Одним из возможных решений (из многих) является выбор окружностей внутри каждой группы:
g.each(function(_, i) {
d3.select(this).selectAll("circle").style("stroke", colors(i))
});
В этом случае второй аргумент (i
) - это индекс каждой группы, который я 'м, чтобы перейти к цветовой шкале.
Вот обновленный JSFiddle: https://jsfiddle.net/3bjwv36r/
А вот фрагмент SO:
var tooltip = d3.select("body").append("div").attr("class", "toolTip");
var color = d3.scale.ordinal().range(["#6b486b", "#a05d56", "#d0743c", "#ff8c00"]);
var colors = d3.scale.ordinal().range(["blue", "orange", "green"]);
var width = 270,
height = 270,
margin = {
top: 20,
right: 10,
bottom: 20,
left: 35
},
n = 10000; // number of samples to generate
var chart = d3.qq()
.width(width)
.height(height)
.domain([-.1, 1.1])
.tickFormat(function(d) {
return ~~(d * 100);
});
var vis = d3.select("body").append("svg")
.attr("width", (width + margin.right + margin.left) * 3)
.attr("height", height + margin.top + margin.bottom)
.append("g")
.attr("transform", "translate(" + margin.left + "," + margin.top + ")")
d3.json("https://api.myjson.com/bins/uti7i", function(error, turkers) {
if (error) throw error;
var tm = mean(turkers[0].percent.values),
td = Math.sqrt(variance(turkers[0].percent.values)),
dd = [
[0.10306430789206111, 0.0036139086950272735, 0.30498647327844536],
[0.5924252668569606, 0.0462763685758622, 0.4340870312025223],
[0.9847627827855167, 2.352350767874714e-4, 0.2609264955190324]
];
var Mean = turkers[0].Mean;
var STDDeviation = turkers[0]["STD Deviation"];
var NormalTestPValue = turkers[0]["Normal test p-value"];
var minValue = turkers[0].percent.minValue;
var maxValue = turkers[0].percent.maxValue;
var g = vis.selectAll("g")
.data([{
x: d3.range(n).map(Math.random),
y: turkers[0].percent.values,
label: "Uniform Distribution"
}, {
x: d3.range(n).map(normal1(tm, td)),
y: turkers[0].percent.values,
label: "Gaussian (Normal) Distribution"
}, {
x: d3.range(n).map(normal3(dd)),
y: turkers[0].percent.values,
label: "Mixture of 3 Gaussians"
}])
.enter().append("g")
.attr("class", "qq")
.attr("r", 5)
.attr("transform", function(d, i) {
return "translate(" + (width + margin.right + margin.left) * i + ")";
})
.on("mousemove", function(d, turkers) {
tooltip
.style("left", d3.event.pageX - (-40) + "px")
.style("top", d3.event.pageY - 70 + "px")
.style("display", "inline-block")
.html(("<table><tr> <td><b>" +
d3.event.clientX + "<sup>th</sup> Percentile: </b>" + d.x[d3.event.clientX].toFixed(2) + " , " + d.y[d3.event.clientY].toFixed(2) + "</td></tr><tr><td><b>Mean : </b>" + Mean +
"</td></tr><tr><td><b>STD Deviation :</b> " + STDDeviation +
"</td></tr><tr><td><b>Normal Test P-Value :</b> " + NormalTestPValue +
"</td></tr><tr><td><b> Min Value : </b>" + minValue.toFixed(2) +
"</td></tr><tr><td><b> Max Value :</b> " + maxValue + "</td></tr><table>"
));
})
.on("mouseout", function(d) {
tooltip.style("display", "none");
});
g.append("rect")
.attr("class", "box")
.attr("width", width)
.attr("height", height);
g.call(chart);
g.append("text")
.attr("dy", "1.3em")
.attr("dx", ".6em")
.text(function(d) {
return d.label;
});
chart.duration(1000);
window.transition = function() {
g.datum(randomize).call(chart);
};
g.each(function(_, i) {
d3.select(this).selectAll("circle").style("stroke", colors(i))
});
});
function randomize(d) {
d.y = d3.range(n).map(Math.random);
return d;
}
// Sample from a normal distribution with mean 0, stddev 1.
function normal() {
var x = 0,
y = 0,
rds, c;
do {
x = Math.random() * 2 - 1;
y = Math.random() * 2 - 1;
rds = x * x + y * y;
} while (rds == 0 || rds > 1);
c = Math.sqrt(-2 * Math.log(rds) / rds); // Box-Muller transform
return x * c; // throw away extra sample y * c
}
// Simple 1D Gaussian (normal) distribution
function normal1(mean, deviation) {
return function() {
return mean + deviation * normal();
};
}
// Gaussian Mixture Model (k=3) fit using E-M algorithm
function normal3(dd) {
return function() {
var r = Math.random(),
i = r < dd[0][2] ? 0 : r < dd[0][2] + dd[1][2] ? 1 : 2,
d = dd[i];
return d[0] + Math.sqrt(d[1]) * normal();
}
}
// Welford's algorithm.
function mean(x) {
var n = x.length;
if (n === 0) return NaN;
var m = 0,
i = -1;
while (++i < n) m += (x[i] - m) / (i + 1);
return m;
}
// Unbiased estimate of a sample's variance.
// Also known as the sample variance, where the denominator is n - 1.
function variance(x) {
var n = x.length;
if (n < 1) return NaN;
if (n === 1) return 0;
var m = mean(x),
i = -1,
s = 0;
while (++i < n) {
var v = x[i] - m;
s += v * v;
}
return s / (n - 1);
}
body {
font: 12px sans-serif;
width: 960px;
height: 310px;
}
.qq .box,
.qq .tick line,
.qq .quantile,
.qq .diagonal {
stroke: #aaa;
fill: none;
}
.qq .quantile {
stroke: #000;
}
.qq .diagonal {
stroke: red;
}
.qq g+g .y.tick {
display: none;
}
.axis path,
.axis line {
fill: none;
stroke: #D4D8DA;
stroke-width: 1px;
shape-rendering: crispEdges;
}
.toolTip {
font: 12px sans-serif;
position: absolute;
display: none;
min-width: 80px;
height: auto;
background: #fff;
color: #000;
border: 1px solid red;
padding: 14px;
text-align: left;
}
<script src="https://cdnjs.cloudflare.com/ajax/libs/d3/3.4.11/d3.min.js"></script>
<script src="https://bl.ocks.org/mbostock/raw/4349187/d3.qq.min.js"></script>