️ This model is deprecated. Please don’t use it as it produces sentence embeddings of low quality. You can find recommended sentence embedding models here: SBERT.net – Pretrained Models

sentence-transformers/distilbert-base-nli-mean-tokens

This is a sentence-transformers model: It maps sentences & paragraphs to a 768 dimensional dense vector space and can be used for tasks like clustering or semantic search.

Usage (Sentence-Transformers)

Using this model becomes easy when you have sentence-transformers installed:
pip install -U sentence-transformers<br />

Then you can use the model like this:
from sentence_transformers import SentenceTransformer<br /> sentences = ["This is an example sentence", "Each sentence is converted"]<br /> model = SentenceTransformer('sentence-transformers/distilbert-base-nli-mean-tokens')<br /> embeddings = model.encode(sentences)<br /> print(embeddings)<br />

Usage (HuggingFace Transformers)

Without sentence-transformers, you can use the model like this: First, you pass your input through the transformer model, then you have to apply the right pooling-operation on-top of the contextualized word embeddings.
from transformers import AutoTokenizer, AutoModel<br /> import torch<br /> #Mean Pooling - Take attention mask into account for correct averaging<br /> def mean_pooling(model_output, attention_mask):<br /> token_embeddings = model_output[0] #First element of model_output contains all token embeddings<br /> input_mask_expanded = attention_mask.unsqueeze(-1).expand(token_embeddings.size()).float()<br /> return torch.sum(token_embeddings * input_mask_expanded, 1) / torch.clamp(input_mask_expanded.sum(1), min=1e-9)<br /> # Sentences we want sentence embeddings for<br /> sentences = ['This is an example sentence', 'Each sentence is converted']<br /> # Load model from HuggingFace Hub<br /> tokenizer = AutoTokenizer.from_pretrained('sentence-transformers/distilbert-base-nli-mean-tokens')<br /> model = AutoModel.from_pretrained('sentence-transformers/distilbert-base-nli-mean-tokens')<br /> # Tokenize sentences<br /> encoded_input = tokenizer(sentences, padding=True, truncation=True, return_tensors='pt')<br /> # Compute token embeddings<br /> with torch.no_grad():<br /> model_output = model(**encoded_input)<br /> # Perform pooling. In this case, max pooling.<br /> sentence_embeddings = mean_pooling(model_output, encoded_input['attention_mask'])<br /> print("Sentence embeddings:")<br /> print(sentence_embeddings)<br />

Evaluation Results

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

Full Model Architecture

SentenceTransformer(<br /> (0): Transformer({'max_seq_length': 128, 'do_lower_case': False}) with Transformer model: DistilBertModel<br /> (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})<br /> )<br />

Citing & Authors

This model was trained by sentence-transformers.
If you find this model helpful, feel free to cite our publication Sentence-BERT: Sentence Embeddings using Siamese BERT-Networks:
@inproceedings{reimers-2019-sentence-bert,<br /> title = "Sentence-BERT: Sentence Embeddings using Siamese BERT-Networks",<br /> author = "Reimers, Nils and Gurevych, Iryna",<br /> booktitle = "Proceedings of the 2019 Conference on Empirical Methods in Natural Language Processing",<br /> month = "11",<br /> year = "2019",<br /> publisher = "Association for Computational Linguistics",<br /> url = "http://arxiv.org/abs/1908.10084",<br /> }<br />