holoviews.core.data package#

Submodules#

Module contents#

class holoviews.core.data.DataConversion(element)[source]#

Bases: object

DataConversion is a very simple container object which can be given an existing Dataset Element and provides methods to convert the Dataset into most other Element types.

class holoviews.core.data.Dataset(data=None, kdims=None, vdims=None, **kwargs)[source]#

Bases: Element

Dataset provides a general baseclass for Element types that contain structured data and supports a range of data formats.

The Dataset class supports various methods offering a consistent way of working with the stored data regardless of the storage format used. These operations include indexing, selection and various ways of aggregating or collapsing the data with a supplied function.

Parameter Definitions

Parameters inherited from:

group = String(constant=True, default='Dataset', label='Group')

A string describing the data wrapped by the object.

datatype = List(bounds=(0, None), default=['dataframe', 'dictionary', 'grid', 'xarray', 'multitabular', 'spatialpandas', 'dask_spatialpandas', 'dask', 'cuDF', 'array', 'ibis'], label='Datatype')

A priority list of the data types to be used for storage on the .data attribute. If the input supplied to the element constructor cannot be put into the requested format, the next format listed will be used until a suitable format is found (or the data fails to be understood).

add_dimension(dimension, dim_pos, dim_val, vdim=False, **kwargs)[source]#

Adds a dimension and its values to the Dataset

Requires the dimension name or object, the desired position in the key dimensions and a key value scalar or array of values, matching the length or shape of the Dataset.

Parameters#

dimension

Dimension or dimension spec to add

dim_posint

Integer index to insert dimension at

dim_valscalar or ndarray

Dimension value(s) to add

vdim

Disabled, this type does not have value dimensions

**kwargs

Keyword arguments passed to the cloned element

Returns#

Cloned object containing the new dimension

aggregate(dimensions=None, function=None, spreadfn=None, **kwargs)[source]#

Aggregates data on the supplied dimensions.

Aggregates over the supplied key dimensions with the defined function or dim_transform specified as a tuple of the transformed dimension name and dim transform.

Parameters#

dimensions

Dimension(s) to aggregate on Default to all key dimensions

function

Aggregation function or transform to apply Supports both simple functions and dimension transforms

spreadfn

Secondary reduction to compute value spread Useful for computing a confidence interval, spread, or standard deviation.

**kwargs

Keyword arguments either passed to the aggregation function or to create new names for the transformed variables

Returns#

Returns the aggregated Dataset

clone(data=None, shared_data=True, new_type=None, link=True, *args, **overrides)[source]#

Clones the object, overriding data and parameters.

Parameters#

data

New data replacing the existing data

shared_databool, optional

Whether to use existing data

new_typeoptional

Type to cast object to

linkbool, optional

Whether clone should be linked Determines whether Streams and Links attached to original object will be inherited.

*args

Additional arguments to pass to constructor

**overrides

New keyword arguments to pass to constructor

Returns#

Cloned object

closest(coords=None, **kwargs)[source]#

Snaps coordinate(s) to closest coordinate in Dataset

Parameters#

coords

List of coordinates expressed as tuples

**kwargs

Coordinates defined as keyword pairs

Returns#

List of tuples of the snapped coordinates

Raises#

NotImplementedError

Raised if snapping is not supported

columns(dimensions=None)[source]#

Convert dimension values to a dictionary.

Returns a dictionary of column arrays along each dimension of the element.

Parameters#

dimensions

Dimensions to return as columns

Returns#

Dictionary of arrays for each dimension

compute()[source]#

Computes the data to a data format that stores the daata in memory, e.g. a Dask dataframe or array is converted to a Pandas DataFrame or NumPy array.

Returns#

Dataset with the data stored in in-memory format

property dataset#

The Dataset that this object was created from

dframe(dimensions=None, multi_index=False)[source]#

Convert dimension values to DataFrame.

Returns a pandas dataframe of columns along each dimension, either completely flat or indexed by key dimensions.

Parameters#

dimensions

Dimensions to return as columns

multi_index

Convert key dimensions to (multi-)index

Returns#

DataFrame of columns corresponding to each dimension

dimension_values(dimension, expanded=True, flat=True)[source]#

Return the values along the requested dimension.

Parameters#

dimension

The dimension to return values for

expandedbool, optional

Whether to expand values Whether to return the expanded values, behavior depends on the type of data:

  • Columnar: If false returns unique values

  • Geometry: If false returns scalar values per geometry

  • Gridded: If false returns 1D coordinates

flatbool, optional

Whether to flatten array

Returns#

NumPy array of values along the requested dimension

get_dimension_type(dim)[source]#

Get the type of the requested dimension.

Type is determined by Dimension.type attribute or common type of the dimension values, otherwise None.

