Cross-Encoder for MS Marco

This model is a generic masked language model fine tuned on stack overflow data. It's base pre-trained model was the cross-encoder/ms-marco-MiniLM-L-12-v2 model.

The model can be used for creating vectors for search applications. It was trained to be used in conjunction with a knn search with OpenSearch for a pet project I've been working on. It's easiest to create document embeddings with the flair package as shown below.

Usage with Transformers

from flair.data import Sentence
from flair.embeddings import TransformerDocumentEmbeddings

sentence = Sentence("Text to be embedded.")
model = TransformerDocumentEmbeddings("model-name")
model.embed(sentence)
embeddings = sentence.embedding
Downloads last month
4
Safetensors
Model size
33.4M params
Tensor type
I64
·
F32
·
Inference Examples
This model does not have enough activity to be deployed to Inference API (serverless) yet. Increase its social visibility and check back later, or deploy to Inference Endpoints (dedicated) instead.