Get your Estimator powered app running in 5 minutes

Live Demo : click me!
(Note: We recommend you open all links on a new tab so you can work side by side with the github README)\

Prerequisite.

• You must have your API key. Don't have one? Try playing with our demo apps on https://estimatorlab.com to earn one.
• Python3 installed. You can install it from https://www.python.org/downloads/

Steps to get your app running

Step 1

Download or clone the repository.

Download: From your web browser, enter as your URL:

https://github.com/throughputer/estimator_tutorials/archive/main.zip

Clone command:

git clone https://github.com/throughputer/estimator_tutorials.git

Step 2

change directory

cd rock-paper-scissor

Step 3

open the file js/rps.js using any editor. E.g.:

vim js/rps.js

Step 4

Go to line number 90

To utilize your access to a ThroughPuter Estimator microservice, provide your API key here, HOWEVER... IMPORTANT!!! It is your responsibility to keep your API key secret. This code is visible in a user's web browser, and this demo is not intended to be hosted publicly with your private key.

this.prediction = new Prediction(
      4,
      true,
      `<URL>`,   // Copy-paste from estimatorlab.com/estimator >> Estimator API Key >> Show API Key >> Copy.
      predictionCB,
      { onopen: wsReady, onclose: onclosefn, onerror: onerrorfn }
    );

Step 5

Replace <URL> (so that the < > marks get eliminated). You can access your URL by navigating to https://estimatorlab.com/estimator and clicking on the Estimator API Key tab on the side, "Show API Key", and "Copy".

Note: Always keep your API key secret. This demo app is NOT built to be hosted publicly. Generally, unlike this demo, only the back end of your web application should connect to your private Estimator microservice.

Step 6

From the rock-paper-scissor directory, start a python3 http server (or any http server). You can install Python3 from here

For mac/linux based systems:

python3 -m http.server

For Windows based systems:

py -m http.server

You should see the following, showing that your http server is running on Port 8000:

Serving HTTP on 0.0.0.0 port 8000 (http://0.0.0.0:8000/) ...

Step 7

Open your favorite browser and enter the URL:

http://localhost:8000/

Confirm that you have connected successfully with your Estimator microservice instance by playing more than four moves of Rock-Paper-Scissors to see that the AI is making moves. (The first four predictions are random while the Estimator is trained.)

All done!

You can see a RPS app running with your API key. Check your Dashboard to view the number of calls made to the estimator microservice & keep track of your free credits consumption.

Note: The first 3 clicks are used to train the online Machine Learning engine and not counted towards your usage.

What's next?

Now you can try your hand at writing your own application. The available Estimator APIs are documented here.

Your trial Estimator microservice supports four variables (in the range 0-255) used to predict from among up to four class values (in the range 0-3). Modifications to these parameters can be requested.

Questions? Contact Us