query2query

This is a sentence-transformers model: It maps queries to a 384 dimensional dense vector space and can be used for tasks like clustering or semantic search over queries.

Checkout this announcing blogpost for more information: https://neeva.com/blog/state-of-the-art-query2query-similarity(https://neeva.com/blog/state-of-the-art-query2query-similarity)

Note: we are releasing this under a license which prevents commercial use. If you want to use it for commercial purposes, please reach out to contact@neeva.co or rajhans@neeva.co with a brief description of what you want to use it for and we will try our best to respond very quickly.

Usage (Sentence-Transformers)

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

pip install -U sentence-transformers

Then you can use the model like this:

from sentence_transformers import SentenceTransformer
queries = ["flight cost from nyc to la", "ticket prices from nyc to la"]

model = SentenceTransformer('neeva/query2query')
embeddings = model.encode(queries)
print(embeddings)

Training

The model was trained for 1M steps with a batch size of 1024 at a learning rate of 2e-5 using a cosine learning rate scheduler with 10000 warmup steps.

Full Model Architecture

SentenceTransformer(
  (0): Transformer({'max_seq_length': 256, 'do_lower_case': False}) with Transformer model: DataParallel 
  (1): Pooling({'word_embedding_dimension': 384, 'pooling_mode_cls_token': False, 'pooling_mode_mean_tokens': True, 'pooling_mode_max_tokens': False, 'pooling_mode_mean_sqrt_len_tokens': False})
  (2): Normalize()
)
Downloads last month
243
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.

Spaces using neeva/query2query 2