[3b4243d] | 1 | """ |
---|
| 2 | SAS data representations. |
---|
| 3 | |
---|
| 4 | Plotting functions for data sets: |
---|
| 5 | |
---|
| 6 | :func:`plot_data` plots the data file. |
---|
| 7 | |
---|
| 8 | :func:`plot_theory` plots a calculated result from the model. |
---|
| 9 | |
---|
| 10 | Wrappers for the sasview data loader and data manipulations: |
---|
| 11 | |
---|
| 12 | :func:`load_data` loads a sasview data file. |
---|
| 13 | |
---|
| 14 | :func:`set_beam_stop` masks the beam stop from the data. |
---|
| 15 | |
---|
| 16 | :func:`set_half` selects the right or left half of the data, which can |
---|
| 17 | be useful for shear measurements which have not been properly corrected |
---|
| 18 | for path length and reflections. |
---|
| 19 | |
---|
| 20 | :func:`set_top` cuts the top part off the data. |
---|
| 21 | |
---|
| 22 | |
---|
| 23 | Empty data sets for evaluating models without data: |
---|
| 24 | |
---|
| 25 | :func:`empty_data1D` creates an empty dataset, which is useful for plotting |
---|
| 26 | a theory function before the data is measured. |
---|
| 27 | |
---|
| 28 | :func:`empty_data2D` creates an empty 2D dataset. |
---|
| 29 | |
---|
| 30 | Note that the empty datasets use a minimal representation of the SasView |
---|
| 31 | objects so that models can be run without SasView on the path. You could |
---|
| 32 | also use these for your own data loader. |
---|
| 33 | |
---|
| 34 | """ |
---|
| 35 | import traceback |
---|
| 36 | |
---|
| 37 | import numpy as np |
---|
| 38 | |
---|
| 39 | def load_data(filename): |
---|
| 40 | """ |
---|
| 41 | Load data using a sasview loader. |
---|
| 42 | """ |
---|
| 43 | from sas.dataloader.loader import Loader |
---|
| 44 | loader = Loader() |
---|
| 45 | data = loader.load(filename) |
---|
| 46 | if data is None: |
---|
| 47 | raise IOError("Data %r could not be loaded" % filename) |
---|
| 48 | return data |
---|
| 49 | |
---|
| 50 | |
---|
| 51 | def set_beam_stop(data, radius, outer=None): |
---|
| 52 | """ |
---|
| 53 | Add a beam stop of the given *radius*. If *outer*, make an annulus. |
---|
| 54 | """ |
---|
| 55 | from sas.dataloader.manipulations import Ringcut |
---|
| 56 | if hasattr(data, 'qx_data'): |
---|
| 57 | data.mask = Ringcut(0, radius)(data) |
---|
| 58 | if outer is not None: |
---|
| 59 | data.mask += Ringcut(outer, np.inf)(data) |
---|
| 60 | else: |
---|
| 61 | data.mask = (data.x < radius) |
---|
| 62 | if outer is not None: |
---|
| 63 | data.mask |= (data.x >= outer) |
---|
| 64 | |
---|
| 65 | |
---|
| 66 | def set_half(data, half): |
---|
| 67 | """ |
---|
| 68 | Select half of the data, either "right" or "left". |
---|
| 69 | """ |
---|
| 70 | from sas.dataloader.manipulations import Boxcut |
---|
| 71 | if half == 'right': |
---|
| 72 | data.mask += \ |
---|
| 73 | Boxcut(x_min=-np.inf, x_max=0.0, y_min=-np.inf, y_max=np.inf)(data) |
---|
| 74 | if half == 'left': |
---|
| 75 | data.mask += \ |
---|
| 76 | Boxcut(x_min=0.0, x_max=np.inf, y_min=-np.inf, y_max=np.inf)(data) |
---|
| 77 | |
---|
| 78 | |
---|
| 79 | def set_top(data, cutoff): |
---|
| 80 | """ |
---|
| 81 | Chop the top off the data, above *cutoff*. |
---|
| 82 | """ |
---|
| 83 | from sas.dataloader.manipulations import Boxcut |
---|
| 84 | data.mask += \ |
---|
| 85 | Boxcut(x_min=-np.inf, x_max=np.inf, y_min=-np.inf, y_max=cutoff)(data) |
