polars-in-the-artic

# Large Example Dataset (2GB of Memory)
example_data <-
    wakefield::r_data_frame(
        n = 15000000,
        wakefield::name(replace = TRUE),
        wakefield::sex_inclusive(),
        wakefield::marital(),
        wakefield::dob(),
        wakefield::education(),
        wakefield::employment(),
        wakefield::car(),
        wakefield::political(),
        wakefield::sat(),
        wakefield::income(),
        wakefield::state()
    )

# Parquet File (200 MB of Space)
rio::export(
    example_data,
    "example_data.parquet"
)

rm(example_data)

# Scan File
example_data <- polars::scan_parquet(
    "example_data.parquet"
)

# Preview
example_data$
    head(5)$
    collect()$
    to_data_frame() |>
    View("Data Preview")

# Get Row Count
example_data$
    collect()$
    height

# Split the Cars Column into Make and Model
result <- example_data$
    with_columns(
        polars::pl$col("Car")$
            cast(polars::pl$dtypes$Utf8)$
            str$split(" ")$
            arr$first()$
            alias("Make"),
        polars::pl$col("Car")$
            cast(polars::pl$dtypes$Utf8)$
            str$split(" ")$
            arr$last()$
            alias("Model")
    )

# Preview
result$
    head(5)$
    collect()$
    to_data_frame() |>
    View("Preview Result")

# Mean Income by State
mean_income_by_state <- example_data$
    groupby("State")$
    agg(
        polars::pl$col("Income")$mean()
    )$
    collect()$
    to_data_frame()

# Median SAT by Political Alignment
median_sat_by_poltics <- example_data$
    groupby("Political")$
    agg(
        polars::pl$col("SAT")$median()
    )$
    collect()$
    to_data_frame()

# Marital Status by Gender
merital_status_by_gender <- example_data$
    groupby("Sex", "Marital")$
    agg(
        polars::pl$first()$len()$alias("Count")
    )$
    sort("Sex")$
    collect()$
    to_data_frame()