| 1 | #!/usr/bin/env python |
|---|
| 2 | """ |
|---|
| 3 | Application to explore the difference between sasview 3.x orientation |
|---|
| 4 | dispersity and possible replacement algorithms. |
|---|
| 5 | """ |
|---|
| 6 | from __future__ import division, print_function |
|---|
| 7 | |
|---|
| 8 | import sys, os |
|---|
| 9 | sys.path.insert(0, os.path.dirname(os.path.dirname(os.path.realpath(__file__)))) |
|---|
| 10 | |
|---|
| 11 | import mpl_toolkits.mplot3d # Adds projection='3d' option to subplot |
|---|
| 12 | import matplotlib.pyplot as plt |
|---|
| 13 | from matplotlib.widgets import Slider, CheckButtons |
|---|
| 14 | from matplotlib import cm |
|---|
| 15 | import numpy as np |
|---|
| 16 | from numpy import pi, cos, sin, sqrt, exp, degrees, radians |
|---|
| 17 | |
|---|
| 18 | SCALED_PHI = True |
|---|
| 19 | |
|---|
| 20 | def draw_beam(ax, view=(0, 0)): |
|---|
| 21 | """ |
|---|
| 22 | Draw the beam going from source at (0, 0, 1) to detector at (0, 0, -1) |
|---|
| 23 | """ |
|---|
| 24 | #ax.plot([0,0],[0,0],[1,-1]) |
|---|
| 25 | #ax.scatter([0]*100,[0]*100,np.linspace(1, -1, 100), alpha=0.8) |
|---|
| 26 | |
|---|
| 27 | steps = 25 |
|---|
| 28 | u = np.linspace(0, 2 * np.pi, steps) |
|---|
| 29 | v = np.linspace(-1, 1, steps) |
|---|
| 30 | |
|---|
| 31 | r = 0.02 |
|---|
| 32 | x = r*np.outer(np.cos(u), np.ones_like(v)) |
|---|
| 33 | y = r*np.outer(np.sin(u), np.ones_like(v)) |
|---|
| 34 | z = 1.3*np.outer(np.ones_like(u), v) |
|---|
| 35 | |
|---|
| 36 | theta, phi = view |
|---|
| 37 | shape = x.shape |
|---|
| 38 | points = np.matrix([x.flatten(), y.flatten(), z.flatten()]) |
|---|
| 39 | points = Rz(phi)*Ry(theta)*points |
|---|
| 40 | x, y, z = [v.reshape(shape) for v in points] |
|---|
| 41 | |
|---|
| 42 | ax.plot_surface(x, y, z, rstride=4, cstride=4, color='y', alpha=0.5) |
|---|
| 43 | |
|---|
| 44 | def draw_jitter(ax, view, jitter, dist='gaussian', size=(0.1, 0.4, 1.0)): |
|---|
| 45 | """ |
|---|
| 46 | Represent jitter as a set of shapes at different orientations. |
|---|
| 47 | """ |
|---|
| 48 | # set max diagonal to 0.95 |
|---|
| 49 | scale = 0.95/sqrt(sum(v**2 for v in size)) |
|---|
| 50 | size = tuple(scale*v for v in size) |
|---|
| 51 | draw_shape = draw_parallelepiped |
|---|
| 52 | #draw_shape = draw_ellipsoid |
|---|
| 53 | |
|---|
| 54 | #np.random.seed(10) |
|---|
| 55 | #cloud = np.random.randn(10,3) |
|---|
| 56 | cloud = [ |
|---|
| 57 | [-1, -1, -1], |
|---|
| 58 | [-1, -1, 0], |
|---|
| 59 | [-1, -1, 1], |
|---|
| 60 | [-1, 0, -1], |
|---|
| 61 | [-1, 0, 0], |
|---|
| 62 | [-1, 0, 1], |
|---|
| 63 | [-1, 1, -1], |
|---|
| 64 | [-1, 1, 0], |
|---|
| 65 | [-1, 1, 1], |
|---|
| 66 | [ 0, -1, -1], |
|---|
| 67 | [ 0, -1, 0], |
|---|
| 68 | [ 0, -1, 1], |
|---|
| 69 | [ 0, 0, -1], |
|---|
| 70 | [ 0, 0, 0], |
|---|
| 71 | [ 0, 0, 1], |
|---|
| 72 | [ 0, 1, -1], |
|---|
| 73 | [ 0, 1, 0], |
|---|
| 74 | [ 0, 1, 1], |
|---|
| 75 | [ 1, -1, -1], |
|---|
| 76 | [ 1, -1, 0], |
|---|
| 77 | [ 1, -1, 1], |
|---|
| 78 | [ 1, 0, -1], |
|---|
| 79 | [ 1, 0, 0], |
|---|
| 80 | [ 1, 0, 1], |
|---|
| 81 | [ 1, 1, -1], |