---|
| 86 | |
---|
| 87 | |
---|
| 88 | class Data1D(object): |
---|
| 89 | def __init__(self, x=None, y=None, dx=None, dy=None): |
---|
| 90 | self.x, self.y, self.dx, self.dy = x, y, dx, dy |
---|
| 91 | self.dxl = None |
---|
| 92 | |
---|
| 93 | def xaxis(self, label, unit): |
---|
| 94 | """ |
---|
| 95 | set the x axis label and unit |
---|
| 96 | """ |
---|
| 97 | self._xaxis = label |
---|
| 98 | self._xunit = unit |
---|
| 99 | |
---|
| 100 | def yaxis(self, label, unit): |
---|
| 101 | """ |
---|
| 102 | set the y axis label and unit |
---|
| 103 | """ |
---|
| 104 | self._yaxis = label |
---|
| 105 | self._yunit = unit |
---|
| 106 | |
---|
| 107 | |
---|
| 108 | |
---|
| 109 | class Data2D(object): |
---|
| 110 | def __init__(self): |
---|
| 111 | self.detector = [] |
---|
| 112 | self.source = Source() |
---|
| 113 | |
---|
| 114 | def xaxis(self, label, unit): |
---|
| 115 | """ |
---|
| 116 | set the x axis label and unit |
---|
| 117 | """ |
---|
| 118 | self._xaxis = label |
---|
| 119 | self._xunit = unit |
---|
| 120 | |
---|
| 121 | def yaxis(self, label, unit): |
---|
| 122 | """ |
---|
| 123 | set the y axis label and unit |
---|
| 124 | """ |
---|
| 125 | self._yaxis = label |
---|
| 126 | self._yunit = unit |
---|
| 127 | |
---|
| 128 | def zaxis(self, label, unit): |
---|
| 129 | """ |
---|
| 130 | set the y axis label and unit |
---|
| 131 | """ |
---|
| 132 | self._zaxis = label |
---|
| 133 | self._zunit = unit |
---|
| 134 | |
---|
| 135 | |
---|
| 136 | class Vector(object): |
---|
| 137 | def __init__(self, x=None, y=None, z=None): |
---|
| 138 | self.x, self.y, self.z = x, y, z |
---|
| 139 | |
---|
| 140 | class Detector(object): |
---|
| 141 | def __init__(self): |
---|
| 142 | self.pixel_size = Vector() |
---|
| 143 | |
---|
| 144 | class Source(object): |
---|
| 145 | pass |
---|
| 146 | |
---|
| 147 | |
---|
| 148 | def empty_data1D(q, resolution=0.05): |
---|
| 149 | """ |
---|
| 150 | Create empty 1D data using the given *q* as the x value. |
---|
| 151 | |
---|
| 152 | *resolution* dq/q defaults to 5%. |
---|
| 153 | """ |
---|
| 154 | |
---|
| 155 | #Iq = 100 * np.ones_like(q) |
---|
| 156 | #dIq = np.sqrt(Iq) |
---|
| 157 | Iq, dIq = None, None |
---|
| 158 | data = Data1D(q, Iq, dx=resolution * q, dy=dIq) |
---|
| 159 | data.filename = "fake data" |
---|
| 160 | data.qmin, data.qmax = q.min(), q.max() |
---|
| 161 | data.mask = np.zeros(len(q), dtype='bool') |
---|
| 162 | return data |
---|
| 163 | |
---|
| 164 | |
---|
| 165 | def empty_data2D(qx, qy=None, resolution=0.05): |
---|
| 166 | """ |
---|
| 167 | Create empty 2D data using the given mesh. |
---|
| 168 | |
---|
| 169 | If *qy* is missing, create a square mesh with *qy=qx*. |
---|
| 170 | |
---|
| 171 | *resolution* dq/q defaults to 5%. |
---|
| 172 | """ |
---|
| 173 | if qy is None: |
---|
| 174 | qy = qx |
---|
| 175 | Qx, Qy = np.meshgrid(qx, qy) |
---|
| 176 | Qx, Qy = Qx.flatten(), Qy.flatten() |
---|
| 177 | Iq = 100 * np.ones_like(Qx) |
---|
| 178 | dIq = np.sqrt(Iq) |
---|
| 179 | mask = np.ones(len(Iq), dtype='bool') |
---|
| 180 | |
---|
| 181 | data = Data2D() |
---|
| 182 | data.filename = "fake data" |
---|
| 183 | data.qx_data = Qx |
---|
| 184 | data.qy_data = Qy |
---|
| 185 | data.data = Iq |
---|
| 186 | data.err_data = dIq |
---|
| 187 | data.mask = mask |
---|
