Vector database sharding
Multimodal search
JaguarDB quantization
JaguarDB Vector API
Best Vector databases
JaguarDB in Docker
Setup JaguarDB with tar package
Setup JaguarDB on multiple nodes
Vector index sharing
How zeromove works
Video introduction
|
|
Example: AI Data Lake
JaguarDB provides data lake capability which is a powerful feature that integrates storage capabilities directly into the vector database system, offering a seamless and unified solution for managing, analyzing, and retrieving both structured and unstructured data. This integration brings efficiency, flexibility, and scalability to the storage and processing of diverse data types within the JaguarDB environment. Vector data is seamlessly integrated with unstructured data, facilitating near-computing for optimal efficiency and peak performance. This integration empowers systems to swiftly analyze and extract valuable insights from both structured and unstructured data sources in one place, significantly enhancing their overall capabilities.
The following example demonstrates how Jaguar stores multiple vectors together with their data sources for analysis from mutiple facets.
create store datalake (vx vector(512,'euclidean_fraction_float'), xray file, vmri vector(512,'euclidean_fraction_float'), mri file, vecg vector(512,'euclidean_fraction_float'), ecg file, vcat vector(512,'euclidean_fraction_float'), cat file, vhist vector(512,'euclidean_fraction_float'), hist char(2048), name char(64), vtime date )
embeddings1 = getEmbeddings('xray1.jpg')
embeddings2 = getEmbeddings('mri1.jpg')
embeddings3 = getEmbeddings('ecg1.jpg')
embeddings4 = getEmbeddings('cat1.jpg')
hist1 = 'This is one record of visit'
embeddings5 = getEmbeddings( hist1 )
insert into datalake values ('embeddings1', 'xray1.jpg', 'embeddings2', 'mri1.jpg', 'embeddings3', 'ecg1.jpg', 'embeddings4', 'cat1.jpg', 'embeddings5', hist1, 'John Doe', '2023-09-18')
embeddings1 = getEmbeddings('xray2.jpg')
embeddings2 = getEmbeddings('mri2.jpg')
embeddings3 = getEmbeddings('ecg2.jpg')
embeddings4 = getEmbeddings('cat2.jpg')
hist2 = 'This is another record of visit'
embeddings5 = getEmbeddings( hist2 )
insert into datalake values ('embeddings1', 'xray2.jpg', 'embeddings2', 'mri2.jpg', 'embeddings3', 'ecg2.jpg', 'embeddings4', 'cat2.jpg', 'embeddings5', hist2, 'John Doe', '2023-09-25')
embeddings1 = getEmbeddings('xray3.jpg')
embeddings2 = getEmbeddings('mri3.jpg')
embeddings3 = getEmbeddings('ecg3.jpg')
embeddings4 = getEmbeddings('cat3.jpg')
hist3 = 'This is third record of visit'
embeddings5 = getEmbeddings( hist2 )
insert into datalake values ('embeddings1', 'xray3.jpg', 'embeddings2', 'mri3.jpg', 'embeddings3', 'ecg3.jpg', 'embeddings4', 'cat3.jpg', 'embeddings5', hist3, 'Jane Doe', '2023-09-28')
The robust Jaguar datalake model intelligently pairs original image files with their corresponding vector data, cohesively storing them together. This integrated approach empowers users to seamlessly conduct diverse operations, including similarity searches, anomaly detection, and multimodal searches, across multiple vector fields with precision and efficiency.
|