|---|
| 82 | [ 1, 1, 0], |
|---|
| 83 | [ 1, 1, 1], |
|---|
| 84 | ] |
|---|
| 85 | dtheta, dphi, dpsi = jitter |
|---|
| 86 | if dtheta == 0: |
|---|
| 87 | cloud = [v for v in cloud if v[0] == 0] |
|---|
| 88 | if dphi == 0: |
|---|
| 89 | cloud = [v for v in cloud if v[1] == 0] |
|---|
| 90 | if dpsi == 0: |
|---|
| 91 | cloud = [v for v in cloud if v[2] == 0] |
|---|
| 92 | draw_shape(ax, size, view, [0, 0, 0], steps=100, alpha=0.8) |
|---|
| 93 | scale = 1/sqrt(3) if dist == 'rectangle' else 1 |
|---|
| 94 | for point in cloud: |
|---|
| 95 | delta = [scale*dtheta*point[0], scale*dphi*point[1], scale*dpsi*point[2]] |
|---|
| 96 | draw_shape(ax, size, view, delta, alpha=0.8) |
|---|
| 97 | for v in 'xyz': |
|---|
| 98 | a, b, c = size |
|---|
| 99 | lim = np.sqrt(a**2+b**2+c**2) |
|---|
| 100 | getattr(ax, 'set_'+v+'lim')([-lim, lim]) |
|---|
| 101 | getattr(ax, v+'axis').label.set_text(v) |
|---|
| 102 | |
|---|
| 103 | def draw_ellipsoid(ax, size, view, jitter, steps=25, alpha=1): |
|---|
| 104 | """Draw an ellipsoid.""" |
|---|
| 105 | a,b,c = size |
|---|
| 106 | u = np.linspace(0, 2 * np.pi, steps) |
|---|
| 107 | v = np.linspace(0, np.pi, steps) |
|---|
| 108 | x = a*np.outer(np.cos(u), np.sin(v)) |
|---|
| 109 | y = b*np.outer(np.sin(u), np.sin(v)) |
|---|
| 110 | z = c*np.outer(np.ones_like(u), np.cos(v)) |
|---|
| 111 | x, y, z = transform_xyz(view, jitter, x, y, z) |
|---|
| 112 | |
|---|
| 113 | ax.plot_surface(x, y, z, rstride=4, cstride=4, color='w', alpha=alpha) |
|---|
| 114 | |
|---|
| 115 | draw_labels(ax, view, jitter, [ |
|---|
| 116 | ('c+', [ 0, 0, c], [ 1, 0, 0]), |
|---|
| 117 | ('c-', [ 0, 0,-c], [ 0, 0,-1]), |
|---|
| 118 | ('a+', [ a, 0, 0], [ 0, 0, 1]), |
|---|
| 119 | ('a-', [-a, 0, 0], [ 0, 0,-1]), |
|---|
| 120 | ('b+', [ 0, b, 0], [-1, 0, 0]), |
|---|
| 121 | ('b-', [ 0,-b, 0], [-1, 0, 0]), |
|---|
| 122 | ]) |
|---|
| 123 | |
|---|
| 124 | def draw_parallelepiped(ax, size, view, jitter, steps=None, alpha=1): |
|---|
| 125 | """Draw a parallelepiped.""" |
|---|
| 126 | a,b,c = size |
|---|
| 127 | x = a*np.array([ 1,-1, 1,-1, 1,-1, 1,-1]) |
|---|
| 128 | y = b*np.array([ 1, 1,-1,-1, 1, 1,-1,-1]) |
|---|
| 129 | z = c*np.array([ 1, 1, 1, 1,-1,-1,-1,-1]) |
|---|
| 130 | tri = np.array([ |
|---|
| 131 | # counter clockwise triangles |
|---|
| 132 | # z: up/down, x: right/left, y: front/back |
|---|
| 133 | [0,1,2], [3,2,1], # top face |
|---|
| 134 | [6,5,4], [5,6,7], # bottom face |
|---|
| 135 | [0,2,6], [6,4,0], # right face |
|---|
| 136 | [1,5,7], [7,3,1], # left face |
|---|
| 137 | [2,3,6], [7,6,3], # front face |
|---|
| 138 | [4,1,0], [5,1,4], # back face |
|---|
| 139 | ]) |
|---|
| 140 | |
|---|
| 141 | x, y, z = transform_xyz(view, jitter, x, y, z) |
|---|
| 142 | ax.plot_trisurf(x, y, triangles=tri, Z=z, color='w', alpha=alpha) |
|---|
| 143 | |
|---|
| 144 | draw_labels(ax, view, jitter, [ |
|---|
| 145 | ('c+', [ 0, 0, c], [ 1, 0, 0]), |
|---|
| 146 | ('c-', [ 0, 0,-c], [ 0, 0,-1]), |
|---|
| 147 | ('a+', [ a, 0, 0], [ 0, 0, 1]), |
|---|
| 148 | ('a-', [-a, 0, 0], [ 0, 0,-1]), |
|---|
