Inheritance diagram for lsst.scarlet.lite.models.parametric.EllipseFrame:
Public Member Functions | |
| __init__ (self, float y0, float x0, float major, float minor, float theta, Box bbox, float r_min=1e-20) | |
| float | grad_x0 (self, np.ndarray input_grad, bool use_r2) |
| float | grad_y0 (self, np.ndarray input_grad, bool use_r2) |
| float | grad_major (self, np.ndarray input_grad, bool use_r2) |
| float | grad_minor (self, np.ndarray input_grad, bool use_r2) |
| float | grad_theta (self, np.ndarray input_grad, bool use_r2) |
| float | x0 (self) |
| float | y0 (self) |
| float | major (self) |
| float | minor (self) |
| float | theta (self) |
| Box | bbox (self) |
| np.ndarray | r_grid (self) |
| np.ndarray | r2_grid (self) |
Protected Attributes | |
| _xMajor | |
| _xMinor | |
| _xa | |
| _yb | |
| _y0 | |
| _x0 | |
| _theta | |
| _major | |
| _minor | |
| _sin | |
| _cos | |
| _bbox | |
| _rMin | |
| _radius2 | |
| _radius | |
Detailed Description
Frame to scale the radius based on the parameters of an ellipse
This frame is used to calculate the coordinates of the
radius and radius**2 from a given center location,
based on the semi-major axis, semi-minor axis, and rotation angle.
It is also used to calculate the gradient wrt either the
radius**2 or radius for all of the model parameters.
Parameters
----------
y0: float
The y-center of the ellipse.
x0: float
The x-center of the ellipse.
major: float
The length of the semi-major axis.
minor: float
The length of the semi-minor axis.
theta: float
The counter-clockwise rotation angle
from the semi-major axis.
bbox: Box
The bounding box that contains the entire frame.
r_min: float
The minimum value of the radius.
This is used to prevent divergences that occur
when calculating the gradient at radius == 0.
Definition at line 104 of file parametric.py.
Constructor & Destructor Documentation
◆ __init__()
| lsst.scarlet.lite.models.parametric.EllipseFrame.__init__ | ( | self, | |
| float | bbox, | ||
| float | x0, | ||
| float | major, | ||
| float | minor, | ||
| float | theta, | ||
| Box | bbox, | ||
| float | r_min = 1e-20 ) |
Parameters
----------
bbox: Box
The bounding box that contains this frame.
Reimplemented from lsst.scarlet.lite.models.parametric.CartesianFrame.
Definition at line 134 of file parametric.py.
143 ):
144 super().__init__(bbox)
145 # Set some useful parameters for derivations
146 sin = np.sin(theta)
147 cos = np.cos(theta)
148
149 # Rotate into the frame with xMajor as the x-axis
150 # and xMinor as the y-axis
151 self._xMajor = (self._x - x0) * cos + (self._y - y0) * sin
152 self._xMinor = -(self._x - x0) * sin + (self._y - y0) * cos
153 # The scaled major and minor axes
154 self._xa = self._xMajor / major
155 self._yb = self._xMinor / minor
156
157 # Store parameters useful for gradient calculation
158 self._y0, self._x0 = y0, x0
159 self._theta = theta
160 self._major = major
161 self._minor = minor
162 self._sin, self._cos = sin, cos
163 self._bbox = bbox
164 self._rMin = r_min
165 # Store the scaled radius**2 and radius
166 self._radius2 = self._xa**2 + self._yb**2
167 self._radius: np.ndarray | None = None
168
Member Function Documentation
◆ bbox()
| Box lsst.scarlet.lite.models.parametric.EllipseFrame.bbox | ( | self | ) |
The bounding box to hold the model
Reimplemented from lsst.scarlet.lite.models.parametric.CartesianFrame.
Definition at line 318 of file parametric.py.
319 """The bounding box to hold the model"""
320 return self._bbox
321
◆ grad_major()
| float lsst.scarlet.lite.models.parametric.EllipseFrame.grad_major | ( | self, | |
| np.ndarray | input_grad, | ||
| bool | use_r2 ) |
The gradient of either the radius or radius**2 wrt.
the semi-major axis
Parameters
----------
input_grad:
Gradient of the likelihood wrt the component model
use_r2:
Whether to calculate the gradient of the radius**2
(``use_r2==True``) or the radius (``use_r2==False``).
Returns
-------
result:
The gradient of the likelihood wrt the semi-major axis.
Definition at line 217 of file parametric.py.
217 def grad_major(self, input_grad: np.ndarray, use_r2: bool) -> float:
218 """The gradient of either the radius or radius**2 wrt.
