Private API reference

generative_model

metric

metrics.mk_betweenAssocRelError(ds[, ...])	Evaluate the relative error of the between-set association strength.
metrics.mk_betweenAssocRelError_cv(ds, cv_assoc)	Evaluate the relative error of the cross-validated between-set association strength.
metrics.mk_meanBetweenAssocRelError(ds, cv_assoc)	Evaluate the relative error of the between-set association strength as the average of the in-sample and cross-validated value.
metrics.mk_weightError(ds[, suffix, prefix])	Evaluate the weight error.
metrics.mk_scoreError(ds[, prefix])	Evaluate the score error.
metrics.mk_loadingError(ds[, prefix])	Evaluate the loading error.
metrics.mk_crossloadingError(ds[, prefix])	Evaluate the cross-loading error.

sample_size

interpolation.calc_n_required(metric, ...[, ...])	Calculate required sample sizes for a given metric.
interpolation.calc_n_required_all_metrics(ds)	Calculate n_required for 5 commonly used metrics, as well as maximum across metrics.
interpolation.calc_max_n_required(*n_requireds)	Finds the maximum sample size across datasets.
linear_model.do_fit_lm(ds, n_reqs[, ...])	Fits a linear model to outcome data.
linear_model.prep_data_for_lm(ds, n_reqs, ...)	Prepare outcome data for use with linear model.
linear_model.fit_linear_model(criterion, model)	Fit a linear model to outcome data.
linear_model.get_lm_coefs(model, criterion, ...)	Get linear model coefficients.
linear_model._save_linear_model(model[, ...])	Save linear model.

data

loaders.load_metaanalysis_outcomes(px, py, n)	Load previously generated outcomes for a specific parameter set.
loaders.load_metaanalysis([data_home, fetch])	Load table of metaanalysis features.
loaders._fetch_synthanad(fname, path)	Retrieve file from remote repository.
loaders._check_data_home(data_home[, subfolder])	Check data_home directory
loaders._check_outcome_data_exists(fname, path)	Check if a given outcome data file exists and possibly retrieve it if not.

util

check_positive_definite(Sigma, noise_level)	Check if a matrix is positive definite.
align_weights(v, vtrue[, copy, return_sign])	Align vectors in rows of v such that they have a positive dot-product with vtrue.
nPerFtr2n(nPerFtr[, py])	Convert a DataArray with elements representing "n_per_ftr" into "n"