| 149 | ('b+', [ 0, b, 0], [-1, 0, 0]), |
|---|
| 150 | ('b-', [ 0,-b, 0], [-1, 0, 0]), |
|---|
| 151 | ]) |
|---|
| 152 | |
|---|
| 153 | def draw_sphere(ax, radius=10., steps=100): |
|---|
| 154 | """Draw a sphere""" |
|---|
| 155 | u = np.linspace(0, 2 * np.pi, steps) |
|---|
| 156 | v = np.linspace(0, np.pi, steps) |
|---|
| 157 | |
|---|
| 158 | x = radius * np.outer(np.cos(u), np.sin(v)) |
|---|
| 159 | y = radius * np.outer(np.sin(u), np.sin(v)) |
|---|
| 160 | z = radius * np.outer(np.ones(np.size(u)), np.cos(v)) |
|---|
| 161 | ax.plot_surface(x, y, z, rstride=4, cstride=4, color='w') |
|---|
| 162 | |
|---|
| 163 | def draw_mesh(ax, view, jitter, radius=1.2, n=11, dist='gaussian'): |
|---|
| 164 | """ |
|---|
| 165 | Draw the dispersion mesh showing the theta-phi orientations at which |
|---|
| 166 | the model will be evaluated. |
|---|
| 167 | """ |
|---|
| 168 | theta, phi, psi = view |
|---|
| 169 | dtheta, dphi, dpsi = jitter |
|---|
| 170 | |
|---|
| 171 | if dist == 'gaussian': |
|---|
| 172 | t = np.linspace(-3, 3, n) |
|---|
| 173 | weights = exp(-0.5*t**2) |
|---|
| 174 | elif dist == 'rectangle': |
|---|
| 175 | # Note: uses sasmodels ridiculous definition of rectangle width |
|---|
| 176 | t = np.linspace(-1, 1, n)*sqrt(3) |
|---|
| 177 | weights = np.ones_like(t) |
|---|
| 178 | else: |
|---|
| 179 | raise ValueError("expected dist to be 'gaussian' or 'rectangle'") |
|---|
| 180 | |
|---|
| 181 | if SCALED_PHI: |
|---|
| 182 | scale_phi = lambda dtheta, dphi: ( |
|---|
| 183 | dphi/abs(cos(radians(dtheta))) if dtheta != 90 |
|---|
| 184 | else 0 if dphi == 0 |
|---|
| 185 | else 4*pi) |
|---|
| 186 | w = np.outer(weights, weights) |
|---|
| 187 | else: |
|---|
| 188 | scale_phi = lambda dtheta, dphe: dphi |
|---|
| 189 | w = np.outer(weights*cos(radians(dtheta*t)), weights) |
|---|
| 190 | |
|---|
| 191 | # mesh in theta, phi formed by rotating z |
|---|
| 192 | z = np.matrix([[0], [0], [radius]]) |
|---|
| 193 | points = np.hstack([Rx(scale_phi(theta_i, phi_j))*Ry(theta_i)*z |
|---|
| 194 | for theta_i in dtheta*t |
|---|
| 195 | for phi_j in dphi*t]) |
|---|
| 196 | # select just the active points (i.e., those with phi < 180 |
|---|
| 197 | active = np.array([abs(scale_phi(theta_i, phi_j)) < 180 |
|---|
| 198 | for theta_i in dtheta*t |
|---|
| 199 | for phi_j in dphi*t]) |
|---|
| 200 | points = points[:, active] |
|---|
| 201 | w = w.flatten()[active] |
|---|
| 202 | |
|---|
| 203 | # rotate relative to beam |
|---|
| 204 | points = orient_relative_to_beam(view, points) |
|---|
| 205 | |
|---|
| 206 | x, y, z = [np.array(v).flatten() for v in points] |
|---|
| 207 | ax.scatter(x, y, z, c=w, marker='o', vmin=0., vmax=1.) |
|---|
| 208 | |
|---|
| 209 | def draw_labels(ax, view, jitter, text): |
|---|
| 210 | """ |
|---|
| 211 | Draw text at a particular location. |
|---|
| 212 | """ |
|---|
| 213 | labels, locations, orientations = zip(*text) |
|---|
| 214 | px, py, pz = zip(*locations) |
|---|
| 215 | dx, dy, dz = zip(*orientations) |
|---|
| 216 | |
|---|
| 217 | px, py, pz = transform_xyz(view, jitter, px, py, pz) |
|---|
