In this example we show how to use Numpy and Matplotlib (Python) from an WebSocket server (NodeJS) in order to build a Polyglot Data Analysis and Visualization API.
Clone the repository:
git clone https://github.com/pawxnsingh/numpy-matplotlib-websocket-node-example
Install MetaCall CLI:
curl -sL https://raw.githubusercontent.com/metacall/install/master/install.sh | shNavigate to the directory:
cd numpy-matplotlib-websocket-node-exampleInstall application dependencies:
metacall pip3 install -r requirements.txt metacall npm install
metacall index.js
For testing it, in another terminal, let's visualize this data calculating the mean, median, standard deviation of this array [10, 12, 14, 15, 10, 11, 150] (you can change the values and size for experimenting)
use postman/hoppscotch(web)
URL: ws://localhost:8080
Message: {"numbers": [10, 12, 14, 15, 10, 11, 150]}It should output something like:
output: look for the output folder for the visualized view
An alternative version with Docker and automated testing is provided.
docker build -t metacall/numpy-matplotlib-websocket-node-example . docker run --rm -v $(pwd)/output:/metacall/output -p 8080:8080 -it metacall/numpy-matplotlib-websocket-node-example