bspline¶
-
class
pydl.pydlutils.bspline.
bspline
(x, nord=4, npoly=1, bkpt=None, bkspread=1.0, **kwargs)[source]¶ Bases:
object
B-spline class.
Functions in the idlutils bspline library are implemented as methods on this class.
- Parameters
x :
numpy.ndarray
The data.
nord :
int
, optionalThe order of the B-spline. Default is 4, which is cubic.
npoly :
int
, optionalPolynomial order to fit over 2nd variable, if supplied. If not supplied the order is 1.
bkpt :
numpy.ndarray
, optionalTo be documented.
bkspread :
float
, optionalTo be documented.
Attributes
breakpoints
To be documented.
nord
To be documented.
npoly
To be documented.
mask
To be documented.
coeff
To be documented.
icoeff
To be documented.
xmin
To be documented.
xmax
To be documented.
funcname
To be documented.
Init creates an object whose attributes are similar to the structure returned by the
create_bsplineset()
function.Methods Summary
action
(x[, x2])Construct banded B-spline matrix, with dimensions [ndata, bandwidth].
bsplvn
(x, ileft)Calculates the value of all possibly nonzero B-splines at
x
of a certain order.fit
(xdata, ydata, invvar[, x2])Calculate a B-spline in the least-squares sense.
intrv
(x)Find the segment between breakpoints which contain each value in the array
x
.maskpoints
(err)Perform simple logic of which breakpoints to mask.
value
(x[, x2, action, lower, upper])Evaluate a B-spline at specified values.
Methods Documentation
-
action
(x, x2=None)[source]¶ Construct banded B-spline matrix, with dimensions [ndata, bandwidth].
- Parameters
x :
numpy.ndarray
Independent variable.
x2 :
numpy.ndarray
, optionalOrthogonal dependent variable for 2d fits.
- Returns
tuple()
A tuple containing the B-spline action matrix; the ‘lower’ parameter, a list of pixel positions, each corresponding to the first occurence of position greater than breakpoint indx; and ‘upper’, Same as lower, but denotes the upper pixel positions.
-
bsplvn
(x, ileft)[source]¶ Calculates the value of all possibly nonzero B-splines at
x
of a certain order.- Parameters
x :
numpy.ndarray
Independent variable.
ileft :
int
Breakpoint segements that contain
x
.- Returns
numpy.ndarray
B-spline values.
-
fit
(xdata, ydata, invvar, x2=None)[source]¶ Calculate a B-spline in the least-squares sense.
Fit is based on two variables:
xdata
which is sorted and spans a large range where breakpoints are requiredydata
which can be described with a low order polynomial.- Parameters
xdata :
numpy.ndarray
Independent variable.
ydata :
numpy.ndarray
Dependent variable.
invvar :
numpy.ndarray
Inverse variance of
ydata
.x2 :
numpy.ndarray
, optionalOrthogonal dependent variable for 2d fits.
- Returns
tuple()
A tuple containing an integer error code, and the evaluation of the b-spline at the input values. An error code of -2 is a failure, -1 indicates dropped breakpoints, 0 is success, and positive integers indicate ill-conditioned breakpoints.
-
intrv
(x)[source]¶ Find the segment between breakpoints which contain each value in the array
x
.The minimum breakpoint is
nbkptord - 1
, and the maximum isnbkpt - nbkptord - 1
.- Parameters
x :
numpy.ndarray
Data values, assumed to be monotonically increasing.
- Returns
numpy.ndarray
Position of array elements with respect to breakpoints.
-
maskpoints
(err)[source]¶ Perform simple logic of which breakpoints to mask.
- Parameters
err :
numpy.ndarray
The list of indexes returned by the cholesky routines.
- Returns
int
An integer indicating the results of the masking. -1 indicates that the error points were successfully masked. -2 indicates failure; the calculation should be aborted.
Notes
The mask attribute is modified, assuming it is possible to create the mask.
-
value
(x, x2=None, action=None, lower=None, upper=None)[source]¶ Evaluate a B-spline at specified values.
- Parameters
x :
numpy.ndarray
Independent variable.
x2 :
numpy.ndarray
, optionalOrthogonal dependent variable for 2d fits.
action :
numpy.ndarray
, optionalAction matrix to use. If not supplied it is calculated.
lower :
numpy.ndarray
, optionalIf the action parameter is supplied, this parameter must also be supplied.
upper :
numpy.ndarray
, optionalIf the action parameter is supplied, this parameter must also be supplied.
- Returns
tuple()
A tuple containing the results of the bspline evaluation and a mask indicating where the evaluation was good.