How to use Shipmaster1/minilm-youtube-adapter with PEFT:
from peft import PeftModel from transformers import AutoModel base_model = AutoModel.from_pretrained("sentence-transformers/all-MiniLM-L6-v2") model = PeftModel.from_pretrained(base_model, "Shipmaster1/minilm-youtube-adapter")
How to use Shipmaster1/minilm-youtube-adapter with sentence-transformers:
from sentence_transformers import SentenceTransformer model = SentenceTransformer("Shipmaster1/minilm-youtube-adapter") sentences = [ "What are some benefits of having a standard like MCP in place for applications leveraging context and context learning, especially in terms of making investment decisions easier for the higher-level executive team?", "Venus back in the 70s, but it got stuck in Earth's orbit and has been on a crash course ever since. But Ready says you shouldn't be worried about it hitting you. the the hope is that most of it will burn down in the earth's atmosphere and then even if something makes it down, it'll fall in the ocean", "project so I know that Vibe coding is kind of like it's about disassociating from you know needing to think about these things but there is still an excit an exciting viby part of like what do I want my project to look like what do I want my starting point to to be what stack do I want to these", "investment from your higher level exec team a little bit easier knowing that there's a standard on the horizon here. uh MCP really provides those standard rules for applications that are leveraging context and context learning etc etc remember we got stateful means it's aware of previous stuff that" ] embeddings = model.encode(sentences) similarities = model.similarity(embeddings, embeddings) print(similarities.shape) # [4, 4]
