yahyaabd/allstats-semantic-base-v1

SentenceTransformer based on sentence-transformers/paraphrase-multilingual-mpnet-base-v2

This is a sentence-transformers model finetuned from sentence-transformers/paraphrase-multilingual-mpnet-base-v2 on the allstats-semantic-synthetic-dataset-v1 dataset. 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: sentence-transformers/paraphrase-multilingual-mpnet-base-v2
• Maximum Sequence Length: 128 tokens
• Output Dimensionality: 768 dimensions
• Similarity Function: Cosine Similarity
• Training Dataset:
  • allstats-semantic-synthetic-dataset-v1

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: XLMRobertaModel 
  (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:

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-semantic-base-v1")
# Run inference
sentences = [
    'analisis kinerja ekspor indonesia feb 2014',
    'Buletin Statistik Perdagangan Luar Negeri Ekspor Menurut Kelompok Komoditi dan Negara Februari 2014',
    'Laporan Bulanan Data Sosial Ekonomi Januari 2019',
]
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]

Evaluation

Metrics

Semantic Similarity

• Datasets: allstats-semantic-base-v1-eval and allstat-semantic-base-v1-test
• Evaluated with EmbeddingSimilarityEvaluator

Metric	allstats-semantic-base-v1-eval	allstat-semantic-base-v1-test
pearson_cosine	0.9866	0.9877
spearman_cosine	0.9033	0.9063

Training Details

Training Dataset

allstats-semantic-synthetic-dataset-v1

• Dataset: allstats-semantic-synthetic-dataset-v1 at d59a245
• Size: 123,640 training samples
• Columns: query, doc, and label
• Approximate statistics based on the first 1000 samples:

  	query	doc	label
  type	string	string	float
  details	min: 6 tokens mean: 10.64 tokens max: 34 tokens	min: 4 tokens mean: 14.06 tokens max: 76 tokens	min: 0.0 mean: 0.49 max: 1.0

• Samples:

  query	doc	label
  Gambaran umum karakteristik usaha di Indonesia	Statistik Karakteristik Usaha 2022/2023	0.9
  Tabel data jumlah sekolah, guru, dan murid MA di bawah Kementerian Agama per provinsi.	Jumlah Sekolah, Guru, dan Murid Madrasah Aliyah (MA) di Bawah Kementerian Agama Menurut Provinsi, tahun ajaran 2005/2006-2015/2016	0.96
  bagaimana kinerja sektor konstruksi indonesia di triwulan ketiga tahun 2008?	Statistik Restoran/Rumah Makan 2007	0.09

• Loss: CosineSimilarityLoss with these parameters:

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

Evaluation Dataset

allstats-semantic-synthetic-dataset-v1

• Dataset: allstats-semantic-synthetic-dataset-v1 at d59a245
• Size: 26,494 evaluation samples
• Columns: query, doc, and label
• Approximate statistics based on the first 1000 samples:

  	query	doc	label
  type	string	string	float
  details	min: 5 tokens mean: 10.48 tokens max: 34 tokens	min: 5 tokens mean: 13.86 tokens max: 58 tokens	min: 0.0 mean: 0.49 max: 1.0

• Samples:

  query	doc	label
  Harga barang konsumsi Indonesia 2022: data per kota	Harga Konsumen Beberapa Barang Kelompok Makanan, Minuman, dan Tembakau 90 Kota di Indonesia 2022	0.92
  data biaya hidup bali 2018	Buletin Statistik Perdagangan Luar Negeri Ekspor Menurut Kelompok Komoditi dan Negara, Maret 2018	0.1
  ekspor barang indonesia november 2011: data lengkap	Buletin Statistik Perdagangan Luar Negeri Ekspor Menurut Kelompok Komoditi dan Negara Februari 2013	0.12

