How Jina is helping in NLP and ML .
Approch which was taken by team developing JINA certainly has
a future as it collects best from various sides to conquer ML and NLP field and especially neuronal based search solutions.
After experiences in Zalando, Tencent own framwork was created.
Supported by by docker or/and kubernetes and recetly by JINAhub which like docker hub make good impresion .
I try to check how modular easy and useful for me is this package
https://github.com/jina-ai/examples
The input query and the output result can be two different data types in a cross-modal search, such as text and images. In this example, I will look at a text-to-image search similar to Google Images, where you enter the description of the image you want to search, and the output is a number of similar imageserybody has lot of differen pictures from smartfone cameras and other devices and posiiblity to index them automatically or temptation of simple searchig not by file name but by content is very high.
This requires t match embedings of text and pictures to be compared . In this case, pre-trained cross-modal models could do that.
Ready to use jupyter notebook prepared by JINA team
There are multiple ready to use sets of data vaailable though to personalise a bit the presentation I decided to scrap quickly( getting pictures from internet with the help of bing engine)
it is enough to install download library
Project is: in progress
Created by: [email protected]