Skip to content

pycauset.vector

pycauset.vector(source, dtype=None, *, max_in_ram_bytes=None, **kwargs)

Create a 1D vector from vector-like input.

This is a data constructor (aligned with 1D np.array([...])).

Parameters

  • source (sequence or numpy.ndarray): 1D data (e.g. list) or a 1D NumPy array.
  • dtype (str or type, optional): Coerce storage dtype (e.g. "float64", "int32", float, int).
  • max_in_ram_bytes (int or None, optional): When constructing from a NumPy array, route through the internal native.asarray import path if the estimated materialized size exceeds this cap to avoid overcommitting RAM (falls back to native regardless when supported dtypes are used).
  • kwargs: Passed through to the backend constructor.

Returns

Examples

import pycauset

v = pycauset.vector([1, 2, 3])
v_c = pycauset.vector([1+2j, 3-4j], dtype="complex_float32")