tlc.client.reduce.pacmap

Dimensionality reduction with the PaCMAP algorithm.

Module Contents

Classes

Class	Description
PaCMAPTableArgs	Arguments specific to the PaCMAP reduction method.
PaCMAPReduction	Perform dimensionality reduction on columns of tables using the PaCMAP algorithm.

API

class tlc.client.reduce.pacmap.PaCMAPTableArgs

Bases: tlc.client.reduce.reduction_method.ReducerArgs

Arguments specific to the PaCMAP reduction method.

See PaCMAPTable for more information.

Initialize self. See help(type(self)) for accurate signature.

n_components: int = None

The number of dimensions in the reduced space (default 2)

n_neighbors: int = None

The number of neighbors to use when constructing the nearest neighbor graph (default to 10 for dataset whose sample size is smaller than 10000. For large dataset whose sample size (n) is larger than 10000, the default value is: 10 + 15 * (log10(n) - 4))

MN_ratio: float = None

the ratio of the number of mid-near pairs to the number of neighbors (default 0.5)

FP_ratio: float = None

the ratio of the number of further pairs to the number of neighbors, (default 2)

distance: str = None

The distance metric to use for the nearest neighbor graph (default ‘euclidean’)

lr: float = None

Learning rate of the AdaGrad optimizer. (default 1)

num_iters: int = None

Number of iterations. Default 450. 450 iterations is enough for most dataset to converge.

verbose: bool = None

Print the progress of PaCMAP. (default False)

apply_pca: bool = None

Whether pacmap should apply PCA to the data before constructing the k-Nearest Neighbor graph (default True)

random_state: int | None = None

Random state for the random number generator (default 0)

source_embedding_column: str | None = None

The name of the column containing the embedding in the input table. If None, all columns are considered.

target_embedding_column: str | None = None

The name of the column containing the embedding in the output table. If None, the column is named after the input column.

retain_source_embedding_column: bool = None

Whether to retain the source embedding column in the output table (default False).

class tlc.client.reduce.pacmap.PaCMAPReduction(reducer_args: tlc.client.reduce.reduction_method._ReducerArgsType | None = None)

Bases: tlc.client.reduce.reduction_method.ReductionMethod[tlc.client.reduce.pacmap.PaCMAPTableArgs]

Perform dimensionality reduction on columns of tables using the PaCMAP algorithm.

Params reducer_args:

A dictionary of arguments which are specific to the reduction method.

default_args() → tlc.client.reduce.pacmap.PaCMAPTableArgs

Returns the default arguments for the PaCMAP reduction method.

fit_and_apply_reduction(producer: tlc.core.objects.table.Table, consumers: list[tlc.core.objects.table.Table]) → dict[tlc.core.url.Url, tlc.core.url.Url]

fit_reduction_method(table: tlc.core.objects.table.Table, column: str) → tlc.core.url.Url

Fits a PaCMAPTable and returns the model URL

apply_reduction_method(table: tlc.core.objects.table.Table, fit_table_url: tlc.core.url.Url, column: str) → tlc.core.url.Url | None