| 218 | dx, dy, dz = transform_xyz(view, jitter, dx, dy, dz) |
|---|
| 219 | |
|---|
| 220 | # TODO: zdir for labels is broken, and labels aren't appearing. |
|---|
| 221 | for label, p, zdir in zip(labels, zip(px, py, pz), zip(dx, dy, dz)): |
|---|
| 222 | zdir = np.asarray(zdir).flatten() |
|---|
| 223 | ax.text(p[0], p[1], p[2], label, zdir=zdir) |
|---|
| 224 | |
|---|
| 225 | # Definition of rotation matrices comes from wikipedia: |
|---|
| 226 | # https://en.wikipedia.org/wiki/Rotation_matrix#Basic_rotations |
|---|
| 227 | def Rx(angle): |
|---|
| 228 | """Construct a matrix to rotate points about *x* by *angle* degrees.""" |
|---|
| 229 | a = radians(angle) |
|---|
| 230 | R = [[1, 0, 0], |
|---|
| 231 | [0, +cos(a), -sin(a)], |
|---|
| 232 | [0, +sin(a), +cos(a)]] |
|---|
| 233 | return np.matrix(R) |
|---|
| 234 | |
|---|
| 235 | def Ry(angle): |
|---|
| 236 | """Construct a matrix to rotate points about *y* by *angle* degrees.""" |
|---|
| 237 | a = radians(angle) |
|---|
| 238 | R = [[+cos(a), 0, +sin(a)], |
|---|
| 239 | [0, 1, 0], |
|---|
| 240 | [-sin(a), 0, +cos(a)]] |
|---|
| 241 | return np.matrix(R) |
|---|
| 242 | |
|---|
| 243 | def Rz(angle): |
|---|
| 244 | """Construct a matrix to rotate points about *z* by *angle* degrees.""" |
|---|
| 245 | a = radians(angle) |
|---|
| 246 | R = [[+cos(a), -sin(a), 0], |
|---|
| 247 | [+sin(a), +cos(a), 0], |
|---|
| 248 | [0, 0, 1]] |
|---|
| 249 | return np.matrix(R) |
|---|
| 250 | |
|---|
| 251 | def transform_xyz(view, jitter, x, y, z): |
|---|
| 252 | """ |
|---|
| 253 | Send a set of (x,y,z) points through the jitter and view transforms. |
|---|
| 254 | """ |
|---|
| 255 | x, y, z = [np.asarray(v) for v in (x, y, z)] |
|---|
| 256 | shape = x.shape |
|---|
| 257 | points = np.matrix([x.flatten(),y.flatten(),z.flatten()]) |
|---|
| 258 | points = apply_jitter(jitter, points) |
|---|
| 259 | points = orient_relative_to_beam(view, points) |
|---|
| 260 | x, y, z = [np.array(v).reshape(shape) for v in points] |
|---|
| 261 | return x, y, z |
|---|
| 262 | |
|---|
| 263 | def apply_jitter(jitter, points): |
|---|
| 264 | """ |
|---|
| 265 | Apply the jitter transform to a set of points. |
|---|
| 266 | |
|---|
| 267 | Points are stored in a 3 x n numpy matrix, not a numpy array or tuple. |
|---|
| 268 | """ |
|---|
| 269 | dtheta, dphi, dpsi = jitter |
|---|
| 270 | points = Rx(dphi)*Ry(dtheta)*Rz(dpsi)*points |
|---|
| 271 | return points |
|---|
| 272 | |
|---|
| 273 | def orient_relative_to_beam(view, points): |
|---|
| 274 | """ |
|---|
| 275 | Apply the view transform to a set of points. |
|---|
| 276 | |
|---|
| 277 | Points are stored in a 3 x n numpy matrix, not a numpy array or tuple. |
|---|
| 278 | """ |
|---|
| 279 | theta, phi, psi = view |
|---|
| 280 | points = Rz(phi)*Ry(theta)*Rz(psi)*points |
|---|
| 281 | return points |
|---|
| 282 | |
|---|
| 283 | # translate between number of dimension of dispersity and the number of |
|---|
| 284 | # points along each dimension. |
|---|
| 285 | PD_N_TABLE = { |
|---|
| 286 | (0, 0, 0): (0, 0, 0), # 0 |
|---|
| 287 | (1, 0, 0): (100, 0, 0), # 100 |
|---|
