31:T1998,md`# Explore data sets on your own Now it's time to play a little bit on your own. This task is pretty open ended, our only suggestion is that you make visalizations using vega-lite in observable. Observable has [many ways](https://observablehq.com/@observablehq/introduction-to-data?collection=@observablehq/working-with-data) to injest data, but it's probably easiest to start by uploading a CSV. ## Step 0: Fork this notebook You'll be modifying this notebook for your exploration, so take a moment and fork it into your account. ## Step 1: Find your dataset When you're looking for data, I'd recommend browsing some of the [sample datasets](https://vincentarelbundock.github.io/Rdatasets/datasets.html) that come with R. (Funny story, when I was learning vega lite, I did a little [meta-exploration](https://observablehq.com/@zekenie/datasets-availible-in-r) of these datasets...). Don't spend too much time picking data, if you're stuck just pick one! ## Step 2: Upload your dataset Once you've got your dataset picked out, you need to upload it to observable. In the uper right hand corner of the screen, click the menu, then select \`Upload Attachments\`. ## Step 3: Import your data For this section, we'll assume your CSV is called \`example.csv\`. Create an observable block that imports the file like this: \`\`\`js data = FileAttachment("example.csv").csv() \`\`\` You should be able to explore the output and see the data. *Note:* If your data is relatively clean, you can pass \`{ typed: true }\` to the \`csv\` function, and it will parse date strings, numbers, etc, so you don't have to do that later *Note:* If you have a JSON file, there's a sibling method of \`csv\` called \`json\`, and it works just how you'd expect. ## Step 4: Add suggested imports We'd recommend starting with the following imports

\`\`\`js import { vl } from "@vega/vega-lite-api" \`\`\`

\`\`\`js import { idea, problem, question, conclusion } from "@zekenie/thought-process" \`\`\`

\`\`\`js import { aq, op } from '@uwdata/arquero' \`\`\`

These imports need to each have their own Observable block. Observable imports are a little confusing. *Note:* You can import either NPM packages _or_ other observable notebooks. If you're interested in reading more about them, you can do that [here](https://observablehq.com/@observablehq/introduction-to-imports?collection=@observablehq/notebook-fundamentals). If you need an npm package, don't use \`import\`, [use require](https://observablehq.com/@observablehq/introduction-to-require). ## Step 5: Go to town Time to play. Look at other examples we've covered, other stuff that [Observable](https://observablehq.com/@observablehq/vega-lite?collection=@observablehq/observable-for-vega-lite) and [UW Data Lab](https://observablehq.com/@uwdata/introduction-to-vega-lite) have put together `32:[] 2f:{"title":"Explore data sets on your own","cells":[{"id":0,"value":"$31","pinned":false,"hidden":false,"mode":"ojs"},{"id":36,"value":"import { aq, op } from '@uwdata/arquero'","pinned":true,"hidden":false,"mode":"ojs"}],"files":"$W32","stdlib":"1","titling":"auto"}