| 188 | data.qmin = 1e-16 |
---|
| 189 | data.qmax = np.inf |
---|
| 190 | |
---|
| 191 | # 5% dQ/Q resolution |
---|
| 192 | if resolution != 0: |
---|
| 193 | # https://www.ncnr.nist.gov/staff/hammouda/distance_learning/chapter_15.pdf |
---|
| 194 | # Should have an additional constant which depends on distances and |
---|
| 195 | # radii of the aperture, pixel dimensions and wavelength spread |
---|
| 196 | # Instead, assume radial dQ/Q is constant, and perpendicular matches |
---|
| 197 | # radial (which instead it should be inverse). |
---|
| 198 | Q = np.sqrt(Qx**2 + Qy**2) |
---|
| 199 | data.dqx_data = resolution * Q |
---|
| 200 | data.dqy_data = resolution * Q |
---|
[ac21c7f] | 201 | else: |
---|
| 202 | data.dqx_data = data.dqy_data = None |
---|
[3b4243d] | 203 | |
---|
| 204 | detector = Detector() |
---|
| 205 | detector.pixel_size.x = 5 # mm |
---|
| 206 | detector.pixel_size.y = 5 # mm |
---|
| 207 | detector.distance = 4 # m |
---|
| 208 | data.detector.append(detector) |
---|
| 209 | data.xbins = qx |
---|
| 210 | data.ybins = qy |
---|
| 211 | data.source.wavelength = 5 # angstroms |
---|
| 212 | data.source.wavelength_unit = "A" |
---|
| 213 | data.Q_unit = "1/A" |
---|
| 214 | data.I_unit = "1/cm" |
---|
| 215 | data.q_data = np.sqrt(Qx ** 2 + Qy ** 2) |
---|
| 216 | data.xaxis("Q_x", "A^{-1}") |
---|
| 217 | data.yaxis("Q_y", "A^{-1}") |
---|
| 218 | data.zaxis("Intensity", r"\text{cm}^{-1}") |
---|
| 219 | return data |
---|
| 220 | |
---|
| 221 | |
---|
| 222 | def plot_data(data, view='log'): |
---|
| 223 | """ |
---|
| 224 | Plot data loaded by the sasview loader. |
---|
| 225 | """ |
---|
| 226 | # Note: kind of weird using the plot result functions to plot just the |
---|
| 227 | # data, but they already handle the masking and graph markup already, so |
---|
| 228 | # do not repeat. |
---|
| 229 | if hasattr(data, 'lam'): |
---|
| 230 | _plot_result_sesans(data, None, None, plot_data=True) |
---|
| 231 | elif hasattr(data, 'qx_data'): |
---|
| 232 | _plot_result2D(data, None, None, view, plot_data=True) |
---|
| 233 | else: |
---|
| 234 | _plot_result1D(data, None, None, view, plot_data=True) |
---|
| 235 | |
---|
| 236 | |
---|
| 237 | def plot_theory(data, theory, resid=None, view='log', plot_data=True): |
---|
| 238 | if hasattr(data, 'lam'): |
---|
| 239 | _plot_result_sesans(data, theory, resid, plot_data=True) |
---|
| 240 | elif hasattr(data, 'qx_data'): |
---|
| 241 | _plot_result2D(data, theory, resid, view, plot_data) |
---|
| 242 | else: |
---|
| 243 | _plot_result1D(data, theory, resid, view, plot_data) |
---|
| 244 | |
---|
| 245 | |
---|
| 246 | def protect(fn): |
---|
| 247 | def wrapper(*args, **kw): |
---|
| 248 | try: |
---|
| 249 | return fn(*args, **kw) |
---|
| 250 | except: |
---|
| 251 | traceback.print_exc() |
---|
| 252 | pass |
---|
| 253 | |
---|
| 254 | return wrapper |
---|
| 255 | |
---|
| 256 | |
---|
| 257 | @protect |
---|
| 258 | def _plot_result1D(data, theory, resid, view, plot_data): |
---|
| 259 | """ |
---|
| 260 | Plot the data and residuals for 1D data. |
---|
| 261 | """ |
---|
| 262 | import matplotlib.pyplot as plt |
---|
| 263 | from numpy.ma import masked_array, masked |
---|
| 264 | |
---|
| 265 | plot_theory = theory is not None |
---|
| 266 | plot_resid = resid is not None |
---|
| 267 | |
---|
| 268 | if data.y is None: |
---|
