metadata

pipeline_tag: sentence-similarity
tags:
  - sentence-transformers
  - feature-extraction
  - sentence-similarity
language:
  - id

indoSBERT-large

This is a sentence-transformers model: It maps sentences & paragraphs to a 256 dimensional dense vector space and can be used for tasks like clustering or semantic search.

IndoSBERT is a modification of https://huggingface.co/indobenchmark/indobert-large-p1 that has been fine-tuned using the siamese network scheme inspired by SBERT (Reimers et al., 2019). This model was fine-tuned with the STS Dataset (2012-2016) which was machine-translated into Indonesian languange.

This model can provide meaningful semantic sentence embeddings for Indonesian sentences.

Usage (Sentence-Transformers)

Using this model becomes easy when you have sentence-transformers installed:

pip install -U sentence-transformers

Then you can use the model like this:

from sentence_transformers import SentenceTransformer
sentences = ["Komposer favorit saya adalah Joe Hisaishi", "Sapo tahu enak banget"]

model = SentenceTransformer('denaya/indoSBERT-large')
embeddings = model.encode(sentences)
print(embeddings)

Training

The model was trained with the parameters:

DataLoader:

torch.utils.data.dataloader.DataLoader of length 1291 with parameters:

{'batch_size': 16, 'sampler': 'torch.utils.data.sampler.RandomSampler', 'batch_sampler': 'torch.utils.data.sampler.BatchSampler'}

Loss:

sentence_transformers.losses.CosineSimilarityLoss.CosineSimilarityLoss

Parameters of the fit()-Method:

{
    "epochs": 50,
    "evaluation_steps": 1,
    "evaluator": "sentence_transformers.evaluation.EmbeddingSimilarityEvaluator.EmbeddingSimilarityEvaluator",
    "max_grad_norm": 1,
    "optimizer_class": "<class 'torch.optim.adamw.AdamW'>",
    "optimizer_params": {
        "lr": 2e-05
    },
    "scheduler": "WarmupLinear",
    "steps_per_epoch": null,
    "warmup_steps": 100,
    "weight_decay": 0.01
}

Full Model Architecture

SentenceTransformer(
  (0): Transformer({'max_seq_length': 256, 'do_lower_case': False}) with Transformer model: BertModel 
  (1): Pooling({'word_embedding_dimension': 1024, 'pooling_mode_cls_token': False, 'pooling_mode_mean_tokens': True, 'pooling_mode_max_tokens': False, 'pooling_mode_mean_sqrt_len_tokens': False})
  (2): Dense({'in_features': 1024, 'out_features': 256, 'bias': True, 'activation_function': 'torch.nn.modules.activation.Tanh'})
)

Citing & Authors

@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",
}

@article{author = {Diana, Denaya}, 
    title = {IndoSBERT: Indonesian SBERT for Semantic Textual Similarity tasks}, 
    year = {2023}, 
    url = {https://huggingface.co/denaya/indoSBERT-large}
}