• 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
• num_train_epochs: 10
• warmup_ratio: 0.1
• fp16: True
• load_best_model_at_end: 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: 10
• 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: False
• 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-base-v1-eval_spearman_cosine	allstat-semantic-base-v1-test_spearman_cosine
0.1294	500	0.0454	0.0267	0.7374	-
0.2588	1000	0.0243	0.0205	0.7527	-
0.3882	1500	0.0199	0.0169	0.7720	-
0.5176	2000	0.0186	0.0164	0.7733	-
0.6470	2500	0.0179	0.0158	0.7806	-
0.7764	3000	0.0158	0.0155	0.7826	-
0.9058	3500	0.0159	0.0155	0.7771	-
1.0352	4000	0.0155	0.0143	0.7847	-
1.1646	4500	0.0133	0.0141	0.7935	-
1.2940	5000	0.0128	0.0132	0.7986	-
1.4234	5500	0.0121	0.0120	0.8148	-
1.5528	6000	0.012	0.0118	0.8030	-
1.6822	6500	0.0118	0.0121	0.8132	-
1.8116	7000	0.0119	0.0109	0.8130	-
1.9410	7500	0.0107	0.0108	0.8132	-
2.0704	8000	0.009	0.0098	0.8181	-
2.1998	8500	0.0082	0.0099	0.8221	-
2.3292	9000	0.008	0.0100	0.8221	-
2.4586	9500	0.008	0.0095	0.8302	-
2.5880	10000	0.0083	0.0090	0.8284	-
2.7174	10500	0.0084	0.0093	0.8261	-
2.8468	11000	0.0084	0.0089	0.8283	-
2.9762	11500	0.0083	0.0093	0.8259	-
3.1056	12000	0.0056	0.0083	0.8362	-
3.2350	12500	0.006	0.0081	0.8357	-
3.3644	13000	0.0057	0.0078	0.8381	-
3.4938	13500	0.006	0.0081	0.8399	-
3.6232	14000	0.0058	0.0078	0.8420	-
3.7526	14500	0.0068	0.0078	0.8303	-
3.8820	15000	0.0056	0.0072	0.8502	-
4.0114	15500	0.0054	0.0073	0.8483	-
4.1408	16000	0.004	0.0068	0.8565	-
4.2702	16500	0.0042	0.0069	0.8493	-
4.3996	17000	0.0043	0.0069	0.8507	-
4.5290	17500	0.0045	0.0069	0.8536	-
4.6584	18000	0.0042	0.0064	0.8602	-
4.7878	18500	0.0043	0.0065	0.8537	-
4.9172	19000	0.0039	0.0062	0.8623	-
5.0466	19500	0.0041	0.0065	0.8601	-
5.1760	20000	0.0032	0.0060	0.8643	-
5.3054	20500	0.0032	0.0064	0.8657	-
5.4348	21000	0.0032	0.0062	0.8669	-
5.5642	21500	0.0031	0.0065	0.8633	-
5.6936	22000	0.003	0.0059	0.8682	-
5.8230	22500	0.0032	0.0057	0.8713	-
5.9524	23000	0.0032	0.0057	0.8688	-
6.0818	23500	0.0026	0.0055	0.8772	-
6.2112	24000	0.0023	0.0056	0.8708	-
6.3406	24500	0.0029	0.0056	0.8734	-
6.4700	25000	0.0027	0.0054	0.8748	-
6.5994	25500	0.0022	0.0054	0.8827	-
6.7288	26000	0.0021	0.0053	0.8823	-
6.8582	26500	0.0021	0.0053	0.8832	-
6.9876	27000	0.0025	0.0052	0.8839	-
7.1170	27500	0.002	0.0051	0.8887	-
7.2464	28000	0.0017	0.0050	0.8869	-
7.3758	28500	0.0019	0.0052	0.8845	-
7.5052	29000	0.0017	0.0051	0.8897	-
7.6346	29500	0.0017	0.0051	0.8920	-
7.7640	30000	0.0018	0.0050	0.8889	-
7.8934	30500	0.0019	0.0050	0.8931	-
8.0228	31000	0.002	0.0049	0.8889	-
8.1522	31500	0.0014	0.0049	0.8912	-
8.2816	32000	0.0013	0.0049	0.8922	-
8.4110	32500	0.0014	0.0049	0.8947	-
8.5404	33000	0.0014	0.0049	0.8960	-
8.6698	33500	0.0014	0.0049	0.8972	-
8.7992	34000	0.0014	0.0048	0.8982	-
8.9286	34500	0.0013	0.0048	0.9003	-
9.0580	35000	0.0014	0.0048	0.9001	-
9.1874	35500	0.0012	0.0048	0.8995	-
9.3168	36000	0.0011	0.0048	0.9008	-
9.4462	36500	0.001	0.0047	0.9015	-
9.5756	37000	0.0011	0.0047	0.9026	-
9.7050	37500	0.0011	0.0047	0.9027	-
9.8344	38000	0.001	0.0047	0.9035	-
9.9638	38500	0.0011	0.0047	0.9033	-
10.0	38640	-	-	-	0.9063

• The bold row denotes the saved checkpoint.

Framework Versions

• Python: 3.10.12
• Sentence Transformers: 3.3.1
• Transformers: 4.47.1
• PyTorch: 2.5.1+cu124
• Accelerate: 1.2.1
• 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",
}