ExploringExplorer

Mix.install([
  {:explorer, "~> 0.10.1"},
  {:kino_explorer, "~> 0.1.24"},
  {:req, "~> 0.5.14"},
  {:tucan, "~> 0.5.0"},
  {:kino_vega_lite, "~> 0.1.13"},
  {:kino_db, "~> 0.3.0"},
  {:exqlite, "~> 0.23.0"}
])

Section

This notebook is for learning purposes. Using Elixir Explorer library and the famous titanic dataset to get familiar with things.

require Explorer.DataFrame, as: DF
alias VegaLite, as: Vl

# Load the data
df = DF.from_csv!("/home/andresh/elixir/live/data/train.csv")

# Question one, how many people survived?

output =
  df
  |> DF.lazy()
  |> DF.group_by(["Sex", "Survived"])
  |> DF.summarise(PassengerId_count: count(col("PassengerId")))
  |> DF.sort_by(asc: col("Sex"))

output |> DF.rename( %{"PassengerId_count" => "count"} ) |> DF.collect()

bar = Tucan.bar(output, "Survived", "PassengerId_count", color_by: "Sex")
bar |> Tucan.set_size(500,500)
bar

Tucan.scatter(df, "Age", "Fare", color_by: "Sex" , shape_by: "Embarked")
|> Tucan.set_size(500,500)