mlx.core.array

class array

An N-dimensional array object.

__init__(self: array, val: scalar | list | tuple | ndarray | array, dtype: Dtype | None = None)

Methods

__init__(self, val[, dtype])
abs(self, *[, stream])	See abs().
all(self[, axis, keepdims, stream])	See all().
any(self[, axis, keepdims, stream])	See any().
argmax(self[, axis, keepdims, stream])	See argmax().
argmin(self[, axis, keepdims, stream])	See argmin().
astype(self, dtype[, stream])	Cast the array to a specified type.
conj(self, *[, stream])	See conj().
cos(self, *[, stream])	See cos().
cummax(self[, axis, reverse, inclusive, stream])	See cummax().
cummin(self[, axis, reverse, inclusive, stream])	See cummin().
cumprod(self[, axis, reverse, inclusive, stream])	See cumprod().
cumsum(self[, axis, reverse, inclusive, stream])	See cumsum().
diag(self[, k, stream])	Extract a diagonal or construct a diagonal matrix.
diagonal(self[, offset, axis1, axis2, stream])	See diagonal().
exp(self, *[, stream])	See exp().
flatten(self[, start_axis, end_axis, stream])	See flatten().
item(self)	Access the value of a scalar array.
log(self, *[, stream])	See log().
log10(self, *[, stream])	See log10().
log1p(self, *[, stream])	See log1p().
log2(self, *[, stream])	See log2().
logcumsumexp(self[, axis, reverse, ...])	See logcumsumexp().
logsumexp(self[, axis, keepdims, stream])	See logsumexp().
max(self[, axis, keepdims, stream])	See max().
mean(self[, axis, keepdims, stream])	See mean().
min(self[, axis, keepdims, stream])	See min().
moveaxis(self, source, destination, *[, stream])	See moveaxis().
prod(self[, axis, keepdims, stream])	See prod().
reciprocal(self, *[, stream])	See reciprocal().
reshape(self, *shape[, stream])	Equivalent to reshape() but the shape can be passed either as a tuple or as separate arguments.
round(self[, decimals, stream])	See round().
rsqrt(self, *[, stream])	See rsqrt().
sin(self, *[, stream])	See sin().
split(self, indices_or_sections[, axis, stream])	See split().
sqrt(self, *[, stream])	See sqrt().
square(self, *[, stream])	See square().
squeeze(self[, axis, stream])	See squeeze().
std(self[, axis, keepdims, ddof, stream])	See std().
sum(self[, axis, keepdims, stream])	See sum().
swapaxes(self, axis1, axis2, *[, stream])	See swapaxes().
tolist(self)	Convert the array to a Python list.
transpose(self, *axes[, stream])	Equivalent to transpose() but the axes can be passed either as a tuple or as separate arguments.
var(self[, axis, keepdims, ddof, stream])	See var().
view(self, dtype, *[, stream])	See view().

Attributes

T	Equivalent to calling self.transpose() with no arguments.
at	Used to apply updates at the given indices.
dtype	The array's Dtype.
imag	The imaginary part of a complex array.
itemsize	The size of the array's datatype in bytes.
nbytes	The number of bytes in the array.
ndim	The array's dimension.
real	The real part of a complex array.
shape	The shape of the array as a Python tuple.
size	Number of elements in the array.