Data visualization and analysis at the University of Washington, Seattle.
Published
Edited
Aug 30, 2021
18 forks
303 stars
dt = aq.table({
u: ['a', 'a', 'b', 'b'],
x: Int32Array.of(1, 2, 3, 4),
y: Float64Array.of(5, 6, 7, 8)
})
dt.view()
[dt.numCols(), dt.numRows()]
dt.columnNames()
dt.get('y', 2)
dt.array('y')
dt.objects()
[...dt] // or, Array.from(dt)
aq.from([{a: 1, b: 0.2}, {a: 4, b: 0.3}])
.view()
aq.from({a: 1, b: 2, c: 3})
.view()
aq.from(new Map([['d', 4], ['e', 5], ['f', 6]]))
.view()
flights = aq.loadArrow('https://vega.github.io/vega-datasets/data/flights-200k.arrow')
flights.view({ limit: 5 }) // view(5) also works as a shorthand
aq.fromJSON('{"id":[1,2],"date":["2020-06-01","2020-09-01"]}')
.view()
aq.table(JSON.parse('{"id":[1,2],"date":["2020-06-01","2020-09-01"]}'))
.view()
viewof beers = aq // viewof shows the table view, but assigns the table value
.fromCSV(await FileAttachment('beer.csv').text())
.view({ height: 240 }) // set maximum height of table viewer in pixels
beers
.orderby('abv')
.view(5) // shorthand for view({ limit: 5 })
beers
.orderby(aq.desc('abv'))
.view(5)
beers
.orderby(aq.desc('ibu'))
.select('name', 'style', 'ibu') // ['name', 'style', 'ibu'] also works
.view(5)
beers
.orderby(aq.desc('ibu'))
.select(0, 1, 4) // [0, 1, 4] also works
.view(5)
beers
.orderby(aq.desc('ibu'))
.select(aq.not('brewery_id')) // 'not' can accept multiple columns, too...
.view(5)
beers
.filter(d => op.includes(op.lower(d.name), 'hop'))
.view()
beers
.derive({ intensity: d => 3 * d.abv + op.log10(d.ibu) / 3 })
.orderby(aq.desc('intensity'))
.view(10)
beers
.rollup({
mean_abv: d => op.mean(d.abv), // op.average is available as a synonym
mean_ibu: d => op.mean(d.ibu),
mean_intensity: d => op.mean(3 * d.abv + op.log10(d.ibu) / 3),
corr_abv_ibu: d => op.corr(d.abv, d.ibu),
count: d => op.count()
})
.view()
beers
.groupby('style')
.rollup({
mean_abv: d => op.mean(d.abv),
mean_ibu: d => op.mean(d.ibu),
mean_intensity: d => op.mean(3 * d.abv + op.log10(d.ibu) / 3),
count: op.count()
})
.filter(d => d.count > 20) // let's not trust results with little support...
.orderby(aq.desc('mean_intensity'))
.view(10)
beers
.groupby('style')
.count()
.view(10)
// group by lower-case values, rename to 'type'
beers.groupby({ type: d => op.lower(d.style) })
beers
.groupby('style')
.derive({ intensity: d => 3 * d.abv + op.log10(d.ibu) / 3 })
.derive({ zscore: ({ intensity: x }) => (x - op.mean(x)) / op.stdev(x) || 0 })
.orderby(aq.desc('zscore'))
.select(aq.not('brewery_id', 'intensity'))
.view()
beers
.groupby('style')
.derive({ intensity: d => 3 * d.abv + op.log10(d.ibu) / 3 })
.derive({ zscore: ({ intensity: x }) => (x - op.mean(x)) / op.stdev(x) || 0 })
.orderby(aq.desc('zscore'))
.filter(d => op.rank() === 1 && op.count() > 20) // filter by rank and count
.select(aq.not('brewery_id', 'intensity'))
.view({ height: 400 }) // set max height to show full table without scroll
viewof breweries = aq
.fromCSV(await FileAttachment('breweries.csv').text())
.view(10)
beers
.join(breweries)
.view(5)
beers
.join(breweries, ['brewery_id', 'brewery_id']) // <- just 'brewery_id' also works!
.view(5)
beers
.join(breweries, 'brewery_id')
.groupby('style')
.rollup({ breweries: op.distinct('brewery_id') } )
.orderby(aq.desc('breweries'))
.view(10)
beers
.join(breweries,
(a, b) => op.equal(a.brewery_id, b.brewery_id),
[aq.all(), aq.not('brewery_id')]
)
.view(5)
breweries
.semijoin(beers) // implicit comparison on 'brewery_id'
.count()
.view()
breweries
.antijoin(beers) // implicit comparison on 'brewery_id'
.count()
.view()
Purpose-built for displays of data
Observable is your go-to platform for exploring data and creating expressive data visualizations. Use reactive JavaScript notebooks for prototyping and a collaborative canvas for visual data exploration and dashboard creation.
