types ¶

Type Aliases:

BotFieldLeftShift –

Type alias for the bottom field's horizontal shift in pixels.
BotFieldTopShift –

Type alias for the bottom field's vertical shift in pixels.
Center –

Type alias for the center point of the sigmoid curve, determining the midpoint of the transition.
FieldShift –

Type alias for shifts in interlaced content.
LeftShift –

Type alias for horizontal shift in pixels (left).
Slope –

Type alias for the slope of the sigmoid curve, controlling the steepness of the transition.
TopFieldLeftShift –

Type alias for the top field's horizontal shift in pixels.
TopFieldTopShift –

Type alias for the top field's vertical shift in pixels.
TopShift –

Type alias for vertical shift in pixels (top).

Classes:

BorderHandling –

Border padding strategy used when a clip requires alignment padding.
SampleGridModel –

Sampling grid alignment model.

BotFieldLeftShift ¶

BotFieldLeftShift = float

Type alias for the bottom field's horizontal shift in pixels.

Used when processing interlaced video to describe the horizontal shift of the bottom field.

BotFieldTopShift ¶

BotFieldTopShift = float

Type alias for the bottom field's vertical shift in pixels.

Used when processing interlaced video to describe the vertical shift of the bottom field.

Center ¶

Center = float

Type alias for the center point of the sigmoid curve, determining the midpoint of the transition.

FieldShift ¶

FieldShift = tuple[
    TopShift | tuple[TopFieldTopShift, BotFieldTopShift],
    LeftShift | tuple[TopFieldLeftShift, BotFieldLeftShift],
]

Type alias for shifts in interlaced content.

Represents separate shifts for top and bottom fields.

LeftShift ¶

LeftShift = float

Type alias for horizontal shift in pixels (left).

Represents the amount of horizontal offset when scaling a video.

Slope ¶

Slope = float

Type alias for the slope of the sigmoid curve, controlling the steepness of the transition.

TopFieldLeftShift ¶

TopFieldLeftShift = float

Type alias for the top field's horizontal shift in pixels.

Used when processing interlaced video to describe the horizontal shift of the top field.

TopFieldTopShift ¶

TopFieldTopShift = float

Type alias for the top field's vertical shift in pixels.

Used when processing interlaced video to describe the vertical shift of the top field.

TopShift ¶

TopShift = float

Type alias for vertical shift in pixels (top).

Represents the amount of vertical offset when scaling a video.

BorderHandling ¶

Bases: CustomIntEnum

Border padding strategy used when a clip requires alignment padding.

Methods:

from_param –

Return the enum value from a parameter.
pad_amount –

Return required padding.
prepare_clip –

Apply required padding and adjust shift.
value –

Attributes:

MIRROR –

Assume the image was resized with mirror padding.
REPEAT –

Assume the image was resized with extend padding, where the outermost row was extended infinitely far.
ZERO –

Assume the image was resized with zero padding.

MIRROR `class-attribute` `instance-attribute` ¶

MIRROR = 0

Assume the image was resized with mirror padding.

REPEAT `class-attribute` `instance-attribute` ¶

REPEAT = 2

Assume the image was resized with extend padding, where the outermost row was extended infinitely far.

ZERO `class-attribute` `instance-attribute` ¶

ZERO = 1

Assume the image was resized with zero padding.

from_param `classmethod` ¶

from_param(value: Any, func_except: FuncExcept | None = None) -> Self

Return the enum value from a parameter.

Parameters:

value ¶
(Any) –

Value to instantiate the enum class.
func_except ¶
(FuncExcept | None, default: None ) –

Exception function.

Returns:

Self –

Enum value.

Raises:

NotFoundEnumValue –

Variable not found in the given enum.

Source code in jetpytools/enums/base.py

@classmethod
def from_param(cls, value: Any, func_except: FuncExcept | None = None) -> Self:
    """
    Return the enum value from a parameter.

    Args:
        value: Value to instantiate the enum class.
        func_except: Exception function.

    Returns:
        Enum value.

    Raises:
        NotFoundEnumValue: Variable not found in the given enum.
    """
    func_except = func_except or cls.from_param

    try:
        return cls(value)
    except (ValueError, TypeError):
        pass

    if isinstance(func_except, tuple):
        func_name, var_name = func_except
    else:
        func_name, var_name = func_except, repr(cls)

    raise NotFoundEnumValueError(
        'The given value for "{var_name}" argument must be a valid {enum_name}, not "{value}"!\n'
        "Valid values are: [{readable_enum}].",
        func_name,
        var_name=var_name,
        enum_name=cls,
        value=value,
        readable_enum=(f"{name} ({value!r})" for name, value in cls.__members__.items()),
        reason=value,
    ) from None

pad_amount ¶

pad_amount(
    clip: VideoNode,
    width: int,
    height: int,
    shift: tuple[TopShift, LeftShift],
    kernel_radius: int,
    src_width: float,
    src_height: float,
) -> tuple[int, int, int, int]

Return required padding.

