# Full Changelog¶

## 0.7¶

- Added new
`matrix_operations_3d`

module in the`utils`

subpackage that contains Numpy functions related to three-dimensional rotations. - Added new
`inertia_tensor_per_object`

function in`mock_observables`

that calculates the inertia tensor for a set of points and masses in a periodic box. - Added new option for the marked correlation functions to accommodate counting pairs of points passing a variable merger ratio criteria
- Added
`fuzzy_digitize`

function to`utils`

sub-package. - Added
`sliding_conditional_percentile`

function to`utils`

sub-package. - Added new
`resample_x_to_match_y`

function to`halotools.utils`

. - Renamed old implementation of
`conditional_abunmatch`

to`conditional_abunmatch_bin_based`

- Added new bin-free implementation of
`conditional_abunmatch`

. - Added new utils function
`bijective_distribution_matching`

- Added new
`load_um_binary_sfr_catalog`

function to load SFR catalogs from UniverseMachine into memory - Added new
`return_indexes`

feature to conditional_abunmatch function. See https://github.com/astropy/halotools/pull/913. - New function
`mean_delta_sigma`

replaces old`delta_sigma`

function. See #955.

## 0.6 (2017-12-15)¶

- Changed the API for mock_observables.pair_counters.npairs_s_mu which now requires
`mu_bins`

to be in the conventional mu=cos(theta_LOS) format instead of mu=sin(theta_LOS). See https://github.com/astropy/halotools/pull/768 - Added new
`mock_observables`

functions`radial_distance`

and`radial_distance_and_velocity`

functions. See https://github.com/astropy/halotools/pull/782 - Added new
`distribution_matching_indices`

function. See https://github.com/astropy/halotools/pull/795. - Removed
`max_sample_size`

keyword argument from all`mock_observables`

functions. - Added new
`weighted_npairs_s_mu`

function to`mock_observables`

sub-package. See Issue #810 - Added standalone function to apply redshift-space distortions. See Issue #806.
- Added
`num_lines_header`

optional keyword argument to TabularAsciiReader - Added
`wp_jackknife`

function. See https://github.com/astropy/halotools/pull/814. - Fixed bug in the normalization of the covariance matrices in the
`tpcf_jackknife`

function. See https://github.com/astropy/halotools/issues/815. - Added
`rp_pi_tpcf_jackknife`

function - See #822 - Fixed bug in estimate_ngals method for case of assembly-biased occupation models. See https://github.com/astropy/halotools/issues/801
- Removed obsolete bounds_enforcing_decorator_factory function - see https://github.com/astropy/halotools/issues/756
- Fixed bug impacting ability of python 3.6 users to access downsampled particle data - See Issues #821, #826, and #831.
- Fixed bug in isolation criteria for calculations without PBCs. See Issue #776.
- Fixed bug in
`weighted_npairs_s_mu`

when called in parallel. See https://github.com/astropy/halotools/issues/837. - Fixed bug in
`radial_profile_3d`

when called in parallel. See https://github.com/astropy/halotools/issues/854. - Fixed bug in
`radial_profile_3d`

when called without periodic boundary conditions. See https://github.com/astropy/halotools/issues/862 - Fixed bug in
`velocity_marked_npairs_3d`

and`velocity_marked_npairs_xy_z`

when called with default arguments. See https://github.com/astropy/halotools/issues/836. - Changed the
`weight_func_id`

numbers associated with weighting functions for`velocity_marked_npairs_3d`

and`velocity_marked_npairs_xy_z`

where e.g. 11 is now 1, etc. - Added new
`test_installation`

feature that dramatically shortens the length of time users need to spend verifying their installation

