Local Embeddings with OpenVINO¶
OpenVINO™ is an open-source toolkit for optimizing and deploying AI inference. The OpenVINO™ Runtime supports various hardware devices including x86 and ARM CPUs, and Intel GPUs. It can help to boost deep learning performance in Computer Vision, Automatic Speech Recognition, Natural Language Processing and other common tasks.
Hugging Face embedding model can be supported by OpenVINO through OpenVINOEmbedding class, and OpenClip model can be through OpenVINOClipEmbedding class.
If you're opening this Notebook on colab, you will probably need to install LlamaIndex 🦙.
%pip install llama-index-embeddings-openvino
!pip install llama-index
Model Exporter¶
It is possible to export your model to the OpenVINO IR format with create_and_save_openvino_model function, and load the model from local folder.
from llama_index.embeddings.huggingface_openvino import OpenVINOEmbedding
OpenVINOEmbedding.create_and_save_openvino_model(
"BAAI/bge-small-en-v1.5", "./bge_ov"
)
Model Loading¶
If you have an Intel GPU, you can specify device="gpu" to run inference on it.
ov_embed_model = OpenVINOEmbedding(model_id_or_path="./bge_ov", device="cpu")
Compiling the model to CPU ...
embeddings = ov_embed_model.get_text_embedding("Hello World!")
print(len(embeddings))
print(embeddings[:5])
384 [-0.003275693394243717, -0.011690815910696983, 0.04155920818448067, -0.03814816474914551, 0.024183083325624466]
OpenClip Model Exporter¶
Class OpenVINOClipEmbedding can support exporting and loading open_clip models with OpenVINO runtime.
%pip install open_clip_torch
from llama_index.embeddings.huggingface_openvino import (
OpenVINOClipEmbedding,
)
OpenVINOClipEmbedding.create_and_save_openvino_model(
"laion/CLIP-ViT-B-32-laion2B-s34B-b79K",
"ViT-B-32-ov",
)
MultiModal Model Loading¶
If you have an Intel GPU, you can specify device="GPU" to run inference on it.
ov_clip_model = OpenVINOClipEmbedding(
model_id_or_path="./ViT-B-32-ov", device="CPU"
)
Embed images and queries with OpenVINO¶
from PIL import Image
import requests
from numpy import dot
from numpy.linalg import norm
image_url = "https://encrypted-tbn0.gstatic.com/images?q=tbn:ANd9GcStMP8S3VbNCqOQd7QQQcbvC_FLa1HlftCiJw&s"
im = Image.open(requests.get(image_url, stream=True).raw)
print("Image:")
display(im)
im.save("logo.jpg")
image_embeddings = ov_clip_model.get_image_embedding("logo.jpg")
print("Image dim:", len(image_embeddings))
print("Image embed:", image_embeddings[:5])
text_embeddings = ov_clip_model.get_text_embedding(
"Logo of a pink blue llama on dark background"
)
print("Text dim:", len(text_embeddings))
print("Text embed:", text_embeddings[:5])
cos_sim = dot(image_embeddings, text_embeddings) / (
norm(image_embeddings) * norm(text_embeddings)
)
print("Cosine similarity:", cos_sim)
Image:
Image dim: 512 Image embed: [-0.03019799292087555, -0.09727513045072556, -0.6659489274024963, -0.025658488273620605, 0.05379948765039444] Text dim: 512 Text embed: [-0.15816599130630493, -0.25564345717430115, 0.22376027703285217, -0.34983670711517334, 0.31968361139297485] Cosine similarity: 0.27307014923203976
For more information refer to:
