ENSIGN Python API¶

`ensign.csv2tensor`¶

Converts tabular data into a sparse tensor.

`ensign.csv2tensor.csv2tensor`(filepaths[, …])	Creates a sparse tensor from one or more CSV files or Bro/Zeek logs.
`ensign.csv2tensor.df2tensor`(dfs[, …])	Variant of csv2tensor where in-memory DataFrames are passed instead of paths to files on disk.

`ensign.cp_decomp`¶

CANDECOMP-PARAFAC (CP) sparse tensor decomposition tools.

This module contains functions for reading, writing, and performing CP decompositions along with a class for representing decomposition results.

`ensign.cp_decomp.CPDecomp`()	Represents the results of a CP decomposition.
`ensign.cp_decomp.CPDecomp.compute_cpd_backtrack`(self)	Computes CP Decomposition backtracking information.
`ensign.cp_decomp.read_cp_decomp_dir`(decomp_dir)	Reads a CP decomposition from the filesystem.
`ensign.cp_decomp.write_cp_decomp_dir`(…[, …])	Writes a CP decomposition to the filesystem.
`ensign.cp_decomp.reconstruct_into`(decomp, tensor)	Reconstructs a CP decomposition ‘into’ a sparse tensor.
`ensign.cp_decomp.get_fit_per_entry`(decomp, …)	Calculates fit for each entry of a reconstructed tensor present in the original tensor.
`ensign.cp_decomp.are_close`(decomp0, decomp1)	Checks if weights and factor matrices are close according to tolerance.
`ensign.cp_decomp.cp_als`(sptensor, rank[, …])	Performs a CP decomposition using the alternating least squares (ALS) method.
`ensign.cp_decomp.cp_als_nn`(sptensor, rank[, …])	Performs a nonnegative CP decomposition using the alternating least squares (ALS) method.
`ensign.cp_decomp.cp_apr`(sptensor, rank[, …])	Performs a CP decomposition using the alternating Poisson regression (APR) method.
`ensign.cp_decomp.cp_apr_pdnr`(sptensor, rank)	Performs a CP decomposition using alternating Poisson regression (APR) with projected damped Newton row (PDNR) subproblem.
`ensign.cp_decomp.cp_apr_pqnr`(sptensor, rank)	Performs a CP decomposition using alternating Poisson regression (APR) with projected quasi-Newton row (PQNR) subproblem.

`ensign.sptensor`¶

Sparse tensor tools.

This module contains functions for reading, writing, and representing sparse tensors.

`ensign.sptensor.SPTensor`([order, nnz, …])	Represents a sparse tensor.
`ensign.sptensor.SPTensor.write`(self, outdir)	Writes a text representation of the sparse tensor to the given directory.
`ensign.sptensor.read_sptensor`(directory[, …])	Reads a sparse tensor file from the filesystem.
`ensign.sptensor.write_sptensor`(directory, …)	Writes a sparse tensor to the filesystem.

`ensign.visualize`¶

Visualize a tensor decomposition by plotting scores for selected components in each mode.

ensign.visualize.plot_component(decomp, comp_id)

Visualizes a specific component of a decomposition using blue-line charts.

`ensign.comp_top_k`¶

Module for investigating the contents of a tensor decomposition by listing the highest scoring labels in a component.

ensign.comp_top_k.get_top_k(factors, labels, …)

Computes the top k labels and corresponding scores for each mode for each component in the list.

`ensign.query_decomp`¶

Return all components containing the query label in the given mode. Up to k components are printed where <query label> has a non-zero score.

ensign.query_decomp.query_decomp(factors, …)

Compute all components containing the query label in the given mode.

`ensign.synchronize_labels`¶

Adjusts labels of multiple tensors and/or CP decompositions such that identical labels (e.g., “1/1/1970”, “10.1.1.1”) map to unique indices (e.g., 5, 437).

`ensign.synchronize_labels.synchronize_labels`(…)	Synchronizes labels between tensors.
`ensign.synchronize_labels.synchronize_labels_stream`(…)	Synchronizes specified modes between a base decomposition and an update tensor.

`ensign.decomp_diff`¶

ensign.decomp_diff.decomp_diff(decomps[, …])

Determine differences(similarities) between decompositions.