Parameters:

clip ¶
(VideoNode) –

Input clip.
width ¶
(int) –

Output width.
height ¶
(int) –

Output height.
shift ¶
(tuple[TopShift, LeftShift]) –

Current (top, left) shift.
kernel_radius ¶
(int) –

Kernel radius.
src_width ¶
(float) –

Width source region.
src_height ¶
(float) –

Height source region.

Returns:

tuple[int, int, int, int] –

Padding amount.

Source code in vskernels/types.py

def pad_amount(
    self,
    clip: vs.VideoNode,
    width: int,
    height: int,
    shift: tuple[TopShift, LeftShift],
    kernel_radius: int,
    src_width: float,
    src_height: float,
) -> tuple[int, int, int, int]:
    """
    Return required padding.

    Args:
        clip: Input clip.
        width: Output width.
        height: Output height.
        shift: Current (top, left) shift.
        kernel_radius: Kernel radius.
        src_width: Width source region.
        src_height: Height source region.

    Returns:
        Padding amount.
    """
    top_shift, left_shift = shift

    w_factor = kernel_radius * max(src_width / width, 1)
    left, right = (
        ceil((w_factor - left_shift) / 2) * 2**clip.format.subsampling_w,
        ceil((w_factor + left_shift) / 2) * 2**clip.format.subsampling_w,
    )

    h_factor = kernel_radius * max(src_height / height, 1)
    top, bottom = (
        ceil((h_factor - top_shift) / 2) * 2**clip.format.subsampling_h,
        ceil((h_factor + top_shift) / 2) * 2**clip.format.subsampling_h,
    )

    return (left, right, top, bottom)

prepare_clip ¶

prepare_clip(
    clip: VideoNode,
    width: int,
    height: int,
    shift: tuple[TopShift, LeftShift],
    kernel_radius: int,
    **kwargs: Any
) -> tuple[VideoNode, tuple[TopShift, LeftShift]]

Apply required padding and adjust shift.

Parameters:

clip ¶
(VideoNode) –

Input clip.
width ¶
(int) –

Output width.
height ¶
(int) –

Output height.
shift ¶
(tuple[TopShift, LeftShift]) –

Current (top, left) shift.
kernel_radius ¶
(int) –

Kernel radius.
**kwargs ¶
(Any, default: {} ) –

Optional src_width/src_height.

Returns:

tuple[VideoNode, tuple[TopShift, LeftShift]] –

(padded clip, updated shift).

Source code in vskernels/types.py

def prepare_clip(
    self,
    clip: vs.VideoNode,
    width: int,
    height: int,
    shift: tuple[TopShift, LeftShift],
    kernel_radius: int,
    **kwargs: Any,
) -> tuple[vs.VideoNode, tuple[TopShift, LeftShift]]:
    """
    Apply required padding and adjust shift.

    Args:
        clip: Input clip.
        width: Output width.
        height: Output height.
        shift: Current (top, left) shift.
        kernel_radius: Kernel radius.
        **kwargs: Optional src_width/src_height.

    Returns:
        (padded clip, updated shift).
    """

    if self is BorderHandling.MIRROR:
        return (clip, shift)

    src_width = fallback(kwargs.get("src_width"), clip.width)
    src_height = fallback(kwargs.get("src_height"), clip.height)

    shift = kwargs.pop("src_stop", shift[0]), kwargs.pop("src_left", shift[1])

    left, right, top, bottom = self.pad_amount(
        clip,
        width,
        height,
        shift,
        kernel_radius,
        src_width,
        src_height,
    )

    match self:
        case BorderHandling.ZERO:
            padded = padder.COLOR(clip, left, right, top, bottom)
        case BorderHandling.REPEAT:
            padded = padder.REPEAT(clip, left, right, top, bottom)
        case _:
            assert_never(self)

    shift = tuple(s + c for s, c in zip(shift, (top, left)))  # type: ignore

    return padded, shift

value ¶

value() -> int

Source code in jetpytools/enums/base.py

@enum_property
def value(self) -> int: ...

SampleGridModel ¶

Bases: CustomIntEnum

Sampling grid alignment model.

While match edges will align the edges of the outermost pixels in the target image, match centers will instead align the centers of the outermost pixels.

