[87985ca] | 1 | """ |
---|
| 2 | Sasview model constructor. |
---|
| 3 | |
---|
| 4 | Given a module defining an OpenCL kernel such as sasmodels.models.cylinder, |
---|
| 5 | create a sasview model class to run that kernel as follows:: |
---|
| 6 | |
---|
[92d38285] | 7 | from sasmodels.sasview_model import load_custom_model |
---|
| 8 | CylinderModel = load_custom_model('sasmodels/models/cylinder.py') |
---|
[87985ca] | 9 | """ |
---|
[4d76711] | 10 | from __future__ import print_function |
---|
[87985ca] | 11 | |
---|
[ce27e21] | 12 | import math |
---|
| 13 | from copy import deepcopy |
---|
[2622b3f] | 14 | import collections |
---|
[4d76711] | 15 | import traceback |
---|
| 16 | import logging |
---|
[724257c] | 17 | from os.path import basename, splitext, abspath, getmtime |
---|
[9f8ade1] | 18 | try: |
---|
| 19 | import _thread as thread |
---|
| 20 | except ImportError: |
---|
| 21 | import thread |
---|
[ce27e21] | 22 | |
---|
[7ae2b7f] | 23 | import numpy as np # type: ignore |
---|
[ce27e21] | 24 | |
---|
[aa4946b] | 25 | from . import core |
---|
[4d76711] | 26 | from . import custom |
---|
[a80e64c] | 27 | from . import product |
---|
[72a081d] | 28 | from . import generate |
---|
[fb5914f] | 29 | from . import weights |
---|
[6d6508e] | 30 | from . import modelinfo |
---|
[bde38b5] | 31 | from .details import make_kernel_args, dispersion_mesh |
---|
[ff7119b] | 32 | |
---|
[2d81cfe] | 33 | # pylint: disable=unused-import |
---|
[fa5fd8d] | 34 | try: |
---|
[2d81cfe] | 35 | from typing import (Dict, Mapping, Any, Sequence, Tuple, NamedTuple, |
---|
| 36 | List, Optional, Union, Callable) |
---|
[fa5fd8d] | 37 | from .modelinfo import ModelInfo, Parameter |
---|
| 38 | from .kernel import KernelModel |
---|
| 39 | MultiplicityInfoType = NamedTuple( |
---|
[a9bc435] | 40 | 'MultiplicityInfo', |
---|
[fa5fd8d] | 41 | [("number", int), ("control", str), ("choices", List[str]), |
---|
| 42 | ("x_axis_label", str)]) |
---|
[60f03de] | 43 | SasviewModelType = Callable[[int], "SasviewModel"] |
---|
[fa5fd8d] | 44 | except ImportError: |
---|
| 45 | pass |
---|
[2d81cfe] | 46 | # pylint: enable=unused-import |
---|
[fa5fd8d] | 47 | |
---|
[724257c] | 48 | logger = logging.getLogger(__name__) |
---|
| 49 | |
---|
[a38b065] | 50 | calculation_lock = thread.allocate_lock() |
---|
| 51 | |
---|
[724257c] | 52 | #: True if pre-existing plugins, with the old names and parameters, should |
---|
| 53 | #: continue to be supported. |
---|
[c95dfc63] | 54 | SUPPORT_OLD_STYLE_PLUGINS = True |
---|
| 55 | |
---|
[fa5fd8d] | 56 | # TODO: separate x_axis_label from multiplicity info |
---|
| 57 | MultiplicityInfo = collections.namedtuple( |
---|
| 58 | 'MultiplicityInfo', |
---|
| 59 | ["number", "control", "choices", "x_axis_label"], |
---|
| 60 | ) |
---|
| 61 | |
---|
[724257c] | 62 | #: set of defined models (standard and custom) |
---|
| 63 | MODELS = {} # type: Dict[str, SasviewModelType] |
---|
[839fd68] | 64 | # TODO: remove unused MODEL_BY_PATH cache once sasview no longer references it |
---|
[724257c] | 65 | #: custom model {path: model} mapping so we can check timestamps |
---|
| 66 | MODEL_BY_PATH = {} # type: Dict[str, SasviewModelType] |
---|
[d321747] | 67 | #: Track modules that we have loaded so we can determine whether the model |
---|
| 68 | #: has changed since we last reloaded. |
---|
| 69 | _CACHED_MODULE = {} # type: Dict[str, "module"] |
---|
[724257c] | 70 | |
---|
[92d38285] | 71 | def find_model(modelname): |
---|
[b32dafd] | 72 | # type: (str) -> SasviewModelType |
---|
| 73 | """ |
---|
| 74 | Find a model by name. If the model name ends in py, try loading it from |