## 0.5 (2017-05-31)¶

- Added SubhaloPhaseSpace class for modeling HOD-style satellite velocities with subhalos.
- The
`Lbox`

attribute of all Halotools halo catalogs is now stored as an array of length 3 instead of a scalar. - Fixed bug in multi-threaded calculations using npairs_per_object_3d function.
- Added
`empirical_models.abunmatch`

sub-package providing conditional abundance matching functionality - Fixed bug in mock population when using the seed keyword, resolving Issue #672, https://github.com/astropy/halotools/issues/672.
- Added
`mock_observables.hod_from_mock`

convenience function to calculate the HOD from a mock galaxy sample - Performance enhancements to marked correlation functions
- Fixed bug in
`mock_observables.pair_counters.npairs_s_mu`

that impacted`mock_observables.s_mu_tpcf`

- Added Zu & Mandelbaum 2015 and Zu & Mandelbaum 2016 HOD models
- Modified internals of
`mock_observables.delta_sigma`

function, including an API change by removing the`pi_max`

argument. Additionally included new`mock_observables.delta_sigma_from_precomputed`

function to compute the results from a set of pre-computed pairs. See https://github.com/astropy/halotools/pull/696. - Fixed factor of 2 error in tpcf_multipole, resolving https://github.com/astropy/halotools/issues/651
- Complete refactoring of the halotools/empirical_models/phase_space_models. No changes to either the API or behavior of any associated classes or functions.
- Addition of two new classes
`halotools.empirical_models.BiasedNFWPhaseSpace`

and`halotools.empirical_models.SFRBiasedNFWPhaseSpace`

for NFW satellites with biased concentrations and Jeans solutions. - The
`gal_types`

attribute of HODModelFactory-produced models is now sorted so that`centrals`

always appears before`satellites`

. This new default behavior is more common for interdependent occupation models, where satellite abundance depends upon central galaxy characteristics, rather than the other way around. See https://github.com/astropy/halotools/pull/729 - Added new keyword arguments to
`return_xyz_formatted_array`

function enabling application of redshift-space distortions for galaxy samples at higher redshift. Previously, the user needed to do this manually). Default behavior of this function is unchanged, provided users had not locally modified the`sim_defaults`

module to have set`default_redshift`

greater than zero.

## 0.4 (2016-08-11)¶

- All models now support an optional
`seed`

keyword argument, allowing for deterministic Monte Carlo realizations of models. As a result of this feature, it is now mandatory that all user-defined models obey a new constraint. Any function appearing in the`mock_generation_calling_sequence`

must now use the kwargs Python syntax to catch any additional inputs passed to these functions by the MockFactory. - Added relative_positions_and_velocities function to mock_observables
- Fixed little h bug in the Behroozi10SmHm class. This impacts the Leauthaud11Cens and Leauthaud11Sats classes, as well as the
`leauthaud11`

composite model. - Fixed bug in mock_observables.pair_counters.npairs_per_object_3d. See https://github.com/astropy/halotools/issues/606.
- New counts_in_cells sub-package in mock_observables
- HodMockFactory has new estimate_ngals feature
- Fixed buggy behavior for two-point functions called for logical branch
`do_auto=True, do_cross=False`

- Performance enhancement of isolation_functions by 50% - 300%, depending on numerical regime.
- Updated all catalogs to version_name=``halotools_v0p4``, resolving the bug pointed out in https://github.com/astropy/halotools/issues/598.
- Performance enhancement of npairs_s_mu function by 10-100x after cleaning cython engine of python objects.

## 0.3 (2016-06-28)¶

- Removed distant_observer_redshift function from mock_survey module
- Removed -march=native compiler flag to resolve installation problems on some architectures

## 0.2 (2016-06-09)¶

- Halotools is now Python 3.x compatible
- Halotools mock_observables package has been given a complete overhaul, adding many new cythonized engines and pair counters (listed below). Functions are 30% - 50x faster, depending on numerical regime. Overhauled engines include velocity_marked_npairs_3d, velocity_marked_npairs_xy_z, npairs_per_object_3d, npairs_s_mu, npairs_jackknife_3d, npairs_projected, npairs_xy_z, npairs_3d, marked_npairs_3d and marked_npairs_xy_z
- Added new utils.crossmatch function
- Added new mock_observables.radial_profile_3d function
- All isolation_functions now return boolean
`is_isolated`

rather than its inverse`has_neighbor`

- Fixed a bug in mock_observables.delta_sigma. See https://github.com/astropy/halotools/issues/523
- Fixed bug in mock_observables.tpcf_jackknife. See https://github.com/astropy/halotools/issues/513
- Deleted mock_observables.nearest_neighbor function

## 0.1 (2016-03-13)¶

- Initial release