TiELECTRA BiEncoder Model

This model is a sentence-transformers model for the Tigrinya language based on TiELECTRA-small. The maps sentences & paragraphs to a 256 dimensional dense vector space and can be used for tasks like clustering or semantic search.

Using Sentence-Transformers

Using this model becomes easy when you have sentence-transformersinstalled:

pip install -U sentence-transformers

Then use the model as follows:

from sentence_transformers import SentenceTransformer
sentences = ["ሓደ ሰብኣይ ፈረስ ይጋልብ ኣሎ።", "ሓንቲ ጓል ክራር ትጻወት ኣላ።"]

model = SentenceTransformer('fgaim/tielectra-bi-encoder')
embeddings = model.encode(sentences)
print(embeddings)

Using 🤗 Transformers

Use the transformers library as follows: Pass the input through the transformer model, then you have to apply the right pooling-operation on-top of the contextualized word embeddings.

import torch
from transformers import AutoModel, AutoTokenizer


# 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 = ["ሓደ ሰብኣይ ፈረስ ይጋልብ ኣሎ።", "ሓንቲ ጓል ክራር ትጻወት ኣላ።"]

# Load model from HuggingFace Hub
tokenizer = AutoTokenizer.from_pretrained("fgaim/tielectra-bi-encoder")
model = AutoModel.from_pretrained("fgaim/tielectra-bi-encoder")

# 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:", sentence_embeddings)

Architecture

Base Model

The model properties:

Model Size	Layers	Attn. Heads	Hidden Size	FFN	Parameters	Max. Seq
SMALL	12	4	256	1024	14M	512

BiEncoder Model

Max Seq Length: 512
Word embedding dimension: 256

SentenceTransformer(
  (0): Transformer({'max_seq_length': 512, 'do_lower_case': False}) with Transformer model: ElectraModel

  (1): Pooling({'word_embedding_dimension': 256, '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})
)

Cite

If you use this model in your product or research, you can cite it as follows:

@article{Fitsum2021TiPLMs,
  author={Fitsum Gaim and Wonsuk Yang and Jong C. Park},
  title={Monolingual Pre-trained Language Models for Tigrinya},
  year=2021,
  publisher={WiNLP 2021 co-located EMNLP 2021}
}