Parameters#

dimension

Dimension to look up by name or by index

Returns#

Declared type of values along the dimension

groupby(dimensions=None, container_type=<class 'holoviews.core.spaces.HoloMap'>, group_type=None, dynamic=False, **kwargs)[source]#

Groups object by one or more dimensions

Applies groupby operation over the specified dimensions returning an object of type container_type (expected to be dictionary-like) containing the groups.

Parameters#

dimensions

Dimension(s) to group by

container_type

Type to cast group container to

group_type

Type to cast each group to

dynamic

Whether to return a DynamicMap

**kwargs

Keyword arguments to pass to each group

Returns#

Returns object of supplied container_type containing the groups. If dynamic=True returns a DynamicMap instead.

property iloc#

Returns iloc indexer with support for columnar indexing.

Returns an iloc object providing a convenient interface to slice and index into the Dataset using row and column indices. Allow selection by integer index, slice and list of integer indices and boolean arrays.

Examples :

  • Index the first row and column:

    dataset.iloc[0, 0]

  • Select rows 1 and 2 with a slice:

    dataset.iloc[1:3, :]

  • Select with a list of integer coordinates:

    dataset.iloc[[0, 2, 3]]

map(map_fn, specs=None, clone=True)[source]#

Map a function to all objects matching the specs

Recursively replaces elements using a map function when the specs apply, by default applies to all objects, e.g. to apply the function to all contained Curve objects:

dmap.map(fn, hv.Curve)

Parameters#

map_fn : Function to apply to each object specs : List of specs to match

List of types, functions or type[.group][.label] specs to select objects to return, by default applies to all objects.

clone

Whether to clone the object or transform inplace

Returns#

Returns the object after the map_fn has been applied

property ndloc#

Returns ndloc indexer with support for gridded indexing.

Returns an ndloc object providing nd-array like indexing for gridded datasets. Follows NumPy array indexing conventions, allowing for indexing, slicing and selecting a list of indices on multi-dimensional arrays using integer indices. The order of array indices is inverted relative to the Dataset key dimensions, e.g. an Image with key dimensions ‘x’ and ‘y’ can be indexed with image.ndloc[iy, ix], where iy and ix are integer indices along the y and x dimensions.

Examples :

  • Index value in 2D array:

    dataset.ndloc[3, 1]

  • Slice along y-axis of 2D array:

    dataset.ndloc[2:5, :]

  • Vectorized (non-orthogonal) indexing along x- and y-axes:

    dataset.ndloc[[1, 2, 3], [0, 2, 3]]

options(*args, clone=True, **kwargs)[source]#

Applies simplified option definition returning a new object.

Applies options on an object or nested group of objects in a flat format returning a new object with the options applied. If the options are to be set directly on the object a simple format may be used, e.g.:

obj.options(cmap=’viridis’, show_title=False)

If the object is nested the options must be qualified using a type[.group][.label] specification, e.g.:

obj.options(‘Image’, cmap=’viridis’, show_title=False)

or using:

obj.options({‘Image’: dict(cmap=’viridis’, show_title=False)})

Identical to the .opts method but returns a clone of the object by default.

Parameters#

*args: Sets of options to apply to object

Supports a number of formats including lists of Options objects, a type[.group][.label] followed by a set of keyword options to apply and a dictionary indexed by type[.group][.label] specs.

backendoptional

Backend to apply options to Defaults to current selected backend

clonebool, optional

Whether to clone object Options can be applied inplace with clone=False

**kwargs: Keywords of options

Set of options to apply to the object

Returns#

Returns the cloned object with the options applied

persist()[source]#

Persists the results of a lazy data interface to memory to speed up data manipulation and visualization. If the particular data backend already holds the data in memory this is a no-op. Unlike the compute method this maintains the same data type.

Returns#

Dataset with the data persisted to memory

property pipeline#

Chain operation that evaluates the sequence of operations that was used to create this object, starting with the Dataset stored in dataset property

range(dim, data_range=True, dimension_range=True)[source]#

Return the lower and upper bounds of values along dimension.

Parameters#

dim

The dimension to compute the range on.

data_rangebool

Compute range from data values

dimension_rangebool

Include Dimension ranges Whether to include Dimension range and soft_range in range calculation

Returns#

Tuple containing the lower and upper bound

reduce(dimensions=None, function=None, spreadfn=None, **reductions)[source]#

Applies reduction along the specified dimension(s).

Allows reducing the values along one or more key dimension with the supplied function. Supports two signatures:

Reducing with a list of dimensions, e.g.:

ds.reduce([‘x’], np.mean)

Defining a reduction using keywords, e.g.:

ds.reduce(x=np.mean)

Parameters#

dimensions

Dimension(s) to apply reduction on Defaults to all key dimensions

function

Reduction operation to apply, e.g. numpy.mean

spreadfn

Secondary reduction to compute value spread Useful for computing a confidence interval, spread, or standard deviation.

