docs(python): Minor tweak in code example in section Coming from Pandas

[jrycw]

While it's true that Polars offers the capability to execute data transformations in parallel using expressions, some of the examples in the documentation don't provide a complete 1-to-1 comparison between Pandas and Polars. To make this comparison more comprehensive, I would recommend contrasting Pandas assign and query methods with Polars with_columns and filter methods.

[MarcoGorelli]

thanks for your PR - generally looks good, just got a minor request

also, will need to address the lint error

[MarcoGorelli]

slightly more idiomatic would be to use pandas.Series.where directly

df.assign(a=lambda df_: df_.a.where(df_.c != 2, df_.b))

[jrycw]

Thanks @MarcoGorelli, I wasn't aware that Pandas dataframe includes its own where method.

[MarcoGorelli]

I think the polars example should probably also use this syntax then, i.e.

df.with_columns(
    tenXValue=pl.col("value") * 10,
    hundredXValue=pl.col("value") * 100,
)

[jrycw]

I'm not entirely certain, but it appears that using .alias is the recommended approach for renaming columns.

[MarcoGorelli]

both work https://pola-rs.github.io/polars/py-polars/html/reference/dataframe/api/polars.DataFrame.with_columns.html

this would just minimise the difference

[jrycw]

Considering this is a brief migration guide to help newcomers from Pandas, I agree with your perspective.

[MarcoGorelli]

Looks like an improvement, and makes the comparisons more "apples to apples" - nice one @jrycw !