---|
| 75 | custom models, otherwise look for it in the list of builtin models. |
---|
| 76 | """ |
---|
[92d38285] | 77 | # TODO: used by sum/product model to load an existing model |
---|
| 78 | # TODO: doesn't handle custom models properly |
---|
| 79 | if modelname.endswith('.py'): |
---|
| 80 | return load_custom_model(modelname) |
---|
| 81 | elif modelname in MODELS: |
---|
| 82 | return MODELS[modelname] |
---|
| 83 | else: |
---|
| 84 | raise ValueError("unknown model %r"%modelname) |
---|
| 85 | |
---|
[56b2687] | 86 | |
---|
[fa5fd8d] | 87 | # TODO: figure out how to say that the return type is a subclass |
---|
[4d76711] | 88 | def load_standard_models(): |
---|
[60f03de] | 89 | # type: () -> List[SasviewModelType] |
---|
[4d76711] | 90 | """ |
---|
| 91 | Load and return the list of predefined models. |
---|
| 92 | |
---|
| 93 | If there is an error loading a model, then a traceback is logged and the |
---|
| 94 | model is not returned. |
---|
| 95 | """ |
---|
| 96 | for name in core.list_models(): |
---|
| 97 | try: |
---|
[92d38285] | 98 | MODELS[name] = _make_standard_model(name) |
---|
[ee8f734] | 99 | except Exception: |
---|
[724257c] | 100 | logger.error(traceback.format_exc()) |
---|
[c95dfc63] | 101 | if SUPPORT_OLD_STYLE_PLUGINS: |
---|
| 102 | _register_old_models() |
---|
| 103 | |
---|
[724257c] | 104 | return list(MODELS.values()) |
---|
[de97440] | 105 | |
---|
[4d76711] | 106 | |
---|
| 107 | def load_custom_model(path): |
---|
[60f03de] | 108 | # type: (str) -> SasviewModelType |
---|
[4d76711] | 109 | """ |
---|
| 110 | Load a custom model given the model path. |
---|
[ff7119b] | 111 | """ |
---|
[724257c] | 112 | #logger.info("Loading model %s", path) |
---|
[d321747] | 113 | |
---|
| 114 | # Load the kernel module. This may already be cached by the loader, so |
---|
| 115 | # only requires checking the timestamps of the dependents. |
---|
[4d76711] | 116 | kernel_module = custom.load_custom_kernel_module(path) |
---|
[d321747] | 117 | |
---|
| 118 | # Check if the module has changed since we last looked. |
---|
| 119 | reloaded = kernel_module != _CACHED_MODULE.get(path, None) |
---|
| 120 | _CACHED_MODULE[path] = kernel_module |
---|
| 121 | |
---|
| 122 | # Turn the module into a model. We need to do this in even if the |
---|
| 123 | # model has already been loaded so that we can determine the model |
---|
| 124 | # name and retrieve it from the MODELS cache. |
---|
| 125 | model = getattr(kernel_module, 'Model', None) |
---|
| 126 | if model is not None: |
---|
[9457498] | 127 | # Old style models do not set the name in the class attributes, so |
---|
| 128 | # set it here; this name will be overridden when the object is created |
---|
| 129 | # with an instance variable that has the same value. |
---|
| 130 | if model.name == "": |
---|
| 131 | model.name = splitext(basename(path))[0] |
---|
[20a70bc] | 132 | if not hasattr(model, 'filename'): |
---|
[724257c] | 133 | model.filename = abspath(kernel_module.__file__).replace('.pyc', '.py') |
---|
[e4bf271] | 134 | if not hasattr(model, 'id'): |
---|
| 135 | model.id = splitext(basename(model.filename))[0] |
---|
[724257c] | 136 | else: |
---|
[56b2687] | 137 | model_info = modelinfo.make_model_info(kernel_module) |
---|
[bcdd6c9] | 138 | model = make_model_from_info(model_info) |
---|
[ed10b57] | 139 | |
---|
[2f2c70c] | 140 | # If a model name already exists and we are loading a different model, |
---|
| 141 | # use the model file name as the model name. |
---|
| 142 | if model.name in MODELS and not model.filename == MODELS[model.name].filename: |
---|
| 143 | _previous_name = model.name |
---|
