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

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

Training Logs

• 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",
}