Describe the problem

We should test on larger datasets that are commonly used in

https://docs.python.org/3/library/typing.html

This way python autocomplete engines work better when the function types are known. It would be a considerable improvement in this library since code can be type checked before running it, not to mention improved auto-complete correctness.

• correcting handling asserts
  • replacing assert condition with the pattern
    if not condition: raise AssertionError()

For discussion, see IndrajeetPatil/ggstatsplot#333 (comment)

• Investigate CSS + nbconvert. Make sure it looks nice, and we can link between pages? (Sudeep)
• Integrate into sbt build (Aish)
• Write general overview on how to plot
• Write withData tutorial (plus Spark integration)
• Write rendering + notebook tutorial

Complementar a base de dados com esses novos CNPJs. Links interessantes:

• Tabelas complementares disponíveis no site da Receita Federal
• Tabela com CNPJ dos Entes Federativos
• [Tabela de Qualificação do Re

Hi,

I am currently going through the tutorial
https://krangl.gitbook.io/docs/tutorials/statistics

At the bottom of the page there is a code listing:

// PC1 vs PC2 scatter
plotOf(projection.transpose().array().withIndex()) {
    mark(MarkType.point)
    encoding(x) { value[0] }
    encoding(y) { value[1] }
    encoding(text){ "PC"+index}
}.render()

This uses plot

Any plans to get this into DefinitelyTyped?

Originally posted by @danielgwilson in Gmousse/dataframe-js#43 (comment)