| 144 | model.name = model.id |
---|
[bf8c271] | 145 | |
---|
[2f2c70c] | 146 | # If the new model name is still in the model list (for instance, |
---|
| 147 | # if we put a cylinder.py in our plug-in directory), then append |
---|
| 148 | # an identifier. |
---|
| 149 | if model.name in MODELS and not model.filename == MODELS[model.name].filename: |
---|
| 150 | model.name = model.id + '_user' |
---|
[724257c] | 151 | logger.info("Model %s already exists: using %s [%s]", |
---|
| 152 | _previous_name, model.name, model.filename) |
---|
[ed10b57] | 153 | |
---|
[d321747] | 154 | # Only update the model if the module has changed |
---|
| 155 | if reloaded or model.name not in MODELS: |
---|
| 156 | MODELS[model.name] = model |
---|
| 157 | |
---|
| 158 | return MODELS[model.name] |
---|
[4d76711] | 159 | |
---|
[87985ca] | 160 | |
---|
[bcdd6c9] | 161 | def make_model_from_info(model_info): |
---|
| 162 | # type: (ModelInfo) -> SasviewModelType |
---|
| 163 | """ |
---|
| 164 | Convert *model_info* into a SasView model wrapper. |
---|
| 165 | """ |
---|
| 166 | def __init__(self, multiplicity=None): |
---|
| 167 | SasviewModel.__init__(self, multiplicity=multiplicity) |
---|
| 168 | attrs = _generate_model_attributes(model_info) |
---|
| 169 | attrs['__init__'] = __init__ |
---|
| 170 | attrs['filename'] = model_info.filename |
---|
| 171 | ConstructedModel = type(model_info.name, (SasviewModel,), attrs) # type: SasviewModelType |
---|
| 172 | return ConstructedModel |
---|
| 173 | |
---|
| 174 | |
---|
[4d76711] | 175 | def _make_standard_model(name): |
---|
[60f03de] | 176 | # type: (str) -> SasviewModelType |
---|
[ff7119b] | 177 | """ |
---|
[4d76711] | 178 | Load the sasview model defined by *name*. |
---|
[72a081d] | 179 | |
---|
[4d76711] | 180 | *name* can be a standard model name or a path to a custom model. |
---|
[87985ca] | 181 | |
---|
[4d76711] | 182 | Returns a class that can be used directly as a sasview model. |
---|
[ff7119b] | 183 | """ |
---|
[4d76711] | 184 | kernel_module = generate.load_kernel_module(name) |
---|
[fa5fd8d] | 185 | model_info = modelinfo.make_model_info(kernel_module) |
---|
[bcdd6c9] | 186 | return make_model_from_info(model_info) |
---|
[72a081d] | 187 | |
---|
| 188 | |
---|
[724257c] | 189 | def _register_old_models(): |
---|
| 190 | # type: () -> None |
---|
| 191 | """ |
---|
| 192 | Place the new models into sasview under the old names. |
---|
| 193 | |
---|
| 194 | Monkey patch sas.sascalc.fit as sas.models so that sas.models.pluginmodel |
---|
| 195 | is available to the plugin modules. |
---|
| 196 | """ |
---|
| 197 | import sys |
---|
| 198 | import sas # needed in order to set sas.models |
---|
| 199 | import sas.sascalc.fit |
---|
| 200 | sys.modules['sas.models'] = sas.sascalc.fit |
---|
| 201 | sas.models = sas.sascalc.fit |
---|
| 202 | import sas.models |
---|
| 203 | from sasmodels.conversion_table import CONVERSION_TABLE |
---|
[e65c3ba] | 204 | |
---|
[724257c] | 205 | for new_name, conversion in CONVERSION_TABLE.get((3, 1, 2), {}).items(): |
---|
| 206 | # CoreShellEllipsoidModel => core_shell_ellipsoid:1 |
---|
| 207 | new_name = new_name.split(':')[0] |
---|
| 208 | old_name = conversion[0] if len(conversion) < 3 else conversion[2] |
---|
| 209 | module_attrs = {old_name: find_model(new_name)} |
---|
| 210 | ConstructedModule = type(old_name, (), module_attrs) |
---|
| 211 | old_path = 'sas.models.' + old_name |
---|
| 212 | setattr(sas.models, old_path, ConstructedModule) |
---|
| 213 | sys.modules[old_path] = ConstructedModule |
---|
| 214 | |
---|
| 215 | |
---|
[a80e64c] | 216 | def MultiplicationModel(form_factor, structure_factor): |