Here's a visual example for a 3x1 image upsampled to 7x1:

Match edges:

+-------------+-------------+-------------+
|      ·      |      ·      |      ·      |
+-------------+-------------+-------------+
↓                                         ↓
+-----+-----+-----+-----+-----+-----+-----+
|  ·  |  ·  |  ·  |  ·  |  ·  |  ·  |  ·  |
+-----+-----+-----+-----+-----+-----+-----+

Match centers:

+-----------------+-----------------+-----------------+
|        ·        |        ·        |        ·        |
+-----------------+-----------------+-----------------+
         ↓                                   ↓
      +-----+-----+-----+-----+-----+-----+-----+
      |  ·  |  ·  |  ·  |  ·  |  ·  |  ·  |  ·  |
      +-----+-----+-----+-----+-----+-----+-----+

For a more detailed explanation, refer to this page: https://entropymine.com/imageworsener/matching/.

The formula for calculating values we can use during desampling is simple:

width: base_width * (target_width - 1) / (base_width - 1)
height: base_height * (target_height - 1) / (base_height - 1)

Methods:

__call__ –

Apply sampling model to sizes and shift.
for_dst –

Apply grid model using destination sizes.
for_src –

Apply grid model using source sizes.
from_param –

Return the enum value from a parameter.
value –

Attributes:

MATCH_CENTERS –

Align pixel centers.
MATCH_EDGES –

Align edges.

MATCH_CENTERS `class-attribute` `instance-attribute` ¶

MATCH_CENTERS = 1

Align pixel centers.

MATCH_EDGES `class-attribute` `instance-attribute` ¶

MATCH_EDGES = 0

Align edges.

call ¶

__call__(
    width: float,
    height: float,
    src_width: float,
    src_height: float,
    shift: tuple[float, float],
    kwargs: dict[str, Any],
) -> tuple[dict[str, Any], tuple[float, float]]

Apply sampling model to sizes and shift.

Parameters:

width ¶
(float) –

Destination width.
height ¶
(float) –

Destination height.
src_width ¶
(float) –

Current source width.
src_height ¶
(float) –

Current source height.
shift ¶
(tuple[float, float]) –

Top, left sampling shift.
kwargs ¶
(dict[str, Any]) –

Parameter dict to update.

Returns:

tuple[dict[str, Any], tuple[float, float]] –

(updated kwargs, updated shift).

Source code in vskernels/types.py

def __call__(
    self,
    width: float,
    height: float,
    src_width: float,
    src_height: float,
    shift: tuple[float, float],
    kwargs: dict[str, Any],
) -> tuple[dict[str, Any], tuple[float, float]]:
    """
    Apply sampling model to sizes and shift.

    Args:
        width: Destination width.
        height: Destination height.
        src_width: Current source width.
        src_height: Current source height.
        shift: Top, left sampling shift.
        kwargs: Parameter dict to update.

    Returns:
        (updated kwargs, updated shift).
    """

    if self is SampleGridModel.MATCH_CENTERS:
        src_width = src_width * (width - 1) / (src_width - 1)
        src_height = src_height * (height - 1) / (src_height - 1)

        shift = kwargs.pop("src_stop", shift[0]), kwargs.pop("src_left", shift[1])

        kwargs.update(src_width=src_width, src_height=src_height)
        shift_x, shift_y, *_ = tuple(
            (x / 2 + sh for x, sh in zip(((height - src_height), (width - src_width)), shift))
        )
        shift = shift_x, shift_y

    return kwargs, shift

for_dst ¶

for_dst(
    clip: VideoNode,
    width: int,
    height: int,
    shift: tuple[float, float],
    **kwargs: Any
) -> tuple[dict[str, Any], tuple[float, float]]

Apply grid model using destination sizes.

Parameters:

clip ¶
(VideoNode) –

Source clip.
width ¶
(int) –

Destination width.
height ¶
(int) –

Destination height.
shift ¶
(tuple[float, float]) –

Current shift.
**kwargs ¶
(Any, default: {} ) –

Optional src_width/src_height.

Returns:

tuple[dict[str, Any], tuple[float, float]] –

(updated kwargs, updated shift).

Source code in vskernels/types.py

def for_dst(
    self, clip: vs.VideoNode, width: int, height: int, shift: tuple[float, float], **kwargs: Any
) -> tuple[dict[str, Any], tuple[float, float]]:
    """
    Apply grid model using destination sizes.

    Args:
        clip: Source clip.
        width: Destination width.
        height: Destination height.
        shift: Current shift.
        **kwargs: Optional src_width/src_height.

    Returns:
        (updated kwargs, updated shift).
    """

    src_width = fallback(kwargs.get("src_width"), width)
    src_height = fallback(kwargs.get("src_height"), height)

    return self(src_width, src_height, width, height, shift, kwargs)

for_src ¶

for_src(
    clip: VideoNode,
    width: int,
    height: int,
    shift: tuple[float, float],
    **kwargs: Any
) -> tuple[dict[str, Any], tuple[float, float]]

Apply grid model using source sizes.

Parameters:

clip ¶
(VideoNode) –

Source clip (fallback for src dimensions).
width ¶
(int) –

Source width.
height ¶
(int) –

Source height.
shift ¶
(tuple[float, float]) –

Current shift.
**kwargs ¶
(Any, default: {} ) –

Optional overrides.

