Tests different invalid parameters that should raise an exception.

Tests different invalid parameters that should raise an exception.

Tests that the correct features are present in the descriptor.

Tests that the correct features are present in the desciptor.

Check that the broadening follows gaussian distribution.

Tests that the r_cut is correctly inferred from the weighting function.

Tests that the completely analytical partial power spectrum with the given basis corresponds to the easier-to-code but less performant numerical

Tests that invalid species throw an appropriate error.

Tests that calculating G4 and G5 works correctly when the j-k distance is larger than cutoff.

Tests that different metrics as defined by scikit-learn can be used.

Tests that a mix of integers and strings throws an appropriate error.

Tests the mu=1, nu=1 Darby et al. feature compression scheme ('mu1nu1' compression).

Tests that the normalization works correctly.

Tests that the normalization works correctly.

Tests that the padding used in constructing extended systems is sufficient.

Tests the correct peak locations and intensities are found. Args: system: The system to test geometry: geometry config gri

Tests the correct peak locations and intensities are found.

Tests that the C++ code performs better than the numpy version.

Tests that periodic images are handled correctly.

Tests that the output spectrum of various supercells of the same crystal is identical after it is normalized.

Tests that the final spectra does not change when translating atoms in a periodic cell. This is not trivially true unless the weight of distan

Test that periodic copies are correctly repeated and included in the output.

Tests that periodicity is not taken into account in Coulomb matrix even if the system is set as periodic.

Tests that the gto radial basis functions are orthonormal.

Test the method set_positions() of the System class

Test the method set_scaled_positions() of the System class

Tests that sparse features may also be used to construct the kernels.

Tests that the species are correctly determined.

Tests that species weights can be supplied and set correctly.

Tests that chemical symbols are correctly transformed into atomic numbers.

Tests the symmetries of the descriptor. Notice that sorted_l2 is not guaranteed to have any of the symmetries due to numerical issues with rows

Tests the symmetries of the descriptor. Notice that sorted_l2 is not guaranteed to have any of the symmetries due to numerical issues with rows

Tests the symmetries of the descriptor. Notice that sorted_l2 is not guaranteed to have any of the symmetries due to numerical issues with rows

Tests the symmetries of the descriptor.

Test that coordinates are correctly transformed from scaled to cartesian and back again.

Tests if arbitrary unit cells are accepted

Tests if arbitrary unit cells are accepted

Tests the verbose flag in create.

Tests that the ValleOganov subclass gives the same output as MBTR with the corresponding parameters.

Tests that the weighting done with C corresponds to the easier-to-code but less performant python version.

Tests that the kernel can be also calculated between two different sets, which is necessary for making predictions with kernel-based methods.