How to use bobox/RoBERTa-base-unsupervised-TSDAE with sentence-transformers:
from sentence_transformers import SentenceTransformer model = SentenceTransformer("bobox/RoBERTa-base-unsupervised-TSDAE") sentences = [ "free in spain? Are Spain free Motorways toll-free Spain, renewing old concessions coming", "how to calculate weighted grade percentage in excel? To find the grade, multiply the grade for each assignment against the weight, and then add these totals all up. So for each cell (in the Total column) we will enter =SUM(Grade Cell * Weight Cell), so my first formula is =SUM(B2*C2), the next one would be =SUM(B3*C3) and so on.", "In Red Dead Redemption 2's story mode, players have to go to \"Story\" in the menu and then click the save icon from there. However, in Red Dead Online, there is no such option. On the contrary, players have no way to manually save their game, which is pretty much par for the course in an online multiplayer experience.", "are motorways free in spain? Are motorways in Spain free? Motorways are 90% toll-free in Spain. Since 2018, Spain isn't renewing old concessions coming to end." ] embeddings = model.encode(sentences) similarities = model.similarity(embeddings, embeddings) print(similarities.shape) # [4, 4]
What is a pickle import?
