utility

Utilities mainly for internal use.

Functions

array_str(a[, nprint])	Stringification of an array.
attribute_repr_string(inst_str, attr_str)	Return a repr string for an attribute that respects line width.
cache_arguments(function)	Decorate function to cache the result with given arguments.
complex_dtype(dtype[, default])	Return complex counterpart of dtype if existing, else default.
conj_exponent(exp)	Conjugate exponent exp / (exp - 1).
dedent(string[, indent_str, max_levels])	Revert the effect of indentation.
dtype_repr(dtype)	Stringify dtype for repr with default for int and float.
dtype_str(dtype)	Stringify dtype for str with default for int and float.
indent(string[, indent_str])	Return a copy of string indented by indent_str.
is_string(obj)	Return True if obj behaves like a string, False else.
method_repr_string(inst_str, meth_str[, ...])	Return a repr string for a method that respects line width.
nd_iterator(shape)	Iterator over n-d cube with shape.
npy_printoptions(**extra_opts)	Context manager to temporarily set NumPy print options.
npy_random_seed(seed)	Context manager to temporarily set the NumPy random generator seed.
pkg_supports(feature, pkg_version, pkg_feat_dict)	Return bool indicating whether a package supports feature.
real_dtype(dtype[, default])	Return the real counterpart of dtype if existing.
repr_string(outer_string, inner_strings[, ...])	Return a pretty string for repr.
run_from_ipython()	If the process is run from IPython.
signature_string(posargs, optargs[, sep, mod])	Return a stringified signature from given arguments.
signature_string_parts(posargs, optargs[, mod])	Return stringified arguments as tuples.
unique(seq)	Return the unique values in a sequence.
writable_array(obj, **kwargs)	Context manager that casts obj to a numpy.array and saves changes.