Sentence Similarity
sentence-transformers
PyTorch
English
deberta-v2
feature-extraction
Generated from Trainer
dataset_size:305010
loss:CachedGISTEmbedLoss
Eval Results (legacy)
text-embeddings-inference
Instructions to use bobox/DeBERTa-small-ST-v1-test-step2 with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- sentence-transformers
How to use bobox/DeBERTa-small-ST-v1-test-step2 with sentence-transformers:
from sentence_transformers import SentenceTransformer model = SentenceTransformer("bobox/DeBERTa-small-ST-v1-test-step2") sentences = [ "how long should a prelude be before a funeral", "Organic 101: What the USDA Organic Label Means. This is the third installment of the Organic 101 series that explores different aspects of the USDA organic regulations. Organic certification requires that farmers and handlers document their processes and get inspected every year.", "The Quadrille. The Quadrille is a historic dance performed by four couples in a rectangular formation, and a precursor to traditional square dancing as well as a style of music. The Quadrille or Quadrille de Contre Danse was originally a card game for four people but the name was given to this dance about 1740.The dance probably derived from the Cotillions of the time. The Quadrille was a very lively dance, unlike the Minuet.Wikipedia states thus: The term quadrille came to exist in the 17th century, within military parades, in which four horsemen and their mounts performed special square-shaped formations or figures.he Quadrille or Quadrille de Contre Danse was originally a card game for four people but the name was given to this dance about 1740. The dance probably derived from the Cotillions of the time. The Quadrille was a very lively dance, unlike the Minuet.", "1 Arrive early. 2 You should always endeavor to arrive at the church or funeral home at between 15 to 20 minutes before the service is scheduled to begin. 3 Take your seat quietly, and reverently await the arrival of the family." ] embeddings = model.encode(sentences) similarities = model.similarity(embeddings, embeddings) print(similarities.shape) # [4, 4] - Notebooks
- Google Colab
- Kaggle