fidanka.fiducial.methods package

Submodules

fidanka.fiducial.methods.mc module

fidanka.fiducial.methods.mc.log_likelihood(theta: ndarray[Any, dtype[float64]], binned_mag: ndarray[Any, dtype[[numpy.ndarray[Any, numpy.dtype[numpy.float64]], numpy.ndarray[Any, numpy.dtype[numpy.float64]]]]], binned_color: ndarray[Any, dtype[[numpy.ndarray[Any, numpy.dtype[numpy.float64]], numpy.ndarray[Any, numpy.dtype[numpy.float64]]]]], binned_color_err: ndarray[Any, dtype[[numpy.ndarray[Any, numpy.dtype[numpy.float64]], numpy.ndarray[Any, numpy.dtype[numpy.float64]]]]]) → float

Calculate the logrithmic likelihood for given parameters

Parameters:

  • theta (FARRAY_1D) – Parameters corresponding to the color (first half) and magnitude ( second half) of each bin

  • binned_mag (FARRAY_2D_2C,) – binned magnitude of all input data

  • binned_color – binned color of all input data

  • binned_color_err (FARRAY_2D_2C,) – binned color of all input data

Returns:

log_like – logrithmic likelihood of given set of parameters

Return type:

float

fidanka.fiducial.methods.mc.log_prior(theta: ndarray[Any, dtype[float64]], binsLeft: ndarray[Any, dtype[float64]], binsRight: ndarray[Any, dtype[float64]], colorLeft: ndarray[Any, dtype[float64]], colorRight: ndarray[Any, dtype[float64]]) → float

provide an uninformative prior for mcmc

Parameters:

  • theta (FARRAY_1D) – Parameters corresponding to the color (first half) and magnitude ( second half) of each bin

  • binsLeft (ndarray[float64]) – left edges of bins in magnitude space

  • binsRight (ndarray[float64]) – right edges of bins in magnitude space

  • colorLeft (ndarray[float64]) – left edges of bins in color space

  • colorRight (ndarray[float64]) – right edges of bins in color space

Returns:

prior – a uniform prior within the boundary

Return type:

float

fidanka.fiducial.methods.mc.log_probability(theta: ndarray[Any, dtype[float64]], binned_color: ndarray[Any, dtype[[numpy.ndarray[Any, numpy.dtype[numpy.float64]], numpy.ndarray[Any, numpy.dtype[numpy.float64]]]]], binned_mag: ndarray[Any, dtype[[numpy.ndarray[Any, numpy.dtype[numpy.float64]], numpy.ndarray[Any, numpy.dtype[numpy.float64]]]]], binned_color_err: ndarray[Any, dtype[[numpy.ndarray[Any, numpy.dtype[numpy.float64]], numpy.ndarray[Any, numpy.dtype[numpy.float64]]]]], binsLeft: ndarray[Any, dtype[float64]], binsRight: ndarray[Any, dtype[float64]], colorLeft: ndarray[Any, dtype[float64]], colorRight: ndarray[Any, dtype[float64]]) → float

Calculate the logrithmic probabilty for the parameter. Note, compare to the verticalized cmd method, this method also take into the consideration of the relation between neighbor bins. The main drawbacks of mcmc is that it is very ineffcient in recovering those parameters when they are highly correlated and the curse of dimensionality also suggests that using mcmc in this case can be very computationally expensive.

Parameters:

  • theta (FARRAY_1D,) – Parameters corresponding to the color (first half) and magnitude ( second half) of each bin

  • binned_mag (FARRAY_2D_2C,) – binned magnitude of all input data

  • binned_color – binned color of all input data

  • binned_color_err (FARRAY_2D_2C,) – binned color of all input data

  • binsLeft (ndarray[float64]) – left edges of bins in magnitude space

  • binsRight (ndarray[float64]) – right edges of bins in magnitude space

  • colorLeft (ndarray[float64]) – left edges of bins in color space

  • colorRight (ndarray[float64]) – right edges of bins in color space

Returns:

logrithmic probability of given sets of parameters

Return type:

log_prop

fidanka.fiducial.methods.mc.plm(color, mag, error1, error2, piecewise_linear, binsLeft, binsRight, i, allowMax)

Module contents