---|
| 217 | # type: ("SasviewModel", "SasviewModel") -> "SasviewModel" |
---|
[e65c3ba] | 218 | """ |
---|
| 219 | Returns a constructed product model from form_factor and structure_factor. |
---|
| 220 | """ |
---|
[a80e64c] | 221 | model_info = product.make_product_info(form_factor._model_info, |
---|
| 222 | structure_factor._model_info) |
---|
[bcdd6c9] | 223 | ConstructedModel = make_model_from_info(model_info) |
---|
[a06af5d] | 224 | return ConstructedModel(form_factor.multiplicity) |
---|
[a80e64c] | 225 | |
---|
[ce27e21] | 226 | |
---|
[fa5fd8d] | 227 | def _generate_model_attributes(model_info): |
---|
| 228 | # type: (ModelInfo) -> Dict[str, Any] |
---|
| 229 | """ |
---|
| 230 | Generate the class attributes for the model. |
---|
| 231 | |
---|
| 232 | This should include all the information necessary to query the model |
---|
| 233 | details so that you do not need to instantiate a model to query it. |
---|
| 234 | |
---|
| 235 | All the attributes should be immutable to avoid accidents. |
---|
| 236 | """ |
---|
| 237 | |
---|
| 238 | # TODO: allow model to override axis labels input/output name/unit |
---|
| 239 | |
---|
[a18c5b3] | 240 | # Process multiplicity |
---|
[fa5fd8d] | 241 | non_fittable = [] # type: List[str] |
---|
[04045f4] | 242 | xlabel = model_info.profile_axes[0] if model_info.profile is not None else "" |
---|
| 243 | variants = MultiplicityInfo(0, "", [], xlabel) |
---|
[a18c5b3] | 244 | for p in model_info.parameters.kernel_parameters: |
---|
[04045f4] | 245 | if p.name == model_info.control: |
---|
[fa5fd8d] | 246 | non_fittable.append(p.name) |
---|
[04045f4] | 247 | variants = MultiplicityInfo( |
---|
[ce176ca] | 248 | len(p.choices) if p.choices else int(p.limits[1]), |
---|
| 249 | p.name, p.choices, xlabel |
---|
[fa5fd8d] | 250 | ) |
---|
| 251 | break |
---|
| 252 | |
---|
[50ec515] | 253 | # Only a single drop-down list parameter available |
---|
| 254 | fun_list = [] |
---|
| 255 | for p in model_info.parameters.kernel_parameters: |
---|
| 256 | if p.choices: |
---|
| 257 | fun_list = p.choices |
---|
| 258 | if p.length > 1: |
---|
| 259 | non_fittable.extend(p.id+str(k) for k in range(1, p.length+1)) |
---|
| 260 | break |
---|
| 261 | |
---|
[a18c5b3] | 262 | # Organize parameter sets |
---|
[fa5fd8d] | 263 | orientation_params = [] |
---|
| 264 | magnetic_params = [] |
---|
| 265 | fixed = [] |
---|
[85fe7f8] | 266 | for p in model_info.parameters.user_parameters({}, is2d=True): |
---|
[fa5fd8d] | 267 | if p.type == 'orientation': |
---|
| 268 | orientation_params.append(p.name) |
---|
| 269 | orientation_params.append(p.name+".width") |
---|
| 270 | fixed.append(p.name+".width") |
---|
[32e3c9b] | 271 | elif p.type == 'magnetic': |
---|
[fa5fd8d] | 272 | orientation_params.append(p.name) |
---|
| 273 | magnetic_params.append(p.name) |
---|
| 274 | fixed.append(p.name+".width") |
---|
[a18c5b3] | 275 | |
---|
[32e3c9b] | 276 | |
---|
[a18c5b3] | 277 | # Build class dictionary |
---|
| 278 | attrs = {} # type: Dict[str, Any] |
---|
| 279 | attrs['_model_info'] = model_info |
---|
| 280 | attrs['name'] = model_info.name |
---|
| 281 | attrs['id'] = model_info.id |
---|
| 282 | attrs['description'] = model_info.description |
---|
| 283 | attrs['category'] = model_info.category |
---|
| 284 | attrs['is_structure_factor'] = model_info.structure_factor |
---|
| 285 | attrs['is_form_factor'] = model_info.ER is not None |
---|
| 286 | attrs['is_multiplicity_model'] = variants[0] > 1 |
---|
| 287 | attrs['multiplicity_info'] = variants |
---|
[fa5fd8d] | 288 | attrs['orientation_params'] = tuple(orientation_params) |
---|
