awkward-array

Build nested records from jagged arrays in Python by using ak.zip to combine multiple variable-length arrays into structured event-like data. This creates a single array of records where each field is accessible by name, enabling structured queries on complex datasets.

Add a derived field to jagged data by applying ak.with_field to an existing Awkward Array. This function returns a new array with the calculated field appended, allowing you to extend records without manually restructuring the nested lists.

Compute pairings and combinations across nested lists using ak.cartesian for cross-list pairings and ak.combinations for n-way combinations within the same list. These functions handle jagged boundaries automatically, returning nested arrays of valid index pairs.

You can use NumPy interoperability when filtering and sorting Awkward arrays by passing Awkward structures directly into compatible NumPy functions. This allows argmin and argmax slicing operations to execute safely while preserving the underlying jagged data structure.

Transform variable-length event data into flat tables for analysis by flattening nested Awkward Array structures. This process unwraps jagged lists into standard tabular rows, making the dataset immediately compatible with standard plotting and analysis libraries.