Provides convenience datasets for unittesting.
Also performs testing of storing/reloading datasets into hdf5 file if cfg.getboolean(‘tests’, ‘use hdf datasets’
Functions
autocorrelated_noise(ds, sr, cutoff[, lfnl, ...]) | Generate a dataset with samples being temporally autocorrelated noise. |
chirp_linear(n_instances[, n_features, ...]) | Generates simple dataset for linear regressions |
dumb_feature_binary_dataset() | Very simple binary (2 labels) dataset |
dumb_feature_dataset() | Create a very simple dataset with 2 features and 3 labels |
generate_testing_datasets(*arg, **kwargs) | |
get_mv_pattern(s2n) | Simple multivariate dataset |
get_random_rotation(ns[, nt, data]) | Return some random rotation (or rotation + dim reduction) matrix |
linear1d_gaussian_noise([size, slope, ...]) | A straight line with some Gaussian noise. |
linear_awgn([size, intercept, slope, ...]) | Generate a dataset from a linear function with AWGN |
load_datadb_demo_blockfmri([path, roi]) | Loads the block-design demo dataset from PyMVPA dataset DB. |
load_datadb_tutorial_data([path, roi]) | Loads the block-design demo dataset from PyMVPA dataset DB. |
load_example_fmri_dataset() | Load minimal fMRI dataset that is shipped with PyMVPA. |
multiple_chunks(func, n_chunks, *args, **kwargs) | Replicate datasets multiple times raising different chunks |
noisy_2d_fx(size_per_fx, dfx, sfx, center[, ...]) | Yet another generator of random dataset |
normal_feature_dataset([perlabel, nlabels, ...]) | Generate a univariate dataset with normal noise and specified means. |
pure_multivariate_signal(patterns[, ...]) | Create a 2d dataset with a clear multivariate signal, but no univariate information. |
random_affine_transformation(ds[, ...]) | Distort a dataset by random scale, shift, and rotation. |
reseed_rng() | Decorator to assure the use of MVPA_SEED while running the test |
saveload_warehouse() | Store all warehouse datasets into HDF5 and reload them. |
sin_modulated(n_instances, n_features[, ...]) | Generate a (quite) complex multidimensional non-linear dataset |
wr1996([size]) | Generate ‘6d robot arm’ dataset (Williams and Rasmussen 1996) |
Classes
Dataset(samples[, sa, fa, a]) | Generic storage class for datasets with multiple attributes. |
HollowSamples([shape, sid, fid, dtype]) | Samples container that doesn’t store samples. |
OddEvenPartitioner([usevalues]) | Create odd and even partitions based on a sample attribute. |