libtcspc.NumericTraits¶

class libtcspc.NumericTraits(**kwargs)[source]¶

Bases: object

Set of integer types used by events and processors that are parameterised by numeric traits.

This is the Python-side counterpart to tcspc::default_numeric_traits on the C++ side. Constructing NumericTraits with no arguments yields the default set; supply one or more keyword arguments to override individual slots. Each override must be a NumPy dtype object, dtype instance, or dtype string, and must satisfy the kind constraint documented below.

Parameters:

abstime_type (dtype-like, optional) – Type used for absolute time (macrotime). Must be a signed or unsigned integer dtype. Default numpy.int64.
channel_type (dtype-like, optional) – Type used for channel numbers. Must be a signed or unsigned integer dtype. Default numpy.int32.
difftime_type (dtype-like, optional) – Type used for difference times (microtime). Must be a signed integer dtype. Default numpy.int32.
count_type (dtype-like, optional) – Type used for counts of detections. Must be a signed or unsigned integer dtype. Default numpy.uint32.
datapoint_type (dtype-like, optional) – Type used for histogram datapoint values. Must be a signed or unsigned integer dtype. Default numpy.int32.
bin_index_type (dtype-like, optional) – Type used for histogram bin indices. Must be an unsigned integer dtype. Default numpy.uint16.
bin_type (dtype-like, optional) – Type used for histogram bin values (counts). Must be a signed or unsigned integer dtype. Default numpy.uint16.
kwargs (Unpack[_NumericTraits])

Notes

Unknown keyword arguments raise TypeError from the underlying TypedDict; values whose dtype kind violates the constraint above raise TypeError from the constructor.