Sentence Similarity
sentence-transformers
Safetensors
xlm-roberta
feature-extraction
dense
Generated from Trainer
dataset_size:7264
loss:MultipleNegativesRankingLoss
text-embeddings-inference
Instructions to use olegGerbylev/e5-large-video-retrieval-ft-v2 with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- sentence-transformers
How to use olegGerbylev/e5-large-video-retrieval-ft-v2 with sentence-transformers:
from sentence_transformers import SentenceTransformer model = SentenceTransformer("olegGerbylev/e5-large-video-retrieval-ft-v2") sentences = [ "How do local Italian tour guides craft a narrative that connects ancient heritage with modern developments?", "what is perspective? how do we draw within it? and what kind of perspectives are there? well, i'm about to show you through some of the diagrams overlaid atop of real-life examples, starting with... one-point perspective with a single vanishing point. two-point perspective with vertical lines remaining parallel to each other, plus an additional vanishing point. three-point perspective with another vanishing point below or above. 4-point perspective connected with the top and bottom points in a warped way and 5-point perspective that is quite similar to 4-point perspective and when looked at a diagram would look something like this. but while knowing these 5 linear perspectives is all well and good, you usually don't use those grids every single time to draw in perspective. so, instead i want to focus on 3 key aspects of perspective drawing, starting with proximity. now, let's look at a cube, which i'm drawing in an orthogonal way, meaning that the vanishing points of the subject are so far out of bounds of our working area that the vanishing lines connecting to those dots are parallel to each other. we might even be able to see that sort of view in a real life when looking at an object from a distance. but if we gradually increase our proximity or shorten the distance between us and the subject matter, you'll notice that the vanishing points of our subject enter", "while they do serve a little bit of everything, i really recommend getting some seafood while you're here. you won't regret it. even though italy is a relatively large country, the train systems are so efficient, it's easy to get around quickly. after venice, take a train ride through milan to lake como or lake maggiore. these gorgeous italian lakes border switzerland and the breathtaking italian alps. stay for two or three days depending on how much hiking and boating you want to do. some of the towns can feel a little touristy, but the rooftop pools and astounding views make up for that. take a day to explore the mountains on some of the abundant hiking trails.", "right way? get ready for the ultimate italian trip itinerary that will change your life. from milan to rome and everything in between, we'll cover the must-see cities, delicious food, and the day-to-day route itinerary that will make your italian experience unforgettable. you're starting in milan. buy your round trip ticket now. this will be the most economical long haul flight option arriving in milan. the other option is flying to rome. but we'll talk about that later. for the sake of this, your ultimate trip itinerary, it begins milan." ] embeddings = model.encode(sentences) similarities = model.similarity(embeddings, embeddings) print(similarities.shape) # [4, 4] - Notebooks
- Google Colab
- Kaggle
Welcome to the community
The community tab is the place to discuss and collaborate with the HF community!