Cython Helper Functions

In addition to the below functions, the lapack, blas, special libraries from Scipy are available, as well as the libc library as math.

starlord.cy_tools.uniform_lpdf(double x, double xmin, double xmax) → double

starlord.cy_tools.uniform_ppf(double x, double xmin, double xmax) → double

starlord.cy_tools.normal_lpdf(double x, double mean, double sigma) → double

starlord.cy_tools.normal_ppf(double p, double mean, double sigma) → double

starlord.cy_tools.normal_cdf(double x, double mean, double sigma) → double

starlord.cy_tools.beta_lpdf(double x, double alpha, double beta) → double

starlord.cy_tools.beta_ppf(double p, double alpha, double beta) → double

starlord.cy_tools.gamma_lpdf(double x, double alpha, double lamb) → double

starlord.cy_tools.gamma_ppf(double p, double alpha, double lamb) → double

starlord.cy_tools.exponential_lpdf(double x, double rate) → double

starlord.cy_tools.exponential_ppf(double p, double rate) → double

starlord.cy_tools.exponential_cdf(double x, double rate) → double

starlord.cy_tools.trunc_power_lpdf(double x, double k, double a, double b) → double

starlord.cy_tools.trunc_power_ppf(double p, double k, double a, double b) → double

starlord.cy_tools.trunc_normal_lpdf(double x, double mean, double sigma, double a, double b) → double

starlord.cy_tools.trunc_normal_ppf(double p, double mean, double sigma, double a, double b) → double

starlord.cy_tools.trunc_exponential_lpdf(double x, double rate, double a, double b) → double

starlord.cy_tools.trunc_exponential_ppf(double p, double rate, double a, double b) → double

starlord.cy_tools.chabrier_lpdf(double log_mass, double log_m_switch, double mean, double sigma, double rate) → double

Initial mass function priors of the form from Chabrier (2002) table 2, as a log PDFs.

That paper lists the following constant values for different groups of stars:

Case	log_m_switch	mean	sigma	power	rate
Disk and Young Clusters	0.0	-1.10237	0.69	1.3	5.295945
Globular Clusters	-0.04575749	-0.48148	0.34	1.3	5.295945
Spheroid	-0.15490195	-0.65757	0.33	1.3	5.295945

In order to turn this into a prior on log_mass, the power-law distribution had to be transformed into an exponential distribution, hence the use of “rate” rather than “power”. The conversion is rate = (1+power)*ln(10).

starlord.cy_tools.chabrier_ppf(double p, double log_m_switch, double mean, double sigma, double rate) → double

Initial mass function priors of the form from Chabrier (2002) table 2, as PPFs.

See starlord.cy_tools.chabrier_lpdf() for more information.