import {vl} from '@vega/vega-lite-api'
import {printTable} from '@uwdata/data-utilities'
d3 = require('d3@5')
md`Upload the file using Shift-Command-u which brings up a file selector dialog box. We are using [Boston Housing Data from Kaggle](https://www.kaggle.com/c/boston-housing).`
fileContents = FileAttachment("Boston@1.csv")
md`Parse the file contents into an array`
housing = d3.csvParse(await fileContents.text(), d3.autoType)
md`Use printTable to look at the first few lines of data`
printTable(housing.slice(0, 10))
vl.markPoint()
.data(housing)
.encode(vl.x().fieldQ('CRIM'), vl.y().fieldQ('Median House Price'))
.render()
scatterPlot = vl
.markPoint()
.data(housing)
.encode(vl.x().fieldQ('CRIM'), vl.y().fieldQ('Median House Price'))
html`<pre>${JSON.stringify(scatterPlot.toJSON(), 0, 2).slice(0, 500)}</pre>`
vl
.markPoint()
.data(housing)
.encode(
vl.y().fieldQ('Median House Price'),
vl.x().fieldQ('CRIM'),
vl
.color()
.fieldQ('DIS')
.scale({ scheme: 'greens' })
)
.render()
md`Most of the points are bunched to the right. Left of the plot. What does this mean?`
vl
.markPoint()
.data(housing)
.transform(vl.filter('datum["CRIM"]>3'))
.encode(
vl.y().fieldQ('Median House Price'),
vl.x().fieldQ('CRIM'),
vl
.color()
.fieldQ('DIS')
.scale({ scheme: 'greens' })
)
.render()
survey = d3.csvParse(
await FileAttachment("student_background.csv").text(),
d3.autoType
)
printTable(survey.slice(0, 10))
vl
.markBar()
.data(survey)
.encode(
vl.y().fieldN('hci'),
vl
.x()
.count()
.title(null) // why is this here? what happens if you change it?
)
.render()
choiceSortOrder = ["no", "no, familiar", "yes, introductory", "yes, advanced"]
vl
.markBar()
.data(survey)
.encode(
vl
.y()
.fieldN('hci')
.sort(choiceSortOrder),
vl
.x()
.count()
.title(null)
)
.render()
{
const base = vl
.markBar()
.data(survey)
.encode(
vl
.x()
.count()
.title(null)
)
.width(100);
const hci = base.encode(vl.y().fieldN('hci').sort(choiceSortOrder));
const cogsci = base.encode(vl.y().fieldN('cogsci').sort(choiceSortOrder));
const stats = base.encode(vl.y().fieldN('stats').sort(choiceSortOrder));
const ml = base.encode(vl.y().fieldN('ml').sort(choiceSortOrder));
return vl
.hconcat(vl.layer(hci), vl.layer(cogsci), vl.layer(stats), vl.layer(ml))
.render();
}
vl.markBar()
.data(survey)
.encode(
vl.x().count().title(null),
vl.y().fieldN(vl.repeat('column')).sort(choiceSortOrder),
)
.width(100)
.repeat({column: ['hci', 'cogsci', 'stats', 'ml']})
.render();
vl.markBar()
.data(survey)
.encode(
vl.x().count().title(null).scale({domain: [0,17]}), // make all x axes the same length
vl.y().fieldN(vl.repeat('column')).sort(choiceSortOrder)
.axis({ titleAngle: 0, titleX: 0, titleY: 0}), // rotate the name of the axis to appear on topo
vl.color().fieldN(vl.repeat('column')).legend(null)
)
.width(100)
.repeat({column: ['hci', 'cogsci', 'stats', 'ml']})
.render();
vl.markPoint()
.data(survey)
.encode(
vl.x().fieldQ('hci_ordinal'),
vl.y().fieldQ('javascript_ordinal'),
vl.size().count()
).render()
vl.markPoint()
.data(survey)
.encode(
vl.x().fieldQ('stats_ordinal'),
vl.y().fieldQ('rlanguage_ordinal'),
vl.size().count().legend(null)
).render()
vl.markPoint()
.data(survey)
.encode(
vl.x().fieldQ('ml_ordinal'),
vl.y().fieldQ('python_ordinal'),
vl.size().count().legend(null)
).render()
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.