| 269 | plot_data = False |
---|
| 270 | scale = data.x**4 if view == 'q4' else 1.0 |
---|
| 271 | |
---|
| 272 | if plot_data or plot_theory: |
---|
| 273 | if plot_resid: |
---|
| 274 | plt.subplot(121) |
---|
| 275 | |
---|
| 276 | #print vmin, vmax |
---|
| 277 | positive = False |
---|
| 278 | if plot_data: |
---|
| 279 | mdata = masked_array(data.y, data.mask) |
---|
| 280 | mdata[~np.isfinite(mdata)] = masked |
---|
| 281 | if view is 'log': |
---|
| 282 | mdata[mdata <= 0] = masked |
---|
| 283 | plt.errorbar(data.x, scale*mdata, yerr=data.dy, fmt='.') |
---|
| 284 | positive = positive or (mdata>0).any() |
---|
| 285 | |
---|
| 286 | if plot_theory: |
---|
| 287 | mtheory = masked_array(theory, data.mask) |
---|
| 288 | if view is 'log': |
---|
| 289 | mtheory[mtheory<= 0] = masked |
---|
| 290 | plt.plot(data.x, scale*mtheory, '-', hold=True) |
---|
| 291 | positive = positive or (mtheory>0).any() |
---|
| 292 | |
---|
| 293 | plt.xscale(view) |
---|
| 294 | plt.yscale('linear' if view == 'q4' or not positive else view) |
---|
| 295 | plt.xlabel('Q') |
---|
| 296 | plt.ylabel('I(Q)') |
---|
| 297 | |
---|
| 298 | if plot_resid: |
---|
| 299 | if plot_data or plot_theory: |
---|
| 300 | plt.subplot(122) |
---|
| 301 | |
---|
| 302 | mresid = masked_array(resid, data.mask) |
---|
| 303 | plt.plot(data.x, mresid, '-') |
---|
| 304 | plt.ylabel('residuals') |
---|
| 305 | plt.xlabel('Q') |
---|
| 306 | plt.xscale(view) |
---|
| 307 | |
---|
| 308 | |
---|
| 309 | @protect |
---|
| 310 | def _plot_result_sesans(data, theory, resid, plot_data): |
---|
| 311 | import matplotlib.pyplot as plt |
---|
| 312 | if data.y is None: |
---|
| 313 | plot_data = False |
---|
| 314 | plot_theory = theory is not None |
---|
| 315 | plot_resid = resid is not None |
---|
| 316 | |
---|
| 317 | if plot_data or plot_theory: |
---|
| 318 | if plot_resid: |
---|
| 319 | plt.subplot(121) |
---|
| 320 | if plot_data: |
---|
| 321 | plt.errorbar(data.x, data.y, yerr=data.dy) |
---|
| 322 | if theory is not None: |
---|
| 323 | plt.plot(data.x, theory, '-', hold=True) |
---|
| 324 | plt.xlabel('spin echo length (nm)') |
---|
| 325 | plt.ylabel('polarization (P/P0)') |
---|
| 326 | |
---|
| 327 | if resid is not None: |
---|
| 328 | if plot_data or plot_theory: |
---|
| 329 | plt.subplot(122) |
---|
| 330 | |
---|
| 331 | plt.plot(data.x, resid, 'x') |
---|
| 332 | plt.xlabel('spin echo length (nm)') |
---|
| 333 | plt.ylabel('residuals (P/P0)') |
---|
| 334 | |
---|
| 335 | |
---|
| 336 | @protect |
---|
| 337 | def _plot_result2D(data, theory, resid, view, plot_data): |
---|
| 338 | """ |
---|
| 339 | Plot the data and residuals for 2D data. |
---|
| 340 | """ |
---|
| 341 | import matplotlib.pyplot as plt |
---|
| 342 | if data.data is None: |
---|
| 343 | plot_data = False |
---|
| 344 | plot_theory = theory is not None |
---|
| 345 | plot_resid = resid is not None |
---|
| 346 | |
---|
| 347 | # Put theory and data on a common colormap scale |
---|
| 348 | vmin, vmax = np.inf, -np.inf |
---|
| 349 | if plot_data: |
---|
| 350 | target = data.data[~data.mask] |
---|
| 351 | datamin = target[target>0].min() if view == 'log' else target.min() |
---|
| 352 | datamax = target.max() |
---|
| 353 | vmin = min(vmin, datamin) |
---|
| 354 | vmax = max(vmax, datamax) |
---|
| 355 | if plot_theory: |
---|
