sciedm.CCM

class sciedm.CCM(columns=None, target=None, E=1, libSizes=None, Tp=0, tau=-1, knn=0, sample=30, random_state=None, exclusionRadius=0, validLib=[], embedded=False, noTime=False, includeData=False, mpMethod=None, sharedMB=0.01, parallel=False, verbose=False)

Convergent Cross Mapping using per-subsample KDTree construction.

Construction stores parameters only. Call Validate and Embed to prepare the precomputed state, or call Project which ensures all preparation stages have run.

Parameters:

dataFrameDataFrame

Input data with named columns.

columnsstr or [str]

Column name(s) for the library variable.

targetstr or [str]

Column name for the target variable. Only target[0] used for Takens

Eint

Embedding dimension.

Tpint

Prediction horizon in time steps. Default 0.

knnint

Number of nearest neighbors. Default 0 is set to E+1 if Takens.

tauint

Embedding delay (negative = rows back in time). Default -1.

exclusionRadiusint

Temporal exclusion radius for neighbor queries. Default 0.

validLibarray-like of bool

Boolean mask of valid library rows. Default: all True.

embeddedbool

If True, columns are already the embedding. Default False.

libSizesstr or [int]

Library sizes. Can be overridden in Project.

sampleint

Number of subsamples per library size. Default 100.

seedint or None

RNG seed for reproducibility.

includeDatabool

Add variance of sample CCM correlation at each libSize. Default False.

parallelbool or int

Worker count. Default True.

mpMethodstr or None

Multiprocessing start method. Default None.

sharedMBfloat

Total worker data size threshold in MB. Below this, worker arrays are passed via Pool initargs (pickle); above, they are placed in OS shared memory for zero-copy access.

Attributes:

is_fitted_bool

A boolean indicating whether the estimator has been fitted.

n_features_in_int

Number of features seen during fit.

feature_names_in_ndarray of shape (n_features_in_,)

Names of features seen during fit. Defined only when X has feature names that are all strings.

libMeans_DataFrame

DataFrame with columns: [“Library Size”, column:target, target:column] listing cross map predictive correlation at each library size

Returns:

self.libMeans_ DataFrame

Examples

>>> from sciedm import CCM
>>> from pandas import DataFrame
>>> df = DataFrame({'time':[t for t in range(1,21)],
                    'x':[1,1,3,4,5,5,6,8,3,3,2,6,5,5,9,3,5,1,8,2],
                    'y':[5,5,7,8,6,6,7,8,2,2,2,8,3,3,7,5,3,1,1,1]})
>>> ccm = CCM(columns='x',target='y',E=2,libSizes=[10,12,16,18])
>>> ccm.fit(df)
CCM(E=2, columns='x', libSizes=[10, 12, 16, 18], target='y')
>>> ccm.transform(df)

Embed()

Build delay embeddings and apply all validity filters.

For each direction, the embedding is the horizontal stack of per-variable E-dimensional time-delay embeddings. Column labels follow the pyEDM convention: 'Var(t-0)', 'Var(t-1)', etc. for tau=-1 shifts.

When embedded=True, the columns listed in columns are taken directly as the pre-built embedding (no delay construction), and their DataFrame column names serve as labels.

Project(libSizes=<object object>, sample=<object object>, random_state=<object object>, includeData=<object object>, parallel=<object object>, mpMethod=<object object>, sharedMB=<object object>, verbose=<object object>)

Run Convergent Cross Mapping over the specified library sizes.

At each library size L and each sample iteration:

• L points are randomly selected as the library.

• ALL M valid points are the prediction set.

• A KDTree is built from the L library embedding vectors.

• All M points query k nearest neighbors from the L-point tree.

• Simplex predictions and Pearson r are computed over all M.

Returns:

self.libMeans_pandas.DataFrame

Validate()

Parse and validate inputs.

fit(X, y=None)

Initialize lib & pred indices, embed data, find neighbors

Parameters:

X{array-like}, shape (n_samples, n_features)

Observed data to be embedded, or used as embedding.

yaccepted but silently ignored — embedding is target-independent

Returns:

selfobject

Returns self.

fit_transform(X, y=None, **fit_params)

Fit to data, then transform it.

Fits transformer to X and y with optional parameters fit_params and returns a transformed version of X.

Parameters:

Xarray-like of shape (n_samples, n_features)

Input samples.

yarray-like of shape (n_samples,) or (n_samples, n_outputs), default=None

Target values (None for unsupervised transformations).

**fit_paramsdict

Additional fit parameters. Pass only if the estimator accepts additional params in its fit method.

Returns:

X_newndarray array of shape (n_samples, n_features_new)

Transformed array.

get_feature_names_out(input_features=None)

set_output for downstream pipeline compatibility

The ‘output’ is a DataFrame with [LibSize, Col:Target, Target:Col]

get_metadata_routing()

Get metadata routing of this object.

Please check User Guide on how the routing mechanism works.

Returns:

routingMetadataRequest

A MetadataRequest encapsulating routing information.

get_params(deep=True)

Get parameters for this estimator.

Parameters:

deepbool, default=True

If True, will return the parameters for this estimator and contained subobjects that are estimators.

Returns:

paramsdict

Parameter names mapped to their values.

set_output(transform='pandas')

Set output container.

See Introducing the set_output API for an example on how to use the API.

Parameters:

transform{“default”, “pandas”, “polars”}, default=None

Configure output of transform and fit_transform.

• "default": Default output format of a transformer

• "pandas": DataFrame output

• "polars": Polars output

• None: Transform configuration is unchanged

Added in version 1.4: "polars" option was added.

Returns:

selfestimator instance

Estimator instance.

set_params(**params)

Set the parameters of this estimator.

The method works on simple estimators as well as on nested objects (such as Pipeline). The latter have parameters of the form <component>__<parameter> so that it’s possible to update each component of a nested object.

Parameters:

**paramsdict

Estimator parameters.

Returns:

selfestimator instance

Estimator instance.

transform(X)

Call CrossMap() for forward and reverse mappings.

The output is independent of the specific X passed here (CCM was fit on the training X); this method exists so the estimator participates correctly in sklearn Pipelines.

Parameters:

X{array-like}, shape (n_samples, n_features)

Not used but accepted for sklearn convention.

Returns:

DataFrame

CCM output: columns [LibSize, columns:target, target:columns]