rc.data.models

Models for data storage.

Classes

DesignMatrix	The familiar user format of DesignMatrix which is fat (has many columns).
NormalDesignMatrix	The internal format of DesignMatrix, which is thin (has few columns).
Normalization	Normalization of a Repo.
Repo	A Repository of data and models. Informally a dataset and all the things we'd like to do to it.

Module Contents

class DesignMatrix(path, data=None, **options)

Bases: rc.base.Table

The familiar user format of DesignMatrix which is fat (has many columns).

Parameters:

• path (Store)

• data (rc.base.definitions.Self | rc.base.definitions.Pd.DataFrame | None)

• options (rc.base.definitions.Any)

Label

Class attribute aliasing acceptable Types for column (or index) labels.

skeleton: rc.base.Pd.DataFrame

DataFrame of the minimal, skeleton DesignMatrix.

defaultOptions: rc.base.MetaData

Default file handling DesignMatrix.Options().

classmethod create(path, src, columns_in_l='')

Reformat the NormalDesignMatrix in src as a Self(DesignMatrix).

Parameters:

• path (rc.base.Store.Path) – The Path to store the DesignMatrix created, overwritten if existing.

• src (NormalDesignMatrix) – The NormalDesignMatrix to reformat.

• columns_in_l (Label)

Return type:

rc.base.Self

Returns: The NormalDesignMatrix created at dst.

classmethod copy(src, dst='')

Reformat this DesignMatrix to a NormalDesignMatrix.

Parameters:

• src (rc.base.Self) – The DesignMatrix to reformat.

• dst (rc.base.Store.Path) – Optional Path to the NormalDesignMatrix. Defaults to '', which overwrites src.

Return type:

NormalDesignMatrix

Returns: The NormalDesignMatrix created at dst.

class NormalDesignMatrix(path, data=None, **options)

Bases: DesignMatrix

The internal format of DesignMatrix, which is thin (has few columns).

Parameters:

• path (Store)

• data (rc.base.definitions.Self | rc.base.definitions.Pd.DataFrame | None)

• options (rc.base.definitions.Any)

create(path, src)

Reformat the NormalDesignMatrix in src as a Self(DesignMatrix).

Parameters:

• path (rc.base.Store.Path) – The Path to store the DesignMatrix created, overwritten if existing.

• src (NormalDesignMatrix) – The NormalDesignMatrix to reformat.

Return type:

rc.base.Self

Returns: The NormalDesignMatrix created at dst.

copy(src, dst='')

Reformat this DesignMatrix to a NormalDesignMatrix.

Parameters:

• src (rc.base.Self) – The DesignMatrix to reformat.

• dst (rc.base.Store.Path) – Optional Path to the NormalDesignMatrix. Defaults to '', which overwrites src.

Return type:

NormalDesignMatrix

Returns: The NormalDesignMatrix created at dst.

class Normalization(path, **tables)

Bases: rc.base.DataBase

Normalization of a Repo.

Parameters:

• path (rc.base.Store.Path)

• tables (rc.base.Table | PD.DataFrame)

class NamedTables

Bases: rc.base.NamedTuple

Must be overridden.

__call__(name)

Returns the Table named name.

Parameters:

name (str)

Return type:

rc.base.Table | rc.base.Matrix | rc.base.MetaData

options: NamedTables[rc.base.MetaData]

Class attribute of the form NamedTables(**{names[i]: options[i], ...}). Override as necessary for bespoke Table.options. Elements of options[i] found in Table.writeOptions populate self[i].options.write, the remainder populate self[i].options.read.

__call__(**meta)

Optimize and update self.

Parameters:

**meta (rc.base.Any) – Optimization MetaData.

Return type:

rc.base.Self

Returns: self

classmethod create(path, data, **meta)

Create a Normalization in path.

Parameters:

• path (rc.base.Store.Path) – The folder to store the Normalization in. Need not exist, any existing Tables will be overwritten if it does.

• **meta (rc.base.Any) – Optimization MetaData.

• data (DesignMatrix)

• **meta

Return type:

rc.base.Self

Returns: The Normalization created.

class Repo(path, **tables)

Bases: rc.base.DataBase

A Repository of data and models. Informally a dataset and all the things we’d like to do to it.

Parameters:

• path (Store)

• tables (Table | rc.base.definitions.Pd.DataFrame)

class NamedTables

Bases: rc.base.NamedTuple

Must be overridden.

__call__(name)

Returns the Table named name.

Parameters:

name (str)

Return type:

rc.base.Table | rc.base.Matrix | rc.base.MetaData

options: NamedTables[rc.base.MetaData]

Class attribute of the form NamedTables(**{names[i]: options[i], ...}). Override as necessary for bespoke Table.options. Elements of options[i] found in Table.writeOptions populate self[i].options.write, the remainder populate self[i].options.read.

property fold

The current fold.

__len__()

1 + K proper folds in self.

Return type:

int

__getitem__(fold)

Indexer returns the Path (s) to the Folds indexed or sliced by fold.

Parameters:

fold (int | slice)

Return type:

rc.base.Path | rc.base.Tuple[rc.base.Path, Ellipsis]

__setitem__(fold, tables)

Indexer creates the Fold (s) named or sliced by name.

Parameters:

• fold (int | slice)

• tables (rc.base.Table | rc.base.Matrix | rc.base.Tuple[rc.base.Table | rc.base.Matrix, Ellipsis])

__call__(**meta)

Optimize and update self.

Parameters:

**meta (rc.base.Any) – Optimization MetaData.

Return type:

rc.base.Self

Returns: self