data = FileAttachment("powerplants_allUS_2010.csv").csv()
viewof table = Inputs.table(data)
Prepping Data: Changing Text to Numbers
If you scan through the data, you'll see there are text and numbers, along many rows and many different dimensions (columns).
The table above doesn't show this well, but if you expand the triangle next to data = > Array(5587) above, you'll see the data in it's more pure form, and you'll see that everything is being brought in from the CSV as a text string. They're not numbers. Further, many of these numbers have commas in them, which is great for human readability (it's actually a human visual processing thing), but not so good for a computer doing calculations. We need to parse this to change some of these text to actual numbers.
mile = "5280" // a variable as a string, similar to how a csv file brings in data.
Number(mile) // change to a number
+mile // change to a number
mile2 = "5,280" // another string, this time with a comma in it.
mile2.replace(/,/g,'') // finds all (the g) instances of the text , between the //, and replaces it with nothing ''
toNum = (text) => Number(text.replace(/,/g,''))
toNum(mile2)
data.forEach(d => {
d["Plant annual net generation (MWh)"] = toNum(d["Plant annual net generation (MWh)"]);
d["Plant annual CO2 equivalent emissions (tons)"] = toNum(d["Plant annual CO2 equivalent emissions (tons)"]);
})
data[1]
data[1]["Plant annual net generation (MWh)"]
data[1]["Plant annual CO2 equivalent emissions (tons)"]
data.forEach(d => {
d["Plant CO2 factor"] = toNum(d["Plant annual CO2 emissions (tons)"]) * 1;
d["Plant CH4 factor"] = toNum(d["Plant annual CH4 emissions (lbs)"]) * 2000 * 25;
})
data[5]["Plant CO2 factor"]
data[5]["Plant CH4 factor"]
data.length
data.filter(d => d["Plant state abbreviation"] == "VA")
data.filter(d => d["Plant annual net generation (MWh)"] > 300000)
data.filter(d => d["Plant primary fuel generation category"] == "GAS")
data.filter(d => d["Plant primary fuel generation category"] == "GAS" && d["Plant annual net generation (MWh)"] > 300000)
filteredData = data.filter(d => d["Plant primary fuel generation category"] == "GAS" && d["Plant annual net generation (MWh)"] > 300000)
minNetGen = d3.min(data, d => d["Plant annual net generation (MWh)"])
maxNetGen = d3.max(data, d => d["Plant annual net generation (MWh)"])
extentsNetGen = d3.extent(data, d => d["Plant annual net generation (MWh)"]) // both min and max
sumNetGen = d3.sum(data, d => d["Plant annual net generation (MWh)"]) // in true Tableau fashion!
meanNetGen = d3.mean(data, d => d["Plant annual net generation (MWh)"])
medianNetGen = d3.median(data, d => d["Plant annual net generation (MWh)"])
plantsByState = d3.group(data, d => d["Plant state abbreviation"])
netGenByState = d3.rollup(data, v => d3.sum(v, d => d["Plant annual net generation (MWh)"]), d => d["Plant state abbreviation"]);
netGenByState.get("VA")
netGenByStateAndFuel = d3.rollup(data, v => d3.sum(v, d => d["Plant annual net generation (MWh)"]), d => d["Plant state abbreviation"], d => d["Plant primary fuel generation category"]);
netGenByStateAndFuel.get("VA")
netGenByStateAndFuel.get("VA").get("GAS")
stateAvgLongs = d3.rollup(data, v => d3.mean(v, d => d["Plant longitude"]), d => d["Plant state abbreviation"]);
stateAvgLongs.get("VA")
//sortedData = d3.sort(data, d => d["Plant annual net generation (MWh)"])
//sortedData = d3.sort(data,(a,b) => d3.ascending(a["Plant annual net generation (MWh)"],b["Plant annual net generation (MWh)"]))
viewof sortedtable = Inputs.table(sortedData)
data.slice(0,1000)
data.slice(1000,5000)
//shuffledData = d3.shuffle(data)
viewof shuffledtable = Inputs.table(shuffledData)
//sample = d3.shuffle(data).slice(0,1000)
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.