| 288 | (0, 1, 0): (0, 100, 0), |
|---|
| 289 | (0, 0, 1): (0, 0, 100), |
|---|
| 290 | (1, 1, 0): (30, 30, 0), # 900 |
|---|
| 291 | (1, 0, 1): (30, 0, 30), |
|---|
| 292 | (0, 1, 1): (0, 30, 30), |
|---|
| 293 | (1, 1, 1): (15, 15, 15), # 3375 |
|---|
| 294 | } |
|---|
| 295 | |
|---|
| 296 | def clipped_range(data, portion=1.0, mode='central'): |
|---|
| 297 | """ |
|---|
| 298 | Determine range from data. |
|---|
| 299 | |
|---|
| 300 | If *portion* is 1, use full range, otherwise use the center of the range |
|---|
| 301 | or the top of the range, depending on whether *mode* is 'central' or 'top'. |
|---|
| 302 | """ |
|---|
| 303 | if portion == 1.0: |
|---|
| 304 | return data.min(), data.max() |
|---|
| 305 | elif mode == 'central': |
|---|
| 306 | data = np.sort(data.flatten()) |
|---|
| 307 | offset = int(portion*len(data)/2 + 0.5) |
|---|
| 308 | return data[offset], data[-offset] |
|---|
| 309 | elif mode == 'top': |
|---|
| 310 | data = np.sort(data.flatten()) |
|---|
| 311 | offset = int(portion*len(data) + 0.5) |
|---|
| 312 | return data[offset], data[-1] |
|---|
| 313 | |
|---|
| 314 | def draw_scattering(calculator, ax, view, jitter, dist='gaussian'): |
|---|
| 315 | """ |
|---|
| 316 | Plot the scattering for the particular view. |
|---|
| 317 | |
|---|
| 318 | *calculator* is returned from :func:`build_model`. *ax* are the 3D axes |
|---|
| 319 | on which the data will be plotted. *view* and *jitter* are the current |
|---|
| 320 | orientation and orientation dispersity. *dist* is one of the sasmodels |
|---|
| 321 | weight distributions. |
|---|
| 322 | """ |
|---|
| 323 | ## Sasmodels use sqrt(3)*width for the rectangle range; scale to the |
|---|
| 324 | ## proper width for comparison. Commented out since now using the |
|---|
| 325 | ## sasmodels definition of width for rectangle. |
|---|
| 326 | #scale = 1/sqrt(3) if dist == 'rectangle' else 1 |
|---|
| 327 | scale = 1 |
|---|
| 328 | |
|---|
| 329 | # add the orientation parameters to the model parameters |
|---|
| 330 | theta, phi, psi = view |
|---|
| 331 | theta_pd, phi_pd, psi_pd = [scale*v for v in jitter] |
|---|
| 332 | theta_pd_n, phi_pd_n, psi_pd_n = PD_N_TABLE[(theta_pd>0, phi_pd>0, psi_pd>0)] |
|---|
| 333 | ## increase pd_n for testing jitter integration rather than simple viz |
|---|
| 334 | #theta_pd_n, phi_pd_n, psi_pd_n = [5*v for v in (theta_pd_n, phi_pd_n, psi_pd_n)] |
|---|
| 335 | |
|---|
| 336 | pars = dict( |
|---|
| 337 | theta=theta, theta_pd=theta_pd, theta_pd_type=dist, theta_pd_n=theta_pd_n, |
|---|
| 338 | phi=phi, phi_pd=phi_pd, phi_pd_type=dist, phi_pd_n=phi_pd_n, |
|---|
| 339 | psi=psi, psi_pd=psi_pd, psi_pd_type=dist, psi_pd_n=psi_pd_n, |
|---|
| 340 | ) |
|---|
| 341 | pars.update(calculator.pars) |
|---|
| 342 | |
|---|
| 343 | # compute the pattern |
|---|
| 344 | qx, qy = calculator._data.x_bins, calculator._data.y_bins |
|---|
| 345 | Iqxy = calculator(**pars).reshape(len(qx), len(qy)) |
|---|
| 346 | |
|---|
| 347 | # scale it and draw it |
|---|
| 348 | Iqxy = np.log(Iqxy) |
|---|
| 349 | if calculator.limits: |
|---|
| 350 | # use limits from orientation (0,0,0) |