| 289 | attrs['magnetic_params'] = tuple(magnetic_params) |
---|
| 290 | attrs['fixed'] = tuple(fixed) |
---|
| 291 | attrs['non_fittable'] = tuple(non_fittable) |
---|
[50ec515] | 292 | attrs['fun_list'] = tuple(fun_list) |
---|
[fa5fd8d] | 293 | |
---|
| 294 | return attrs |
---|
[4d76711] | 295 | |
---|
[ce27e21] | 296 | class SasviewModel(object): |
---|
| 297 | """ |
---|
| 298 | Sasview wrapper for opencl/ctypes model. |
---|
| 299 | """ |
---|
[fa5fd8d] | 300 | # Model parameters for the specific model are set in the class constructor |
---|
| 301 | # via the _generate_model_attributes function, which subclasses |
---|
| 302 | # SasviewModel. They are included here for typing and documentation |
---|
| 303 | # purposes. |
---|
| 304 | _model = None # type: KernelModel |
---|
| 305 | _model_info = None # type: ModelInfo |
---|
| 306 | #: load/save name for the model |
---|
| 307 | id = None # type: str |
---|
| 308 | #: display name for the model |
---|
| 309 | name = None # type: str |
---|
| 310 | #: short model description |
---|
| 311 | description = None # type: str |
---|
| 312 | #: default model category |
---|
| 313 | category = None # type: str |
---|
| 314 | |
---|
| 315 | #: names of the orientation parameters in the order they appear |
---|
[724257c] | 316 | orientation_params = None # type: List[str] |
---|
[fa5fd8d] | 317 | #: names of the magnetic parameters in the order they appear |
---|
[724257c] | 318 | magnetic_params = None # type: List[str] |
---|
[fa5fd8d] | 319 | #: names of the fittable parameters |
---|
[724257c] | 320 | fixed = None # type: List[str] |
---|
[fa5fd8d] | 321 | # TODO: the attribute fixed is ill-named |
---|
| 322 | |
---|
| 323 | # Axis labels |
---|
| 324 | input_name = "Q" |
---|
| 325 | input_unit = "A^{-1}" |
---|
| 326 | output_name = "Intensity" |
---|
| 327 | output_unit = "cm^{-1}" |
---|
| 328 | |
---|
| 329 | #: default cutoff for polydispersity |
---|
| 330 | cutoff = 1e-5 |
---|
| 331 | |
---|
| 332 | # Note: Use non-mutable values for class attributes to avoid errors |
---|
| 333 | #: parameters that are not fitted |
---|
| 334 | non_fittable = () # type: Sequence[str] |
---|
| 335 | |
---|
| 336 | #: True if model should appear as a structure factor |
---|
| 337 | is_structure_factor = False |
---|
| 338 | #: True if model should appear as a form factor |
---|
| 339 | is_form_factor = False |
---|
| 340 | #: True if model has multiplicity |
---|
| 341 | is_multiplicity_model = False |
---|
[1f35235] | 342 | #: Multiplicity information |
---|
[fa5fd8d] | 343 | multiplicity_info = None # type: MultiplicityInfoType |
---|
| 344 | |
---|
| 345 | # Per-instance variables |
---|
| 346 | #: parameter {name: value} mapping |
---|
| 347 | params = None # type: Dict[str, float] |
---|
| 348 | #: values for dispersion width, npts, nsigmas and type |
---|
| 349 | dispersion = None # type: Dict[str, Any] |
---|
| 350 | #: units and limits for each parameter |
---|
[60f03de] | 351 | details = None # type: Dict[str, Sequence[Any]] |
---|
| 352 | # # actual type is Dict[str, List[str, float, float]] |
---|
[04dc697] | 353 | #: multiplicity value, or None if no multiplicity on the model |
---|
[fa5fd8d] | 354 | multiplicity = None # type: Optional[int] |
---|
[04dc697] | 355 | #: memory for polydispersity array if using ArrayDispersion (used by sasview). |
---|
| 356 | _persistency_dict = None # type: Dict[str, Tuple[np.ndarray, np.ndarray]] |
---|
[fa5fd8d] | 357 | |
---|
| 358 | def __init__(self, multiplicity=None): |
---|
[04dc697] | 359 | # type: (Optional[int]) -> None |
---|
[2622b3f] | 360 | |
---|
[04045f4] | 361 | # TODO: _persistency_dict to persistency_dict throughout sasview |
