Public
Edited
Jul 25, 2023
3 stars
data
Type Table, then Shift-Enter. Ctrl-space for more options.
selectedData.tipChannels
iris = ({
data: irisData,
zFeature: "species",
scale: false
})
wine = ({
data: wineData,
zFeature: "Wine",
scale: true
})
import { data as hawkerData } from "@yizhe-ang/sg-hawker-food-nutrition"
hawker = ({
data: hawkerData,
zFeature: "type",
scale: true,
toDrop: "food"
})
happinessData = whr_2023
whr_2023.csv
Type Table, then Shift-Enter. Ctrl-space for more options.
happiness = ({
data: happinessData,
// zFeature: "Ladder score",
zFeature: null,
scale: true,
toDrop: "Country name",
tipChannels: { Country: "Country name" }
})
data = selectedData.data
df = new dfd.DataFrame(data)
correlations = d3.cross(corrFeatures, corrFeatures).map(([a, b]) => ({
a,
b,
correlation: corr(Plot.valueof(data, a), Plot.valueof(data, b))
}))
scaledDf = {
const dfCopy = df.copy();
const dfToScale = dfCopy.drop({ columns: featuresToDrop });
const dfNotToScale = dfCopy.loc({ columns: featuresToDrop });
const scaler = new dfd.StandardScaler();
scaler.fit(dfToScale);
const scaledDf = scaler.transform(dfToScale);
return dfd.concat({ dfList: [dfNotToScale, scaledDf], axis: 1 });
}
scaledData = {
// const scaler = new dfd.StandardScaler();
// scaler.fit(dataForPca);
// return scaler.transform(dataForPca);
return dfd.toJSON(scaledDf);
}
scaledDataSample = {
const sample = await scaledDf.sample(20, { seed: 5 });
return dfd.toJSON(sample);
}
feature1Weight = featureScale * Math.sin((featureDirection * Math.PI) / 180)
feature2Weight = featureScale * Math.cos((featureDirection * Math.PI) / 180)
feature1WeightAlt = Math.sin((featureDirectionAlt * Math.PI) / 180)
feature2WeightAlt = Math.cos((featureDirectionAlt * Math.PI) / 180)
dfForPca = df.drop({ columns: featuresToDrop })
featuresForPca = dfForPca.columns
// dataForPca = df.loc({ columns: featuresForPca }).values
dataForPca = dfForPca.values
// Scaling is important!
pca = new ML.PCA(dataForPca, { scale: selectedData.scale })
pcaData = pca.predict(dataForPca).data
// biplotVectors = pca.getLoadings().transpose().data
biplotVectors = pca.getEigenvectors().data
ML = require("https://www.lactame.com/lib/ml/6.0.0/ml.min.js")
import { PlotMatrix } with { data } from "@observablehq/autoplot-matrix"
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.