| 356 | theorymin = theory[theory>0].min() if view == 'log' else theory.min() |
---|
| 357 | theorymax = theory.max() |
---|
| 358 | vmin = min(vmin, theorymin) |
---|
| 359 | vmax = max(vmax, theorymax) |
---|
| 360 | |
---|
| 361 | if plot_data: |
---|
| 362 | if plot_theory and plot_resid: |
---|
| 363 | plt.subplot(131) |
---|
| 364 | elif plot_theory or plot_resid: |
---|
| 365 | plt.subplot(121) |
---|
| 366 | _plot_2d_signal(data, target, view=view, vmin=vmin, vmax=vmax) |
---|
| 367 | plt.title('data') |
---|
| 368 | plt.colorbar() |
---|
| 369 | |
---|
| 370 | if plot_theory: |
---|
| 371 | if plot_data and plot_resid: |
---|
| 372 | plt.subplot(132) |
---|
| 373 | elif plot_data: |
---|
| 374 | plt.subplot(122) |
---|
| 375 | elif plot_resid: |
---|
| 376 | plt.subplot(121) |
---|
| 377 | _plot_2d_signal(data, theory, view=view, vmin=vmin, vmax=vmax) |
---|
| 378 | plt.title('theory') |
---|
| 379 | plt.colorbar() |
---|
| 380 | |
---|
| 381 | #if plot_data or plot_theory: |
---|
| 382 | # plt.colorbar() |
---|
| 383 | |
---|
| 384 | if plot_resid: |
---|
| 385 | if plot_data and plot_theory: |
---|
| 386 | plt.subplot(133) |
---|
| 387 | elif plot_data or plot_theory: |
---|
| 388 | plt.subplot(122) |
---|
| 389 | _plot_2d_signal(data, resid, view='linear') |
---|
| 390 | plt.colorbar() |
---|
| 391 | plt.title('residuals') |
---|
| 392 | |
---|
| 393 | |
---|
| 394 | @protect |
---|
| 395 | def _plot_2d_signal(data, signal, vmin=None, vmax=None, view='log'): |
---|
| 396 | """ |
---|
| 397 | Plot the target value for the data. This could be the data itself, |
---|
| 398 | the theory calculation, or the residuals. |
---|
| 399 | |
---|
| 400 | *scale* can be 'log' for log scale data, or 'linear'. |
---|
| 401 | """ |
---|
| 402 | import matplotlib.pyplot as plt |
---|
| 403 | from numpy.ma import masked_array |
---|
| 404 | |
---|
| 405 | image = np.zeros_like(data.qx_data) |
---|
| 406 | image[~data.mask] = signal |
---|
| 407 | valid = np.isfinite(image) |
---|
| 408 | if view == 'log': |
---|
| 409 | valid[valid] = (image[valid] > 0) |
---|
| 410 | image[valid] = np.log10(image[valid]) |
---|
| 411 | elif view == 'q4': |
---|
| 412 | image[valid] *= (data.qx_data[valid]**2+data.qy_data[valid]**2)**2 |
---|
| 413 | image[~valid | data.mask] = 0 |
---|
| 414 | #plottable = Iq |
---|
| 415 | plottable = masked_array(image, ~valid | data.mask) |
---|
| 416 | xmin, xmax = min(data.qx_data), max(data.qx_data) |
---|
| 417 | ymin, ymax = min(data.qy_data), max(data.qy_data) |
---|
| 418 | # TODO: fix vmin, vmax so it is shared for theory/resid |
---|
| 419 | vmin = vmax = None |
---|
| 420 | try: |
---|
| 421 | if vmin is None: vmin = image[valid & ~data.mask].min() |
---|
| 422 | if vmax is None: vmax = image[valid & ~data.mask].max() |
---|
| 423 | except: |
---|
| 424 | vmin, vmax = 0, 1 |
---|
| 425 | plt.imshow(plottable.reshape(128, 128), |
---|
| 426 | interpolation='nearest', aspect=1, origin='upper', |
---|
| 427 | extent=[xmin, xmax, ymin, ymax], vmin=vmin, vmax=vmax) |
---|
| 428 | |
---|
| 429 | |
---|
| 430 | def demo(): |
---|
| 431 | data = load_data('DEC07086.DAT') |
---|
| 432 | set_beam_stop(data, 0.004) |
---|
| 433 | plot_data(data) |
---|
| 434 | import matplotlib.pyplot as plt; plt.show() |
---|
| 435 | |
---|
| 436 | |
---|
| 437 | if __name__ == "__main__": |
---|
| 438 | demo() |
---|