|---|
| 351 | vmin, vmax = calculator.limits |
|---|
| 352 | else: |
|---|
| 353 | vmin, vmax = clipped_range(Iqxy, portion=0.95, mode='top') |
|---|
| 354 | #print("range",(vmin,vmax)) |
|---|
| 355 | #qx, qy = np.meshgrid(qx, qy) |
|---|
| 356 | if 0: |
|---|
| 357 | level = np.asarray(255*(Iqxy - vmin)/(vmax - vmin), 'i') |
|---|
| 358 | level[level<0] = 0 |
|---|
| 359 | colors = plt.get_cmap()(level) |
|---|
| 360 | ax.plot_surface(qx, qy, -1.1, rstride=1, cstride=1, facecolors=colors) |
|---|
| 361 | elif 1: |
|---|
| 362 | ax.contourf(qx/qx.max(), qy/qy.max(), Iqxy, zdir='z', offset=-1.1, |
|---|
| 363 | levels=np.linspace(vmin, vmax, 24)) |
|---|
| 364 | else: |
|---|
| 365 | ax.pcolormesh(qx, qy, Iqxy) |
|---|
| 366 | |
|---|
| 367 | def build_model(model_name, n=150, qmax=0.5, **pars): |
|---|
| 368 | """ |
|---|
| 369 | Build a calculator for the given shape. |
|---|
| 370 | |
|---|
| 371 | *model_name* is any sasmodels model. *n* and *qmax* define an n x n mesh |
|---|
| 372 | on which to evaluate the model. The remaining parameters are stored in |
|---|
| 373 | the returned calculator as *calculator.pars*. They are used by |
|---|
| 374 | :func:`draw_scattering` to set the non-orientation parameters in the |
|---|
| 375 | calculation. |
|---|
| 376 | |
|---|
| 377 | Returns a *calculator* function which takes a dictionary or parameters and |
|---|
| 378 | produces Iqxy. The Iqxy value needs to be reshaped to an n x n matrix |
|---|
| 379 | for plotting. See the :class:`sasmodels.direct_model.DirectModel` class |
|---|
| 380 | for details. |
|---|
| 381 | """ |
|---|
| 382 | from sasmodels.core import load_model_info, build_model |
|---|
| 383 | from sasmodels.data import empty_data2D |
|---|
| 384 | from sasmodels.direct_model import DirectModel |
|---|
| 385 | |
|---|
| 386 | model_info = load_model_info(model_name) |
|---|
| 387 | model = build_model(model_info) #, dtype='double!') |
|---|
| 388 | q = np.linspace(-qmax, qmax, n) |
|---|
| 389 | data = empty_data2D(q, q) |
|---|
| 390 | calculator = DirectModel(data, model) |
|---|
| 391 | |
|---|
| 392 | # stuff the values for non-orientation parameters into the calculator |
|---|
| 393 | calculator.pars = pars.copy() |
|---|
| 394 | calculator.pars.setdefault('backgound', 1e-3) |
|---|
| 395 | |
|---|
| 396 | # fix the data limits so that we can see if the pattern fades |
|---|
| 397 | # under rotation or angular dispersion |
|---|
| 398 | Iqxy = calculator(theta=0, phi=0, psi=0, **calculator.pars) |
|---|
| 399 | Iqxy = np.log(Iqxy) |
|---|
| 400 | vmin, vmax = clipped_range(Iqxy, 0.95, mode='top') |
|---|
| 401 | calculator.limits = vmin, vmax+1 |
|---|
| 402 | |
|---|
| 403 | return calculator |
|---|
| 404 | |
|---|
| 405 | def select_calculator(model_name, n=150, size=(10,40,100)): |
|---|
| 406 | """ |
|---|
| 407 | Create a model calculator for the given shape. |
|---|
| 408 | |
|---|
| 409 | *model_name* is one of sphere, cylinder, ellipsoid, triaxial_ellipsoid, |
|---|
| 410 | parallelepiped or bcc_paracrystal. *n* is the number of points to use |
|---|
