inspect_viz

Core

Data

Data source for visualizations.

Data sources can be created from any standard Python data frame (e.g. Pandas, Polars, etc.) or from a path pointing to a data file in a standard format (e.g. csv, parquet, etc.)

Source

class Data

Attributes

columns list[str]

Column names for data source.

Methods

from_dataframe

Create Data from a standard Python data frame (e.g. Pandas, Polars, PyArrow, etc.).

Source

@classmethod
def from_dataframe(cls, df: IntoDataFrame) -> "Data"

df IntoDataFrame

Data frame to read.

from_file

Create Data from a data file (e.g. csv, parquet, feather, etc.).

Source

@classmethod
def from_file(cls, file: str | PathLike[str]) -> "Data"

file str | PathLike[str]

File to read data from. Supported formats include csv, json, xslx, parquet, feather, sas7bdat, dta, and fwf.

Component

Data visualization component (input, plot, mark, table, layout, etc.).

Visualization components are Jupyter widgets that can be used in any notebook or Jupyter based publishing system.

See the documentation on inputs, plots, marks, and interactors for details on available components.

Source

class Component(AnyWidget)

Params

Selection

Selection that can be filtered by inputs and other selections.

Selection types include:

• Selection.intersect() for intersecting clauses (logical “and”)
• Selection.union() for unionone clauses (logical “or”)
• Selection.single() for a single clause only
• Selection.crossfilter() for a cross-filtered intersection

Source

class Selection(str)

Methods

intersect

Create a new Selection instance with an intersect (conjunction) resolution strategy.

Source

@classmethod
def intersect(
    cls,
    cross: bool = False,
    empty: bool = False,
    include: Union["Selection", list["Selection"]] | None = None,
) -> "Selection"

cross bool

Boolean flag indicating cross-filtered resolution. If true, selection clauses will not be applied to the clients they are associated with.

empty bool

Boolean flag indicating if a lack of clauses should correspond to an empty selection with no records. This setting determines the default selection state.

include Union[Selection, list[Selection]] | None

Upstream selections whose clauses should be included as part of the new selection. Any clauses published to upstream selections will be relayed to the new selection.

union

Create a new Selection instance with a union (disjunction) resolution strategy.

Source

@classmethod
def union(
    cls,
    cross: bool = False,
    empty: bool = False,
    include: Union["Selection", list["Selection"]] | None = None,
) -> "Selection"

cross bool

Boolean flag indicating cross-filtered resolution. If true, selection clauses will not be applied to the clients they are associated with.

empty bool

Boolean flag indicating if a lack of clauses should correspond to an empty selection with no records. This setting determines the default selection state.

include Union[Selection, list[Selection]] | None

Upstream selections whose clauses should be included as part of the new selection. Any clauses published to upstream selections will be relayed to the new selection.

single

Create a new Selection instance with a singular resolution strategy that keeps only the most recent selection clause.

Source

@classmethod
def single(
    cls,
    cross: bool = False,
    empty: bool = False,
    include: Union["Selection", list["Selection"]] | None = None,
) -> "Selection"

cross bool

Boolean flag indicating cross-filtered resolution. If true, selection clauses will not be applied to the clients they are associated with.

empty bool

Boolean flag indicating if a lack of clauses should correspond to an empty selection with no records. This setting determines the default selection state.

include Union[Selection, list[Selection]] | None

Upstream selections whose clauses should be included as part of the new selection. Any clauses published to upstream selections will be relayed to the new selection.

crossfilter

Create a new Selection instance with a cross-filtered intersect resolution strategy.

Source

@classmethod
def crossfilter(
    cls,
    empty: bool = False,
    include: Union["Selection", list["Selection"]] | None = None,
) -> "Selection"

empty bool

Boolean flag indicating if a lack of clauses should correspond to an empty selection with no records. This setting determines the default selection state.

include Union[Selection, list[Selection]] | None

Upstream selections whose clauses should be included as part of the new selection. Any clauses published to upstream selections will be relayed to the new selection.

Param

Parameter that can be bound from inputs.

Source

class Param(str)

Attributes

id str

Unique id (automatically generated).

default ParamValue

Default value.

ParamValue

Type alias for parameter values (scalar or sequence of scalars).

Source

ParamValue: TypeAlias = (
    int | float | bool | str | datetime | Sequence[int | float | bool | str]
)

Options

options

Inspect Viz global options.

Source

options: Options = Options(output_format="auto")

options_context

Context manager for temporarily overriding global options.

Source

@contextmanager
def options_context(**kwargs: Unpack[OptionsArgs]) -> Iterator[None]

**kwargs Unpack[OptionsArgs]

Options to override within the context.

Options

Inspect Viz global options.

Source

class Options(SimpleNamespace)

Attributes

output_format Literal['auto', 'js', 'png']

Output format for components.

Defaults to “auto”, which resolves to “js” (interactive plots and tables) in all contexts except Quarto PDF output (which uses “png”). Specify “png” to always write static PNG images instead (interactive features will be disabled in this case).