fiery.bounds.padding

This module reimplements torch.nn.functional.pad and torch.roll with a larger set of boundary conditions.

Functions:

Name	Description
`pad`	Pad a tensor
`roll`	Roll a tensor
`ensure_shape`	Pad/crop a tensor so that it has a given shape

Functions:

pad

pad(inp: Tensor, padsize: SequenceOrScalar[int], mode: SequenceOrScalar[BoundLike] = 'constant', value: Number = 0, side: str | None = None) -> Tensor

Pad a tensor.

This function is a bit more generic than torch's native pad (torch.nn.functional.pad), but probably a bit slower:

works with any input type
works with arbitrarily large padding size
crops the tensor for negative padding values
implements additional padding modes

When used with defaults parameters (side=None), it behaves exactly like torch.nn.functional.pad

Boundary modes

Like in PyTorch's pad, boundary modes include:

'circular' (or 'dft')
'mirror' (or 'dct1')
'reflect' (or 'dct2')
'replicate' (or 'nearest')
'constant' (or 'zero')

as well as the following new modes:

'antimirror' (or 'dst1')
'antireflect' (or 'dst2')

Side modes

Side modes are 'pre' (or 'left'), 'post' (or 'right'), 'both' or None.

If side is not None, inp.dim() values (or less) should be provided.
If side is None, twice as many values should be provided, indicating different padding sizes for the 'pre' and 'post' sides.
If the number of padding values is less than the dimension of the input tensor, zeros are prepended.

Parameters:

Name	Type	Description	Default
`inp`	`tensor`	Input tensor	required
`padsize`	`SequenceOrScalar[int]`	Amount of padding in each dimension.	required
`mode`	`SequenceOrScalar[BoundLike]`	Padding mode	`'constant'`
`value`	`scalar`	Value to pad with in mode `'constant'`.	`0`
`side`	`{'pre', 'post', 'both', None}`	Use padsize to pad on left side (`'pre'`), right side (`'post'`) or both sides (`'both'`). If `None`, the padding side for the left and right sides should be provided in alternate order.	`None`

Returns:

Name	Type	Description
`out`	`tensor`	Padded tensor.

ensure_shape

ensure_shape(inp: Tensor, shape: SequenceOrScalar[int | None], mode: SequenceOrScalar[BoundLike] = 'constant', value: Number = 0, side: str = 'post', ceil: bool = False) -> Tensor

Pad/crop a tensor so that it has a given shape

Parameters:

Name	Type	Description	Default
`inp`	`tensor`	Input tensor	required
`shape`	`SequenceOrScalar[int]`	Output shape. A value of `None` for a given dimension keeps that dimension's current size.	required
`mode`	`SequenceOrScalar[BoundLike]`	Boundary mode	`'constant'`
`value`	`scalar`	Value for mode `'constant'`	`0`
`side`	`{'pre', 'post', 'both'}`	Side to crop/pad	`'post'`
`ceil`	`bool`	When `side='both'` and the amount to crop/pad cannot be split evenly, put the extra unit on the `'pre'` side (`True`) rather than the `'post'` side (`False`).	`False`

Returns:

Name	Type	Description
`out`	`tensor`	Padded tensor with shape `shape`

roll

roll(inp: Tensor, shifts: SequenceOrScalar[int] = 1, dims: SequenceOrScalar[int] | None = None, bound: SequenceOrScalar[BoundLike] = 'circular') -> Tensor

Like torch.roll, but with any boundary condition

Warning

When dims is None, we do not flatten but shift all dimensions. This differs from the behavior of torch.roll .

Parameters:

Name	Type	Description	Default
`inp`	`tensor`	Input	required
`shifts`	`SequenceOrScalar[int]`	Amount by which to roll. Positive shifts to the right, negative to the left.	`1`
`dims`	`SequenceOrScalar[int]`	Dimensions to roll. By default, shifts apply to all dimensions if a scalar, or to the last N if a sequence.	`None`
`bound`	`SequenceOrScalar[BoundLike]`	Boundary condition	`'circular'`

Returns:

Name	Type	Description
`out`	`tensor`	Rolled tensor