chart = {
const wtcut = 0;
const links = data.links.map(d => Object.create(d));
const nodes = data.nodes.map(d => Object.create(d));
const simulation = d3.forceSimulation(nodes)
.force('x', d3.forceX().strength(2).x(d =>width/(maxLayer+1) * (d.LayerNum+1)))
.force("link", d3.forceLink(links).id(d => d.id))
.force("charge", d3.forceManyBody().strength(-250))
.force("center", d3.forceCenter(width / 2, height / 2));
const svg = d3.select(DOM.svg(width, height));
const link = svg.append("g")
.selectAll("line")
.data(links)
.join("line")
.attr("class", "link")
.attr("stroke", d => d.weight> 0? "#FF374C":"#6495ED")
.attr("stroke-opacity", d => Math.abs(d.weight)/maxwt)
.attr("stroke-width", d => Math.abs(d.weight)>wtcut? Math.abs(d.weight)*2 : 0);
const node = svg.append("g")
.attr("stroke", "#fff")
.attr("stroke-width", 1.5)
.selectAll("circle")
.data(nodes)
.join("circle")
.attr("r", d => Math.abs(d.valueScaled)*10+2)
.attr("fill", d => d.valueScaled> 0? "#FF8C00":"#00CED1")
//.attr("fill", color)
//.attr("opacity", d => Math.abs(d.valueScaled)*0.6)
.call(drag(simulation));
const bar = svg.append("g")
bar.append("rect")
.attr(x)
var x = d3.scaleLinear()
.domain([-0.1, 1.1])
.range([100, width-100]);
var brush = d3.brushX()
.extent([[100, height*9/10 - 30], [width-100, height*9/10]])
.on("start brush", brushed);
const axis = bar.append("g")
.attr("transform", "translate(0," + height*9/10 + ")")
.call(d3.axisBottom(x));
var dot = bar.append("g")
.selectAll(".dot")
.data(links)
.enter()
.append("circle")
.attr("class","dot")
.attr("cx", d=>x(Math.abs(d.weight)/maxwt))
.attr("cy", d => d.weight> 0? height*9/10 -10 : height*9/10 -20)
.attr("fill", d => d.weight>= 0? "#FF374C":"#6495ED")
.attr("r", 1.5)
.attr("fill-opacity", d=> Math.abs(d.weight)/maxwt);
bar.append("g")
.call(brush)
.call(brush.move, [x(0),x(1.1)])
.selectAll(".overlay")
.each(function(d) { d.type = "selection"; }) // Treat overlay interaction as move.
.on("mousedown touchstart", brushclicked); // Recenter before brushing.
function brushclicked() {
var cx = d3.mouse(this)[0];
d3.select(this.parentNode).call(brush.move, cx > x(1.1) ? [x(0), x(1.1)] : cx < x(-0.1) ? [x(-0.1), x(1.1)] : [cx, x(1.1)]);
}
function brushed() {
var extent = d3.event.selection;
d3.selectAll(".link").attr("stroke-width", function(d) {
if(extent[0] <= x(Math.abs(d.weight)/maxwt) && x(Math.abs(d.weight)/maxwt) <= extent[1])
{
return Math.abs(d.weight)*2;
}
else
{
return 0;
}
});
}
node.append("title")
.text(d => ['layer:'+d.LayerNum, ' node:'+d.NodeNum, ' value:'+d.value]);
simulation.on("tick", () => {
link
.attr("x1", d => d.source.x)
.attr("y1", d => d.source.y)
.attr("x2", d => d.target.x)
.attr("y2", d => d.target.y);
node
.attr("cx", d => d.x)
.attr("cy", d => d.y);
});
invalidation.then(() => simulation.stop());
return svg.node();
}
viewof model = {
const model = select({
title: "Which model?",
options: [
{ label: "🤷", value: "https://raw.githubusercontent.com/cse512-19s/FP-Visualizing-neural-network-architecture/master/docs/FeedforwardNN.json?token=AHF5OJCAMRERWY3PDMXKYIS473BMA" },
],
value: "https://raw.githubusercontent.com/cse512-19s/FP-Visualizing-neural-network-architecture/master/docs/FeedforwardNN.json?token=AHF5OJCAMRERWY3PDMXKYIS473BMA"
});
model.input.style.fontSize = "30px";
model.input.style.marginTop = "8px";
return model;
}
InputValues = [phi0, theta0, x99, y99, phi99, theta99, xdot0, ydot0, phidot0, thetadot0]
md `## Graph Plot Setting`
height = 800
width =1000
maxLayer =d3.max(data.nodes, d => d.LayerNum)
maxwt = d3.max(data.links, d => Math.abs(d.weight))
maxnode = d3.max(data.nodes, d => Math.abs(d.value))
color = {
const scale = d3.scaleOrdinal(d3.schemeCategory10);
return d => scale(d.LayerNum);
}
drag = simulation => {
function dragstarted(d) {
if (!d3.event.active) simulation.alphaTarget(3).restart();
d.fx = d.x;
d.fy = d.y;
}
function dragged(d) {
d.fx = d3.event.x;
d.fy = d3.event.y;
}
function dragended(d) {
if (!d3.event.active) simulation.alphaTarget(0);
d.fx = null;
d.fy = null;
}
return d3.drag()
.on("start", dragstarted)
.on("drag", dragged)
.on("end", dragended);
}
d3 = require("d3@5")
import {number} from "@jashkenas/inputs"
import {slider} from "@jashkenas/inputs"
import {select} from "@jashkenas/inputs"
md `## Graph Generation`
NN = d3.json(model)
InputNames = ['phi_0', 'theta_0', 'x_99', 'y_99', 'phi_99', 'theta_99', 'xdot_0', 'ydot_0', 'phidot_0', 'thetadot_0']
data = ComputeLayerValues(InputValues)
One platform to build and deploy the best data apps
Experiment and prototype by building visualizations in live JavaScript notebooks. Collaborate with your team and decide which concepts to build out.
Use Observable Framework to build data apps locally. Use data loaders to build in any language or library, including Python, SQL, and R.
Seamlessly deploy to Observable. Test before you ship, use automatic deploy-on-commit, and ensure your projects are always up-to-date.