Bertimbau Base Portuguese Sts

🚀 juridics/bertimbau-base-portuguese-sts-scale

This is a sentence-transformers model that maps sentences and paragraphs to a 768-dimensional dense vector space, which can be used for tasks such as clustering or semantic search.

🚀 Quick Start

This model simplifies the process of sentence embedding generation. It maps sentences and paragraphs into a 768-dimensional dense vector space, which can be used for tasks such as clustering or semantic search.

✨ Features

• Dense Vector Mapping: Maps sentences and paragraphs to a 768-dimensional dense vector space.
• Versatile Applications: Suitable for tasks like clustering and semantic search.

📦 Installation

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

pip install -U sentence-transformers

💻 Usage Examples

Basic Usage

If you have sentence-transformers installed, you can use the model like this:

from sentence_transformers import SentenceTransformer
sentences = ["This is an example sentence", "Each sentence is converted"]

model = SentenceTransformer('juridics/bertimbau-base-portuguese-sts-scale')
embeddings = model.encode(sentences)
print(embeddings)

Advanced Usage

Without sentence-transformers, you can use the model by passing your input through the transformer model and then applying the right pooling-operation on top of the contextualized word embeddings:

from transformers import AutoTokenizer, AutoModel
import torch


#Mean Pooling - Take attention mask into account for correct averaging
def mean_pooling(model_output, attention_mask):
    token_embeddings = model_output[0] #First element of model_output contains all token embeddings
    input_mask_expanded = attention_mask.unsqueeze(-1).expand(token_embeddings.size()).float()
    return torch.sum(token_embeddings * input_mask_expanded, 1) / torch.clamp(input_mask_expanded.sum(1), min=1e-9)


# Sentences we want sentence embeddings for
sentences = ['This is an example sentence', 'Each sentence is converted']

# Load model from HuggingFace Hub
tokenizer = AutoTokenizer.from_pretrained('juridics/bertimbau-base-portuguese-sts-scale')
model = AutoModel.from_pretrained('juridics/bertimbau-base-portuguese-sts-scale')

# Tokenize sentences
encoded_input = tokenizer(sentences, padding=True, truncation=True, return_tensors='pt')

# Compute token embeddings
with torch.no_grad():
    model_output = model(**encoded_input)

# Perform pooling. In this case, mean pooling.
sentence_embeddings = mean_pooling(model_output, encoded_input['attention_mask'])

print("Sentence embeddings:")
print(sentence_embeddings)

📚 Documentation

Evaluation Results

For an automated evaluation of this model, see the Sentence Embeddings Benchmark: https://seb.sbert.net

Training

The model was trained with the following parameters:

DataLoader: torch.utils.data.dataloader.DataLoader of length 2492 with parameters:

{'batch_size': 8, '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": 3,
    "evaluation_steps": 2492,
    "evaluator": "sentence_transformers.evaluation.EmbeddingSimilarityEvaluator.EmbeddingSimilarityEvaluator",
    "max_grad_norm": 1,
    "optimizer_class": "<class 'torch.optim.adamw.AdamW'>",
    "optimizer_params": {
        "lr": 5e-05
    },
    "scheduler": "WarmupLinear",
    "steps_per_epoch": null,
    "warmup_steps": 748,
    "weight_decay": 0.01
}

Full Model Architecture

SentenceTransformer(
  (0): Transformer({'max_seq_length': 384, 'do_lower_case': False}) with Transformer model: BertModel 
  (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})
)

Citing & Authors

If you use our work, please cite:

@incollection{Viegas_2023,
	doi = {10.1007/978-3-031-36805-9_24},
	url = {https://doi.org/10.1007%2F978-3-031-36805-9_24},
	year = 2023,
	publisher = {Springer Nature Switzerland},
	pages = {349--365},
	author = {Charles F. O. Viegas and Bruno C. Costa and Renato P. Ishii},
	title = {{JurisBERT}: A New Approach that~Converts a~Classification Corpus into~an~{STS} One},
	booktitle = {Computational Science and Its Applications {\textendash} {ICCSA} 2023}
}