**reductions

Keyword argument defining reduction Allows reduction to be defined as keyword pair of dimension and function

Returns#

The Dataset after reductions have been applied.

reindex(kdims=None, vdims=None)[source]#

Reindexes Dataset dropping static or supplied kdims

Creates a new object with a reordered or reduced set of key dimensions. By default drops all non-varying key dimensions.x

Parameters#

kdimsoptional

New list of key dimensionsx

vdimsoptional

New list of value dimensions

Returns#

Reindexed object

relabel(label=None, group=None, depth=0)[source]#

Clone object and apply new group and/or label.

Applies relabeling to children up to the supplied depth.

Parameters#

labelstr, optional

New label to apply to returned object

groupstr, optional

New group to apply to returned object

depthint, optional

Depth to which relabel will be applied If applied to container allows applying relabeling to contained objects up to the specified depth

Returns#

Returns relabelled object

sample(samples=None, bounds=None, closest=True, **kwargs)[source]#

Samples values at supplied coordinates.

Allows sampling of element with a list of coordinates matching the key dimensions, returning a new object containing just the selected samples. Supports multiple signatures:

Sampling with a list of coordinates, e.g.:

ds.sample([(0, 0), (0.1, 0.2), …])

Sampling a range or grid of coordinates, e.g.:

1D : ds.sample(3) 2D : ds.sample((3, 3))

Sampling by keyword, e.g.:

ds.sample(x=0)

Parameters#

samples : List of nd-coordinates to sample bounds

Bounds of the region to sample Defined as two-tuple for 1D sampling and four-tuple for 2D sampling.

closest

Whether to snap to closest coordinates

**kwargs

Coordinates specified as keyword pairs Keywords of dimensions and scalar coordinates

Returns#

Element containing the sampled coordinates

select(selection_expr=None, selection_specs=None, **selection)[source]#

Applies selection by dimension name

Applies a selection along the dimensions of the object using keyword arguments. The selection may be narrowed to certain objects using selection_specs. For container objects the selection will be applied to all children as well.

Selections may select a specific value, slice or set of values:

  • value: Scalar values will select rows along with an exact

    match, e.g.:

    ds.select(x=3)

  • slice: Slices may be declared as tuples of the upper and

    lower bound, e.g.:

    ds.select(x=(0, 3))

  • values: A list of values may be selected using a list or

    set, e.g.:

    ds.select(x=[0, 1, 2])

  • predicate expression: A holoviews.dim expression, e.g.:

    from holoviews import dim ds.select(selection_expr=dim(‘x’) % 2 == 0)

Parameters#

selection_exprholoviews.dim predicate expression

specifying selection.

selection_specsList of specs to match on

A list of types, functions, or type[.group][.label] strings specifying which objects to apply the selection on.

**selection: Dictionary declaring selections by dimension

Selections can be scalar values, tuple ranges, lists of discrete values and boolean arrays

Returns#

Returns an Dimensioned object containing the selected data or a scalar if a single value was selected

property shape#

Returns the shape of the data.

sort(by=None, reverse=False)[source]#

Sorts the data by the values along the supplied dimensions.

Parameters#

by

Dimension(s) to sort by

reversebool, optional

Reverse sort order

Returns#

Sorted Dataset

property to#

Returns the conversion interface with methods to convert Dataset

transform(*args, **kwargs)[source]#

Transforms the Dataset according to a dimension transform.

Transforms may be supplied as tuples consisting of the dimension(s) and the dim transform to apply or keyword arguments mapping from dimension(s) to dim transforms. If the arg or kwarg declares multiple dimensions the dim transform should return a tuple of values for each.

A transform may override an existing dimension or add a new one in which case it will be added as an additional value dimension.

Parameters#

args

Specify the output arguments and transforms as a tuple of dimension specs and dim transforms

dropbool

Whether to drop all variables not part of the transform

keep_indexbool
Whether to keep indexes

Whether to apply transform on datastructure with index, e.g. pandas.Series or xarray.DataArray, (important for dask datastructures where index may be required to align datasets).

kwargs

Specify new dimensions in the form new_dim=dim_transform

Returns#

Transformed dataset with new dimensions

class holoviews.core.data.PipelineMeta(classname, bases, classdict)[source]#

Bases: ParameterizedMetaclass

holoviews.core.data.concat(datasets, datatype=None)[source]#

Concatenates collection of datasets along NdMapping dimensions.

Concatenates multiple datasets wrapped in an NdMapping type along all of its dimensions. Before concatenation all datasets are cast to the same datatype, which may be explicitly defined or implicitly derived from the first datatype that is encountered. For columnar data concatenation adds the columns for the dimensions being concatenated along and then concatenates all the old and new columns. For gridded data a new axis is created for each dimension being concatenated along and then hierarchically concatenates along each dimension.

Parameters#

datasets

NdMapping of Datasets to concatenate

datatype

Datatype to cast data to before concatenation

Returns#

Concatenated dataset

holoviews.core.data.disable_pipeline()[source]#

Disable PipelineMeta class from storing pipelines.