| 411 | in the q range. *qmax* is chosen based on model parameters for the |
|---|
| 412 | given model to show something intersting. |
|---|
| 413 | |
|---|
| 414 | Returns *calculator* and tuple *size* (a,b,c) giving minor and major |
|---|
| 415 | equitorial axes and polar axis respectively. See :func:`build_model` |
|---|
| 416 | for details on the returned calculator. |
|---|
| 417 | """ |
|---|
| 418 | a, b, c = size |
|---|
| 419 | if model_name == 'sphere': |
|---|
| 420 | calculator = build_model('sphere', n=n, radius=c) |
|---|
| 421 | a = b = c |
|---|
| 422 | elif model_name == 'bcc_paracrystal': |
|---|
| 423 | calculator = build_model('bcc_paracrystal', n=n, dnn=c, |
|---|
| 424 | d_factor=0.06, radius=40) |
|---|
| 425 | a = b = c |
|---|
| 426 | elif model_name == 'cylinder': |
|---|
| 427 | calculator = build_model('cylinder', n=n, qmax=0.3, radius=b, length=c) |
|---|
| 428 | a = b |
|---|
| 429 | elif model_name == 'ellipsoid': |
|---|
| 430 | calculator = build_model('ellipsoid', n=n, qmax=1.0, |
|---|
| 431 | radius_polar=c, radius_equatorial=b) |
|---|
| 432 | a = b |
|---|
| 433 | elif model_name == 'triaxial_ellipsoid': |
|---|
| 434 | calculator = build_model('triaxial_ellipsoid', n=n, qmax=0.5, |
|---|
| 435 | radius_equat_minor=a, |
|---|
| 436 | radius_equat_major=b, |
|---|
| 437 | radius_polar=c) |
|---|
| 438 | elif model_name == 'parallelepiped': |
|---|
| 439 | calculator = build_model('parallelepiped', n=n, a=a, b=b, c=c) |
|---|
| 440 | else: |
|---|
| 441 | raise ValueError("unknown model %s"%model_name) |
|---|
| 442 | |
|---|
| 443 | return calculator, (a, b, c) |
|---|
| 444 | |
|---|
| 445 | def main(model_name='parallelepiped', size=(10, 40, 100)): |
|---|
| 446 | """ |
|---|
| 447 | Show an interactive orientation and jitter demo. |
|---|
| 448 | |
|---|
| 449 | *model_name* is one of the models available in :func:`select_model`. |
|---|
| 450 | """ |
|---|
| 451 | # set up calculator |
|---|
| 452 | calculator, size = select_calculator(model_name, n=150, size=size) |
|---|
| 453 | |
|---|
| 454 | ## uncomment to set an independent the colour range for every view |
|---|
| 455 | ## If left commented, the colour range is fixed for all views |
|---|
| 456 | calculator.limits = None |
|---|
| 457 | |
|---|
| 458 | ## use gaussian distribution unless testing integration |
|---|
| 459 | #dist = 'rectangle' |
|---|
| 460 | dist = 'gaussian' |
|---|
| 461 | |
|---|
| 462 | ## initial view |
|---|
| 463 | #theta, dtheta = 70., 10. |
|---|
| 464 | #phi, dphi = -45., 3. |
|---|
| 465 | #psi, dpsi = -45., 3. |
|---|
| 466 | theta, phi, psi = 0, 0, 0 |
|---|
| 467 | dtheta, dphi, dpsi = 0, 0, 0 |
|---|
| 468 | |
|---|
| 469 | ## create the plot window |
|---|
| 470 | #plt.hold(True) |
|---|
| 471 | plt.set_cmap('gist_earth') |
|---|
| 472 | plt.clf() |
|---|
| 473 | #gs = gridspec.GridSpec(2,1,height_ratios=[4,1]) |
|---|
| 474 | #ax = plt.subplot(gs[0], projection='3d') |
|---|
| 475 | ax = plt.axes([0.0, 0.2, 1.0, 0.8], projection='3d') |
|---|
| 476 | try: # CRUFT: not all versions of matplotlib accept 'square' 3d projection |
|---|
