All Mpnet Base V2 Feature Extraction

🚀 all-mpnet-base-v2 clone

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

🚀 Quick Start

This section provides a quick guide on how to use the all-mpnet-base-v2 clone model.

✨ Features

• Maps sentences and paragraphs to a 768-dimensional dense vector space.
• Can be used for clustering and semantic search tasks.
• The only difference from the official model is the change of pipeline_tag: feature-extraction in the README.md.

📦 Installation

If you want to use this model, you need to install the sentence-transformers library. You can use the following command to install it:

pip install -U sentence-transformers

💻 Usage Examples

Basic Usage

If you have the sentence-transformers library installed, you can use the model as follows:

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

model = SentenceTransformer('sentence-transformers/all-mpnet-base-v2')
embeddings = model.encode(sentences)
print(embeddings)

Advanced Usage

Even if you don't have the sentence-transformers library installed, you can still use the model. First, pass your input through the transformer model, and then apply the appropriate pooling operation to the contextualized word embeddings.

from transformers import AutoTokenizer, AutoModel
import torch
import torch.nn.functional as F

#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('sentence-transformers/all-mpnet-base-v2')
model = AutoModel.from_pretrained('sentence-transformers/all-mpnet-base-v2')

# 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
sentence_embeddings = mean_pooling(model_output, encoded_input['attention_mask'])

# Normalize embeddings
sentence_embeddings = F.normalize(sentence_embeddings, p=2, dim=1)

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

📚 Documentation

Evaluation Results

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

Background

The project aims to train sentence embedding models on very large sentence-level datasets using a self-supervised contrastive learning objective. We used the pretrained microsoft/mpnet-base model and fine-tuned it on a 1B sentence pairs dataset. We used a contrastive learning objective: given a sentence from the pair, the model should predict which out of a set of randomly sampled other sentences was actually paired with it in our dataset.

We developed this model during the Community week using JAX/Flax for NLP & CV, organized by Hugging Face. We developed this model as part of the project: Train the Best Sentence Embedding Model Ever with 1B Training Pairs. We benefited from efficient hardware infrastructure to run the project: 7 TPUs v3-8, as well as intervention from Google's Flax, JAX, and Cloud team members about efficient deep learning frameworks.

Intended uses

Our model is intended to be used as a sentence and short paragraph encoder. Given an input text, it outputs a vector that captures the semantic information. The sentence vector can be used for information retrieval, clustering, or sentence similarity tasks.

By default, input text longer than 384 word pieces is truncated.

Training procedure

Pre-training

We used the pretrained microsoft/mpnet-base model. Please refer to the model card for more detailed information about the pre-training procedure.

Fine-tuning

We fine-tuned the model using a contrastive objective. Formally, we computed the cosine similarity from each possible sentence pair in the batch. We then applied the cross-entropy loss by comparing with the true pairs.

Hyperparameters

We trained our model on a TPU v3-8. We trained the model for 100k steps using a batch size of 1024 (128 per TPU core). We used a learning rate warm-up of 500. The sequence length was limited to 128 tokens. We used the AdamW optimizer with a 2e-5 learning rate. The full training script is available in this repository: train_script.py.

Training data

We used the concatenation of multiple datasets to fine-tune our model. The total number of sentence pairs is over 1 billion. We sampled each dataset with a weighted probability, and the configuration is detailed in the data_config.json file.

📄 License

This project is licensed under the Apache-2.0 license.