QueryIt Smart is the demonstration application where it combines BigQuery for large scale data analytics and Cloud Machine Learning Engine for scalable machine learning analytics. See What is it? page for details of the technology.
You need to have a Google Cloud Platform project and get service account's JSON key.
Pleace it to config/service_account.json
.
For Map API used in demand forecast, you should generate API key from Google Cloud Developers Console.
docker build -t queryit-smart:latest .
At the top directory of this project, run the following command.
Please replace MY_API_KEY
with your API Key generated in Setup section.
Replace MY_USER
and MY_PASS
with your user/password for HTTP basic authentication.
docker run -d -v $PWD/config:/usr/app/config -p8080:8080 -e BASIC_USER=MY_USER -e BASIC_PASSWORD=MY_PASS -e API_KEY=MY_API_KEY queryit-smart:latest
Open http://localhost:8080
(if you use Virtual Machine or CCE to run docker container, replace localhost
with your VM address) in browser.
See models/README.md to know how to customize datasets for Image search and Document search.