| 477 | ax.axis('square') |
|---|
| 478 | except Exception: |
|---|
| 479 | pass |
|---|
| 480 | |
|---|
| 481 | axcolor = 'lightgoldenrodyellow' |
|---|
| 482 | |
|---|
| 483 | ## add control widgets to plot |
|---|
| 484 | axtheta = plt.axes([0.1, 0.15, 0.45, 0.04], axisbg=axcolor) |
|---|
| 485 | axphi = plt.axes([0.1, 0.1, 0.45, 0.04], axisbg=axcolor) |
|---|
| 486 | axpsi = plt.axes([0.1, 0.05, 0.45, 0.04], axisbg=axcolor) |
|---|
| 487 | stheta = Slider(axtheta, 'Theta', -90, 90, valinit=theta) |
|---|
| 488 | sphi = Slider(axphi, 'Phi', -180, 180, valinit=phi) |
|---|
| 489 | spsi = Slider(axpsi, 'Psi', -180, 180, valinit=psi) |
|---|
| 490 | |
|---|
| 491 | axdtheta = plt.axes([0.75, 0.15, 0.15, 0.04], axisbg=axcolor) |
|---|
| 492 | axdphi = plt.axes([0.75, 0.1, 0.15, 0.04], axisbg=axcolor) |
|---|
| 493 | axdpsi= plt.axes([0.75, 0.05, 0.15, 0.04], axisbg=axcolor) |
|---|
| 494 | # Note: using ridiculous definition of rectangle distribution, whose width |
|---|
| 495 | # in sasmodels is sqrt(3) times the given width. Divide by sqrt(3) to keep |
|---|
| 496 | # the maximum width to 90. |
|---|
| 497 | dlimit = 30 if dist == 'gaussian' else 90/sqrt(3) |
|---|
| 498 | sdtheta = Slider(axdtheta, 'dTheta', 0, 2*dlimit, valinit=dtheta) |
|---|
| 499 | sdphi = Slider(axdphi, 'dPhi', 0, 2*dlimit, valinit=dphi) |
|---|
| 500 | sdpsi = Slider(axdpsi, 'dPsi', 0, 2*dlimit, valinit=dpsi) |
|---|
| 501 | |
|---|
| 502 | ## callback to draw the new view |
|---|
| 503 | def update(val, axis=None): |
|---|
| 504 | view = stheta.val, sphi.val, spsi.val |
|---|
| 505 | jitter = sdtheta.val, sdphi.val, sdpsi.val |
|---|
| 506 | # set small jitter as 0 if multiple pd dims |
|---|
| 507 | dims = sum(v > 0 for v in jitter) |
|---|
| 508 | limit = [0, 0, 2, 5][dims] |
|---|
| 509 | jitter = [0 if v < limit else v for v in jitter] |
|---|
| 510 | ax.cla() |
|---|
| 511 | draw_beam(ax, (0, 0)) |
|---|
| 512 | draw_jitter(ax, view, jitter, dist=dist, size=size) |
|---|
| 513 | #draw_jitter(ax, view, (0,0,0)) |
|---|
| 514 | draw_mesh(ax, view, jitter, dist=dist, n=30) |
|---|
| 515 | draw_scattering(calculator, ax, view, jitter, dist=dist) |
|---|
| 516 | plt.gcf().canvas.draw() |
|---|
| 517 | |
|---|
| 518 | ## bind control widgets to view updater |
|---|
| 519 | stheta.on_changed(lambda v: update(v,'theta')) |
|---|
| 520 | sphi.on_changed(lambda v: update(v, 'phi')) |
|---|
| 521 | spsi.on_changed(lambda v: update(v, 'psi')) |
|---|
| 522 | sdtheta.on_changed(lambda v: update(v, 'dtheta')) |
|---|
| 523 | sdphi.on_changed(lambda v: update(v, 'dphi')) |
|---|
| 524 | sdpsi.on_changed(lambda v: update(v, 'dpsi')) |
|---|
| 525 | |
|---|
| 526 | ## initialize view |
|---|
| 527 | update(None, 'phi') |
|---|
| 528 | |
|---|
| 529 | ## go interactive |
|---|
| 530 | plt.show() |
|---|
| 531 | |
|---|
| 532 | if __name__ == "__main__": |
|---|
| 533 | model_name = sys.argv[1] if len(sys.argv) > 1 else 'parallelepiped' |
|---|
| 534 | size = tuple(int(v) for v in sys.argv[2].split(',')) if len(sys.argv) > 2 else (10, 40, 100) |
|---|
| 535 | main(model_name, size) |
|---|
