengeluaran untuk Konsumsi Penduduk Indonesia, Maret 2018	0.9

• Loss: CosineSimilarityLoss with these parameters:

  {
      "loss_fct": "torch.nn.modules.loss.MSELoss"
  }
  

Training Hyperparameters

Non-Default Hyperparameters

• eval_strategy: steps
• per_device_train_batch_size: 32
• per_device_eval_batch_size: 32
• learning_rate: 1e-05
• warmup_ratio: 0.1
• fp16: True
• load_best_model_at_end: True
• label_smoothing_factor: 0.01
• 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: 1e-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: 0.1
• 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: 0.01
• 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	sts-dev_spearman_cosine	sts-test_spearman_cosine
0	0	-	0.3773	0.8467	-
0.0395	10	0.3676	0.3628	0.8469	-
0.0791	20	0.3166	0.3161	0.8474	-
0.1186	30	0.2743	0.2423	0.8483	-
0.1581	40	0.1933	0.1625	0.8494	-
0.1976	50	0.1473	0.1154	0.8507	-
0.2372	60	0.1046	0.1020	0.8514	-
0.2767	70	0.0839	0.0878	0.8519	-
0.3162	80	0.0839	0.0759	0.8519	-
0.3557	90	0.0756	0.0667	0.8521	-
0.3953	100	0.0611	0.0597	0.8522	-
0.4348	110	0.0562	0.0554	0.8520	-
0.4743	120	0.0557	0.0518	0.8517	-
0.5138	130	0.0411	0.0482	0.8514	-
0.5534	140	0.0481	0.0454	0.8510	-
0.5929	150	0.0474	0.0423	0.8500	-
0.6324	160	0.0433	0.0404	0.8498	-
0.6719	170	0.0389	0.0390	0.8502	-
0.7115	180	0.0423	0.0373	0.8503	-
0.7510	190	0.0348	0.0360	0.8495	-
0.7905	200	0.0404	0.0346	0.8492	-
0.8300	210	0.0285	0.0334	0.8494	-
0.8696	220	0.0322	0.0317	0.8482	-
0.9091	230	0.0311	0.0305	0.8469	-
0.9486	240	0.027	0.0298	0.8462	-
0.9881	250	0.03	0.0292	0.8462	-
1.0277	260	0.0245	0.0292	0.8458	-
1.0672	270	0.026	0.0290	0.8447	-
1.1067	280	0.0325	0.0279	0.8466	-
1.1462	290	0.0208	0.0274	0.8458	-
1.1858	300	0.0249	0.0271	0.8451	-
1.2253	310	0.026	0.0264	0.8444	-
1.2648	320	0.0234	0.0261	0.8469	-
1.3043	330	0.024	0.0267	0.8482	-
1.3439	340	0.0212	0.0254	0.8480	-
1.3834	350	0.033	0.0247	0.8473	-
1.4229	360	0.0246	0.0244	0.8473	-
1.4625	370	0.0241	0.0242	0.8477	-
1.5020	380	0.0187	0.0237	0.8473	-
1.5415	390	0.0228	0.0235	0.8474	-
1.5810	400	0.0169	0.0234	0.8475	-
1.6206	410	0.0249	0.0233	0.8470	-
1.6601	420	0.0223	0.0234	0.8475	-
1.6996	430	0.0174	0.0232	0.8477	-
1.7391	440	0.0249	0.0229	0.8480	-
1.7787	450	0.0243	0.0229	0.8483	-
1.8182	460	0.0203	0.0232	0.8485	-
1.8577	470	0.0198	0.0226	0.8477	-
1.8972	480	0.019	0.0223	0.8464	-
1.9368	490	0.0185	0.0218	0.8465	-
1.9763	500	0.0168	0.0218	0.8468	-
2.0158	510	0.019	0.0217	0.8472	-
2.0553	520	0.0194	0.0217	0.8476	-
2.0949	530	0.0192	0.0216	0.8475	-
2.1344	540	0.0175	0.0215	0.8473	-
2.1739	550	0.013	0.0214	0.8477	-
2.2134	560	0.017	0.0212	0.8478	-
2.2530	570	0.0157	0.0212	0.8478	-
2.2925	580	0.0169	0.0211	0.8473	-
2.3320	590	0.0192	0.0210	0.8475	-
2.3715	600	0.0116	0.0208	0.8472	-
2.4111	610	0.0151	0.0207	0.8473	-
2.4506	620	0.0182	0.0205	0.8472	-
2.4901	630	0.0143	0.0205	0.8471	-
2.5296	640	0.0193	0.0204	0.8470	-
2.5692	650	0.0194	0.0203	0.8469	-
2.6087	660	0.0132	0.0204	0.8469	-
2.6482	670	0.0208	0.0204	0.8464	-
2.6877	680	0.0155	0.0203	0.8461	-
2.7273	690	0.0142	0.0203	0.8461	-
2.7668	700	0.0162	0.0203	0.8460	-
2.8063	710	0.0198	0.0203	0.8461	-
2.8458	720	0.0138	0.0204	0.8465	-
2.8854	730	0.0145	0.0204	0.8465	-
2.9249	740	0.0129	0.0204	0.8466	-
2.9644	750	0.0108	0.0204	0.8465	-
-1	-1	-	-	-	0.8497

• 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",
}