219 the semi-major axis
220
221 Parameters
222 ----------
223 input_grad:
224 Gradient of the likelihood wrt the component model
225 use_r2:
226 Whether to calculate the gradient of the radius**2
227 (``use_r2==True``) or the radius (``use_r2==False``).
228
229 Returns
230 -------
231 result:
232 The gradient of the likelihood wrt the semi-major axis.
233 """
234 grad = -2 / self._major * self._xa**2
235 if use_r2:
236 grad *= 2
237 else:
238 r = self.r_grid
239 grad *= 1 / r
240 return np.sum(grad * input_grad) # type: ignore
241
◆ grad_minor()
| float lsst.scarlet.lite.models.parametric.EllipseFrame.grad_minor | ( | self, | |
| np.ndarray | input_grad, | ||
| bool | use_r2 ) |
The gradient of either the radius or radius**2 wrt.
the semi-minor axis
Parameters
----------
input_grad:
Gradient of the likelihood wrt the component model
use_r2:
Whether to calculate the gradient of the radius**2
(``useR2==True``) or the radius (``useR2==False``).
Returns
-------
result:
The gradient of the likelihood wrt the semi-minor axis.
Definition at line 242 of file parametric.py.
242 def grad_minor(self, input_grad: np.ndarray, use_r2: bool) -> float:
243 """The gradient of either the radius or radius**2 wrt.
244 the semi-minor axis
245
246 Parameters
247 ----------
248 input_grad:
249 Gradient of the likelihood wrt the component model
250 use_r2:
251 Whether to calculate the gradient of the radius**2
252 (``useR2==True``) or the radius (``useR2==False``).
253
254 Returns
255 -------
256 result:
257 The gradient of the likelihood wrt the semi-minor axis.
258 """
259 grad = -2 / self._minor * self._yb**2
260 if use_r2:
261 grad *= 2
262 else:
263 r = self.r_grid
264 grad *= 1 / r
265 return np.sum(grad * input_grad) # type: ignore
266
◆ grad_theta()
| float lsst.scarlet.lite.models.parametric.EllipseFrame.grad_theta | ( | self, | |
| np.ndarray | input_grad, | ||
| bool | use_r2 ) |
The gradient of either the radius or radius**2 wrt.
the rotation angle
Parameters
----------
input_grad:
Gradient of the likelihood wrt the component model
use_r2:
Whether to calculate the gradient of the radius**2
(``useR2==True``) or the radius (``useR2==False``).
Returns
-------
result:
The gradient of the likelihood wrt the rotation angle.
Definition at line 267 of file parametric.py.
267 def grad_theta(self, input_grad: np.ndarray, use_r2: bool) -> float:
268 """The gradient of either the radius or radius**2 wrt.
269 the rotation angle
270
271 Parameters
272 ----------
273 input_grad:
274 Gradient of the likelihood wrt the component model
275 use_r2:
276 Whether to calculate the gradient of the radius**2
277 (``useR2==True``) or the radius (``useR2==False``).
278
279 Returns
280 -------
281 result:
282 The gradient of the likelihood wrt the rotation angle.
283 """
284 grad = self._xa * self._xMinor / self._major - self._yb * self._xMajor / self._minor
285 if use_r2:
286 grad *= 2
287 else:
288 r = self.r_grid
289 grad *= 1 / r
290 return np.sum(grad * input_grad) # type: ignore
291
◆ grad_x0()
| float lsst.scarlet.lite.models.parametric.EllipseFrame.grad_x0 | ( | self, | |
| np.ndarray | input_grad, | ||
| bool | use_r2 ) |
The gradient of either the radius or radius**2 wrt. the x-center
Parameters
----------
input_grad:
Gradient of the likelihood wrt the component model
use_r2:
Whether to calculate the gradient of the radius**2
(``useR2==True``) or the radius (``useR2==False``).
Returns
-------
result:
The gradient of the likelihood wrt x0.
Definition at line 169 of file parametric.py.
169 def grad_x0(self, input_grad: np.ndarray, use_r2: bool) -> float:
170 """The gradient of either the radius or radius**2 wrt. the x-center
171
172 Parameters
173 ----------
174 input_grad:
175 Gradient of the likelihood wrt the component model
176 use_r2:
177 Whether to calculate the gradient of the radius**2
178 (``useR2==True``) or the radius (``useR2==False``).
179
180 Returns
181 -------
182 result:
183 The gradient of the likelihood wrt x0.
184 """
185 grad = -self._xa * self._cos / self._major + self._yb * self._sin / self._minor
186 if use_r2:
187 grad *= 2
188 else:
189 r = self.r_grid
190 grad *= 1 / r
191 return np.sum(grad * input_grad) # type: ignore
192
◆ grad_y0()
| float lsst.scarlet.lite.models.parametric.EllipseFrame.grad_y0 | ( | self, | |
| np.ndarray | input_grad, | ||
| bool | use_r2 ) |
The gradient of either the radius or radius**2 wrt. the y-center
Parameters
----------
input_grad:
Gradient of the likelihood wrt the component model
use_r2:
Whether to calculate the gradient of the radius**2
(``useR2==True``) or the radius (``useR2==False``).
