Flaubert Base Uncased Xnli Sts

🚀 inokufu/flaubert-base-uncased-xnli-sts

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

🚀 Quick Start

This model is based on the French flaubert - base - uncased pre - trained model [1, 2]. It was first fine - tuned on a natural language inference task (XNLI) [3], which aims to train the model to recognize relations between sentences (contradiction, neutral, implication). Then, it was fine - tuned on a text semantic similarity task (on STS - fr data) [4], training the model to estimate the similarity between two sentences. This fine - tuning process enables the model to have a much better semantic representation of words than the base model.

✨ Features

• Maps sentences and paragraphs to a 768 - dimensional dense vector space.
• Suitable for tasks like clustering and semantic search.
• Fine - tuned on natural language inference and text semantic similarity tasks.

📦 Installation

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

pip install -U sentence-transformers

💻 Usage Examples

Basic Usage

When sentence-transformers is installed, you can use the model as follows:

from sentence_transformers import SentenceTransformer
sentences = ["Apprendre le python", "Devenir expert en comptabilité"]

model = SentenceTransformer('inokufu/flaubert-base-uncased-xnli-sts')
embeddings = model.encode(sentences)
print(embeddings)

Advanced Usage

Without sentence-transformers, you can use the model like this: First, pass your input through the transformer model, then apply 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 = ["Apprendre le python", "Devenir expert en comptabilité"]

# Load model from HuggingFace Hub
tokenizer = AutoTokenizer.from_pretrained('inokufu/flaubert-base-uncased-xnli-sts')
model = AutoModel.from_pretrained('inokufu/flaubert-base-uncased-xnli-sts')

# 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

STS (fr) score: 83.07%

Model Architecture

SentenceTransformer(
  (0): Transformer({'max_seq_length': 512, 'do_lower_case': True}) with Transformer model: FlaubertModel
  (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})
)

📄 License

No license information provided in the original document.

🔧 Technical Details

The model is based on the French flaubert - base - uncased pre - trained model. Through fine - tuning on the XNLI natural language inference task and the STS - fr text semantic similarity task, it can better understand the semantic relationships between sentences. The fine - tuning process adjusts the model's parameters to adapt to specific tasks, resulting in a more accurate semantic representation of words.

References

[1] https://hal.archives - ouvertes.fr/hal - 02784776v3/document
[2] https://huggingface.co/flaubert/flaubert_base_uncased
[3] https://arxiv.org/abs/1809.05053
[4] https://huggingface.co/datasets/stsb_multi_mt