깔끔한 필터를 위한 query와 isin

Long chains of ampersands get hard to read. Two tools, query and isin, make complex filtering far more pleasant. ✨

The query method lets you describe the filter as plain text, so you skip the repeated df name and bracket noise.

Inside query you write natural words like and and or, instead of the symbols you needed in bracket filtering.

Pull an outside value into a query string by prefixing it with an at sign. This keeps thresholds out of the literal text.

A big win of query is comparing columns to each other directly, naming both inside the same readable string.

To match any value from a list, isin beats chaining many OR conditions. It returns a clean boolean mask.

Combine the tilde with isin to keep rows that are not in your list, a tidy way to exclude several values at once.

You can even call isin from within a query string, blending the two tools for readable multi-value filters.

Remember that isin tests each cell against a fixed list of allowed values. It is not meant for comparing one column to another column.

Like bracket filtering, query leaves the original untouched and hands back a fresh DataFrame to assign or chain.

Reach for query when logic is long and English-like, and for isin when one column must match many values. Pick whatever reads clearest.

Choose the cleanest tool for the job.

Use query for readable English-like filters with @variables, and isin to match many values at once. Your filters stay clean and clear. ✅