Returns
-------
result:
The gradient of the likelihood wrt y0.
Definition at line 193 of file parametric.py.
193 def grad_y0(self, input_grad: np.ndarray, use_r2: bool) -> float:
194 """The gradient of either the radius or radius**2 wrt. the y-center
195
196 Parameters
197 ----------
198 input_grad:
199 Gradient of the likelihood wrt the component model
200 use_r2:
201 Whether to calculate the gradient of the radius**2
202 (``useR2==True``) or the radius (``useR2==False``).
203
204 Returns
205 -------
206 result:
207 The gradient of the likelihood wrt y0.
208 """
209 grad = -(self._xa * self._sin / self._major + self._yb * self._cos / self._minor)
210 if use_r2:
211 grad *= 2
212 else:
213 r = self.r_grid
214 grad *= 1 / r
215 return np.sum(grad * input_grad) # type: ignore
216
◆ major()
| float lsst.scarlet.lite.models.parametric.EllipseFrame.major | ( | self | ) |
The semi-major axis
Definition at line 303 of file parametric.py.
303 def major(self) -> float:
304 """The semi-major axis"""
305 return self._major
306
◆ minor()
| float lsst.scarlet.lite.models.parametric.EllipseFrame.minor | ( | self | ) |
The semi-minor axis
Definition at line 308 of file parametric.py.
308 def minor(self) -> float:
309 """The semi-minor axis"""
310 return self._minor
311
◆ r2_grid()
| np.ndarray lsst.scarlet.lite.models.parametric.EllipseFrame.r2_grid | ( | self | ) |
The radius squared located at each pixel
Definition at line 330 of file parametric.py.
330 def r2_grid(self) -> np.ndarray:
331 """The radius squared located at each pixel"""
332 return self._radius2 + self._rMin**2
333
334
◆ r_grid()
| np.ndarray lsst.scarlet.lite.models.parametric.EllipseFrame.r_grid | ( | self | ) |
The radial coordinates of each pixel
Definition at line 323 of file parametric.py.
323 def r_grid(self) -> np.ndarray:
324 """The radial coordinates of each pixel"""
325 if self._radius is None:
326 self._radius = np.sqrt(self.r2_grid)
327 return self._radius
328
◆ theta()
| float lsst.scarlet.lite.models.parametric.EllipseFrame.theta | ( | self | ) |
The rotation angle
Definition at line 313 of file parametric.py.
313 def theta(self) -> float:
314 """The rotation angle"""
315 return self._theta
316
◆ x0()
| float lsst.scarlet.lite.models.parametric.EllipseFrame.x0 | ( | self | ) |
The x-center
Definition at line 293 of file parametric.py.
293 def x0(self) -> float:
294 """The x-center"""
295 return self._x0
296
◆ y0()
| float lsst.scarlet.lite.models.parametric.EllipseFrame.y0 | ( | self | ) |
The y-center
Definition at line 298 of file parametric.py.
298 def y0(self) -> float:
299 """The y-center"""
300 return self._y0
301
Member Data Documentation
◆ _bbox
|
protected |
Definition at line 163 of file parametric.py.
◆ _cos
|
protected |
Definition at line 162 of file parametric.py.
◆ _major
|
protected |
Definition at line 160 of file parametric.py.
◆ _minor
|
protected |
Definition at line 161 of file parametric.py.
◆ _radius
|
protected |
Definition at line 326 of file parametric.py.
◆ _radius2
|
protected |
Definition at line 166 of file parametric.py.
◆ _rMin
|
protected |
Definition at line 164 of file parametric.py.
◆ _sin
|
protected |
Definition at line 162 of file parametric.py.
◆ _theta
|
protected |
Definition at line 159 of file parametric.py.
◆ _x0
|
protected |
Definition at line 158 of file parametric.py.
◆ _xa
|
protected |
Definition at line 154 of file parametric.py.
◆ _xMajor
|
protected |
Definition at line 151 of file parametric.py.
◆ _xMinor
|
protected |
Definition at line 152 of file parametric.py.
◆ _y0
|
protected |
Definition at line 158 of file parametric.py.
◆ _yb
|
protected |
Definition at line 155 of file parametric.py.
The documentation for this class was generated from the following file:
- /j/snowflake/release/lsstsw/stack/lsst-scipipe-8.0.0/Linux64/scarlet_lite/gae0086650b+585e252eca/python/lsst/scarlet/lite/models/parametric.py