Returns:

tuple[dict[str, Any], tuple[float, float]] –

(updated kwargs, updated shift).

Source code in vskernels/types.py

def for_src(
    self, clip: vs.VideoNode, width: int, height: int, shift: tuple[float, float], **kwargs: Any
) -> tuple[dict[str, Any], tuple[float, float]]:
    """
    Apply grid model using source sizes.

    Args:
        clip: Source clip (fallback for src dimensions).
        width: Source width.
        height: Source height.
        shift: Current shift.
        **kwargs: Optional overrides.

    Returns:
        (updated kwargs, updated shift).
    """

    src_width = fallback(kwargs.get("src_width"), clip.width)
    src_height = fallback(kwargs.get("src_height"), clip.height)

    return self(width, height, src_width, src_height, shift, kwargs)

from_param `classmethod` ¶

from_param(value: Any, func_except: FuncExcept | None = None) -> Self

Return the enum value from a parameter.

Parameters:

value ¶
(Any) –

Value to instantiate the enum class.
func_except ¶
(FuncExcept | None, default: None ) –

Exception function.

Returns:

Self –

Enum value.

Raises:

NotFoundEnumValue –

Variable not found in the given enum.

Source code in jetpytools/enums/base.py

@classmethod
def from_param(cls, value: Any, func_except: FuncExcept | None = None) -> Self:
    """
    Return the enum value from a parameter.

    Args:
        value: Value to instantiate the enum class.
        func_except: Exception function.

    Returns:
        Enum value.

    Raises:
        NotFoundEnumValue: Variable not found in the given enum.
    """
    func_except = func_except or cls.from_param

    try:
        return cls(value)
    except (ValueError, TypeError):
        pass

    if isinstance(func_except, tuple):
        func_name, var_name = func_except
    else:
        func_name, var_name = func_except, repr(cls)

    raise NotFoundEnumValueError(
        'The given value for "{var_name}" argument must be a valid {enum_name}, not "{value}"!\n'
        "Valid values are: [{readable_enum}].",
        func_name,
        var_name=var_name,
        enum_name=cls,
        value=value,
        readable_enum=(f"{name} ({value!r})" for name, value in cls.__members__.items()),
        reason=value,
    ) from None

value ¶

value() -> int

Source code in jetpytools/enums/base.py

@enum_property
def value(self) -> int: ...

types ¶

BotFieldLeftShift ¶

BotFieldTopShift ¶

Center ¶

FieldShift ¶

LeftShift ¶

Slope ¶

TopFieldLeftShift ¶

TopFieldTopShift ¶

TopShift ¶

BorderHandling ¶

MIRROR class-attribute instance-attribute ¶

REPEAT class-attribute instance-attribute ¶

ZERO class-attribute instance-attribute ¶

from_param classmethod ¶

value ¶

func_except ¶

pad_amount ¶

clip ¶

width ¶

height ¶

shift ¶

kernel_radius ¶

src_width ¶

src_height ¶

prepare_clip ¶

clip ¶

width ¶

height ¶

shift ¶

kernel_radius ¶

**kwargs ¶

value ¶

SampleGridModel ¶

MATCH_CENTERS class-attribute instance-attribute ¶

MATCH_EDGES class-attribute instance-attribute ¶

__call__ ¶

width ¶

height ¶

src_width ¶

src_height ¶

shift ¶

kwargs ¶

for_dst ¶

clip ¶

width ¶

height ¶

shift ¶

**kwargs ¶

for_src ¶

clip ¶

width ¶

height ¶

shift ¶

**kwargs ¶

from_param classmethod ¶

value ¶

func_except ¶

value ¶

MIRROR `class-attribute` `instance-attribute` ¶

REPEAT `class-attribute` `instance-attribute` ¶

ZERO `class-attribute` `instance-attribute` ¶

from_param `classmethod` ¶

`value` ¶

`func_except` ¶

`clip` ¶

`width` ¶

`height` ¶

`shift` ¶

`kernel_radius` ¶

`src_width` ¶

`src_height` ¶

`clip` ¶

`width` ¶

`height` ¶

`shift` ¶

`kernel_radius` ¶

`kwargs`** ¶

MATCH_CENTERS `class-attribute` `instance-attribute` ¶

MATCH_EDGES `class-attribute` `instance-attribute` ¶

call ¶

`width` ¶

`height` ¶

`src_width` ¶

`src_height` ¶

`shift` ¶

`kwargs` ¶

`clip` ¶

`width` ¶

`height` ¶

`shift` ¶

`kwargs`** ¶

`clip` ¶

`width` ¶

`height` ¶

`shift` ¶

`kwargs`** ¶

from_param `classmethod` ¶

`value` ¶

`func_except` ¶