Author: | cudbg |
---|---|
Views Total: | 474 views |
Official Page: | Go to website |
Last Update: | June 10, 2016 |
License: | MIT |
Preview:

Description:
mpredict.js is a no-dependency JavaScript library used to predict the mouse position.
See also:
- JavaScript Library For Mouse Position Tracking And User Input Prediction – Trial.js
- JavaScript Library To Track Mouse Position Relative To Element – spot.js
How to use it:
Make sure the mpredict.js and mpredict-templates.json files are placed at the same folder. Then include the mpredict.js script on your webpage as follow:
<script src="mpredict.js"></script>
Start predicting mouse position:
mPredict.start();
Possible options with default values:
mPredict.start({ // The time interval(by milliseconds) of sampling the mouse trace sampleInterval: 10, // he time threshold of pausing; when the time between two consecutive mouse events is larger than this threshold, the second mouse event will be regarded as the beginning of a new trace pauseThreshold: 50, // The number of nearest neighbors in the KNN algorithm K: 5, // The path to the JSON file of templates used for the KNN algorithm pathToTemplates: './mpredict-templates.json', // The selector of target element to record mouse trace from targetElement: 'document' });
API.
// Get the current mouse trace recorded and sampled by MPredict.js on the target DOM element mPredict.getCurrentTrace() // Sample a mouse trace for prediction // trace : Array: [[x0, y0, t0], ..., [xn, yn, tn]], xi and yi are mouse positions(in pixels), ti is the timestamp mPredict.sampleTrace(trace) // Predict the mouse position for the given trace after deltaTime // trace : sampled mouse trace // deltaTime : integer(in milliseconds), deltaTime <= 0 means predicting the endpoint for the given trace mPredict.predictPosition(trace, deltaTime)