I have two time series for example:
s1:
2017-01-06 18:39:30 100
2017-01-07 18:39:28 101
and
s2:
2017-01-07 18:00:00 90
2017-01-08 18:00:00 105
I want to plot these in a Chartjs chart, however it seems that Chartjs only takes one x-axis array (or label in Chartjs terminology).
So my question is what is the best way to plot both of these?
My approach was to write a function (in python, although the language doesn't really matter for this part) that iterates through both time series and creates 3 new arrays which is in the format apparently Chartjs needs based off the 1st example here: https://www.sitepoint.com/introduction-chart-js-2-0-six-examples/
The algorithm (in sudo code) goes like:
# inputs are initial time series s1 and s2
y1 = [] # to hold new s1 data values
y2 = [] # to hold new s2 data values
x = [] # to hold times
# iterate through longest series s1 or s2
if s1[idx].time > s2[idx].time
x.append(s1[idx].time)
y1.append(s1[idx].data)
# y2 appends the linear interpolation of
# of closest y2 points
if (s1[idx].time < s2[idx].time)
x.append(s2[idx].time)
# opposite of above. ie. swap y1<->y2, s1->s2
else # they have the same time
x.append(s1[idx].time)
y1.append(s1[idx].data)
y2.append(s2[idx].data)
There are a couple other conditional checks for when data runs out of the shorter series but that is the main logic. After which I have 3 arrays that I can now add to chart js via one time/label array/x-axis and two data arrays. However this seems WAY more complicated than it should be considering how common I assume this use case is. Any help or advice is greatly appreciated.
In ChartJS, label is a Category Cartesian Axis. Since you mentioned linear interpolation in your code, I assume the strings like 2017-01-06 18:39:30
are not categories, they represent the numeric values of the x-axis. So we need to inform ChartJS that the strings in the x axis are actually time. We do this in the scale options.
var s1 = {
label: 's1',
borderColor: 'blue',
data: [
{ x: '2017-01-06 18:39:30', y: 100 },
{ x: '2017-01-07 18:39:28', y: 101 },
]
};
var s2 = {
label: 's2',
borderColor: 'red',
data: [
{ x: '2017-01-07 18:00:00', y: 90 },
{ x: '2017-01-08 18:00:00', y: 105 },
]
};
var ctx = document.getElementById('myChart').getContext('2d');
var chart = new Chart(ctx, {
type: 'line',
data: { datasets: [s1, s2] },
options: {
scales: {
xAxes: [{
type: 'time'
}]
}
}
});
<script src="https://cdnjs.cloudflare.com/ajax/libs/moment.js/2.20.1/moment.min.js"></script>
<script src="https://cdnjs.cloudflare.com/ajax/libs/Chart.js/2.7.1/Chart.min.js"></script>
<canvas id="myChart"></canvas>
You can find more information in Chart.js documentation.
You can have data of the form [{x:"value", y:"value"}] when your graph is of type scatter.
So to make your graph work, do this.
var canvas = document.getElementById("graph");
var s1 = [{x:"2017-01-06 18:39:30",y:"100"},{x:"2017-01-07 18:39:28",y:"101"}];
var s2 = [{x:"2017-01-07 18:00:00",y:"90"},{x:"2017-01-08 18:00:00",y:"105"}];
var graphParams = {
type:"scatter",
data:{
datasets: [{
label:"Series 1",
data:s1,
borderColor:"red",
backgroundColor:"transparent",
},
{
label:"Series 2",
data:s2,
borderColor:"blue",
backgroundColor:"transparent",
}],
},
options:{
maintainAspectRatio:false,
responsive:false,
scales:{
xAxes:[{
type:"time",
distribution: "series",
}],
}
}
}
ctx = new Chart(canvas, graphParams);
<canvas id="graph" height="500" width="700"/>
<script src="https://cdnjs.cloudflare.com/ajax/libs/Chart.js/2.2.1/Chart.bundle.min.js"></script>
If you love us? You can donate to us via Paypal or buy me a coffee so we can maintain and grow! Thank you!
Donate Us With