Computer Vision Data Science Deep Learning Machine-Learning
From scientific applications to consumer software to internal systems in the enterprises, deep learning technology is transforming how we interact with and make use of all the data. Especially with the rapid growth of media and metadata in these sectors, there is an evolving need for search systems to go beyond the conventional search approach (symbolic search) and towards semantic search, or neural search.
Deep learning technology provides a good base to semantically search for information. However, building a neural search system is non-trivial for researchers and engineers without good understanding of both search workflows and deep learning models.
In this workshop, we will show you step by step how you can make use of open source tools (Jina + BigTransfer model) to build a neural search system that performs image-to-image search with an open source dataset, in an easier way.
Jina is an open source neural search framework that empowers software developers to utilize deep learning models to build search systems that are fast, scalable, and work with any kind of data.
Development environment: WSL 2 (Windows) / Linux / MacOS with Python 3.7
Intro to Neural Search
Advanced configuration of Jina (replacing model, replacing preprocessing pipeline etc.)
Pipeline hyperparameter optimization
Jina Box (web interface)
Deployment to AWS
Type: Training (180 mins); Python level: Beginner; Domain level: Intermediate
I'm a Machine Learning Engineer with a practical mindset and a passion for the field.
I work at Jina AI, in Berlin, where I am making neural search open source and universal.
I have six years of professional experience across the stack (Machine Learning, Backend, DevOps, Frontend).