API reference

Triangle and diagonal indices

Name	Summary
`np.trilIndices`	Row and column indices of the lower triangle of an `n × m` array.
`np.triuIndices`	Row and column indices of the upper triangle of an `n × m` array.
`np.trilIndicesFrom`	`trilIndices` for the shape of a 2-D array.
`np.triuIndicesFrom`	`triuIndices` for the shape of a 2-D array.
`np.diagIndices`	Indices of the main diagonal of an `ndim`-dimensional array with all sides `n`.
`np.diagIndicesFrom`	`diagIndices` for an array whose sides are all equal (at least 2-D).
`np.maskIndices`	Indices where `maskFunc(np.ones([n, n]), k)` is nonzero, e.g.
`np.fillDiagonal`	Fills the main diagonal of `a` in place (values repeat cyclically).

`np.trilIndices`

np.trilIndices(n, [k], [m])

Row and column indices of the lower triangle of an n × m array.

NumPynumpy.tril_indices

Parameters

nnumber: Rows.
[k]number: Diagonal offset: 0 main, > 0 above, < 0 below. Default 0.
[m]number | null: Columns. Default n.

Returns

[NDArray, NDArray]

Example

TypeScript

np.trilIndices(3).map((a) => a.toArray()); // => [[0, 1, 1, 2, 2, 2], [0, 0, 1, 0, 1, 2]]

TypeScript declaration

np.trilIndices(n: number, k?: number | undefined, m?: number | null | undefined): NDArray[]

`np.triuIndices`

np.triuIndices(n, [k], [m])

Row and column indices of the upper triangle of an n × m array.

NumPynumpy.triu_indices

Parameters

nnumber: Rows.
[k]number: Diagonal offset: 0 main, > 0 above, < 0 below. Default 0.
[m]number | null: Columns. Default n.

Returns

[NDArray, NDArray]

Example

TypeScript

np.triuIndices(3, 1).map((a) => a.toArray()); // => [[0, 0, 1], [1, 2, 2]]

TypeScript declaration

np.triuIndices(n: number, k?: number | undefined, m?: number | null | undefined): NDArray[]

`np.trilIndicesFrom`

np.trilIndicesFrom(arr, [k])

trilIndices for the shape of a 2-D array.

NumPynumpy.tril_indices_from

Parameters

arrNDArray: 2-D array.
[k]number: Diagonal offset: 0 main, > 0 above, < 0 below. Default 0.

Returns

[NDArray, NDArray]

Example

TypeScript

np.trilIndicesFrom(np.zeros([2, 3])).map((a) => a.toArray()); // => [[0, 1, 1], [0, 0, 1]]

TypeScript declaration

np.trilIndicesFrom(arr: NDArray, k?: number | undefined): NDArray[]

`np.triuIndicesFrom`

np.triuIndicesFrom(arr, [k])

triuIndices for the shape of a 2-D array.

NumPynumpy.triu_indices_from

Parameters

arrNDArray: 2-D array.
[k]number: Diagonal offset: 0 main, > 0 above, < 0 below. Default 0.

Returns

[NDArray, NDArray]

Example

TypeScript

np.triuIndicesFrom(np.zeros([2, 3])).map((a) => a.toArray()); // => [[0, 0, 0, 1, 1], [0, 1, 2, 1, 2]]

TypeScript declaration

np.triuIndicesFrom(arr: NDArray, k?: number | undefined): NDArray[]

`np.diagIndices`

np.diagIndices(n, [ndim])

Indices of the main diagonal of an ndim-dimensional array with all sides n.

NumPynumpy.diag_indices

Parameters

nnumber: Side length.
[ndim]number: Dimensions. Default 2.

Returns

NDArray[]

Example

TypeScript

np.diagIndices(3).map((a) => a.toArray()); // => [[0, 1, 2], [0, 1, 2]]

TypeScript declaration

np.diagIndices(n: number, ndim?: number | undefined): NDArray[]

`np.diagIndicesFrom`

np.diagIndicesFrom(arr)

diagIndices for an array whose sides are all equal (at least 2-D).

NumPynumpy.diag_indices_from

Parameters

arrNDArray: Input array.

Returns

NDArray[]

Example

TypeScript

np.diagIndicesFrom(np.zeros([2, 2, 2])).map((a) => a.toArray()); // => [[0, 1], [0, 1], [0, 1]]

TypeScript declaration

np.diagIndicesFrom(arr: NDArray): NDArray[]

`np.maskIndices`

np.maskIndices(n, maskFunc, [k])

Indices where maskFunc(np.ones([n, n]), k) is nonzero, e.g. np.triu.

NumPynumpy.mask_indices

Parameters

nnumber: Side length.
maskFunc(m: NDArray, k: number) => NDArray: Mask function such as np.tril.
[k]number: Diagonal offset: 0 main, > 0 above, < 0 below. Default 0.

Returns

NDArray[]

Example

TypeScript

np.maskIndices(3, np.triu, 1).map((a) => a.toArray()); // => [[0, 0, 1], [1, 2, 2]]

TypeScript declaration

np.maskIndices(n: number, maskFunc: (m: NDArray, k: number) => NDArray, k?: number | undefined): NDArray[]

`np.fillDiagonal`

np.fillDiagonal(a, val, [options])

Fills the main diagonal of a in place (values repeat cyclically). For tall 2-D arrays, wrap: true restarts the diagonal after every ncols + 1 rows, like NumPy.

NumPynumpy.fill_diagonal

Parameters

aNDArray: Writeable array, at least 2-D (all sides equal when > 2-D).
valArrayLike: Scalar or values to write.
[options.wrap]boolean: Default false.

Returns

void

Example

TypeScript

const a = np.zeros([3, 3]);
np.fillDiagonal(a, [1, 2]);
a; // => [[1, 0, 0], [0, 2, 0], [0, 0, 1]]
const b = np.zeros([5, 2]);
np.fillDiagonal(b, 7, { wrap: true });
b; // => [[7, 0], [0, 7], [0, 0], [7, 0], [0, 7]]

TypeScript declaration

np.fillDiagonal(a: NDArray, val: ArrayInput, options?: { wrap?: boolean | undefined; } | undefined): void