---|
| 362 | # TODO: refactor multiplicity to encompass variants |
---|
| 363 | # TODO: dispersion should be a class |
---|
[fa5fd8d] | 364 | # TODO: refactor multiplicity info |
---|
| 365 | # TODO: separate profile view from multiplicity |
---|
| 366 | # The button label, x and y axis labels and scale need to be under |
---|
| 367 | # the control of the model, not the fit page. Maximum flexibility, |
---|
| 368 | # the fit page would supply the canvas and the profile could plot |
---|
| 369 | # how it wants, but this assumes matplotlib. Next level is that |
---|
| 370 | # we provide some sort of data description including title, labels |
---|
| 371 | # and lines to plot. |
---|
| 372 | |
---|
[1f35235] | 373 | # Get the list of hidden parameters given the multiplicity |
---|
[04045f4] | 374 | # Don't include multiplicity in the list of parameters |
---|
[fa5fd8d] | 375 | self.multiplicity = multiplicity |
---|
[04045f4] | 376 | if multiplicity is not None: |
---|
| 377 | hidden = self._model_info.get_hidden_parameters(multiplicity) |
---|
| 378 | hidden |= set([self.multiplicity_info.control]) |
---|
| 379 | else: |
---|
| 380 | hidden = set() |
---|
[8f93522] | 381 | if self._model_info.structure_factor: |
---|
| 382 | hidden.add('scale') |
---|
| 383 | hidden.add('background') |
---|
| 384 | self._model_info.parameters.defaults['background'] = 0. |
---|
[04045f4] | 385 | |
---|
[04dc697] | 386 | self._persistency_dict = {} |
---|
[fa5fd8d] | 387 | self.params = collections.OrderedDict() |
---|
[b3a85cd] | 388 | self.dispersion = collections.OrderedDict() |
---|
[fa5fd8d] | 389 | self.details = {} |
---|
[8977226] | 390 | for p in self._model_info.parameters.user_parameters({}, is2d=True): |
---|
[04045f4] | 391 | if p.name in hidden: |
---|
[fa5fd8d] | 392 | continue |
---|
[fcd7bbd] | 393 | self.params[p.name] = p.default |
---|
[fa5fd8d] | 394 | self.details[p.id] = [p.units, p.limits[0], p.limits[1]] |
---|
[fb5914f] | 395 | if p.polydisperse: |
---|
[fa5fd8d] | 396 | self.details[p.id+".width"] = [ |
---|
| 397 | "", 0.0, 1.0 if p.relative_pd else np.inf |
---|
| 398 | ] |
---|
[fb5914f] | 399 | self.dispersion[p.name] = { |
---|
| 400 | 'width': 0, |
---|
| 401 | 'npts': 35, |
---|
| 402 | 'nsigmas': 3, |
---|
| 403 | 'type': 'gaussian', |
---|
| 404 | } |
---|
[ce27e21] | 405 | |
---|
[de97440] | 406 | def __get_state__(self): |
---|
[fa5fd8d] | 407 | # type: () -> Dict[str, Any] |
---|
[de97440] | 408 | state = self.__dict__.copy() |
---|
[4d76711] | 409 | state.pop('_model') |
---|
[de97440] | 410 | # May need to reload model info on set state since it has pointers |
---|
| 411 | # to python implementations of Iq, etc. |
---|
| 412 | #state.pop('_model_info') |
---|
| 413 | return state |
---|
| 414 | |
---|
| 415 | def __set_state__(self, state): |
---|
[fa5fd8d] | 416 | # type: (Dict[str, Any]) -> None |
---|
[de97440] | 417 | self.__dict__ = state |
---|
[fb5914f] | 418 | self._model = None |
---|
[de97440] | 419 | |
---|
[ce27e21] | 420 | def __str__(self): |
---|
[fa5fd8d] | 421 | # type: () -> str |
---|
[ce27e21] | 422 | """ |
---|
| 423 | :return: string representation |
---|
| 424 | """ |
---|
| 425 | return self.name |
---|
| 426 | |
---|
| 427 | def is_fittable(self, par_name): |
---|
[fa5fd8d] | 428 | # type: (str) -> bool |
---|
[ce27e21] | 429 | """ |
---|
| 430 | Check if a given parameter is fittable or not |
---|
| 431 | |
---|
| 432 | :param par_name: the parameter name to check |
---|
| 433 | """ |
---|
[e758662] | 434 | return par_name in self.fixed |
---|
[ce27e21] | 435 | #For the future |
---|
| 436 | #return self.params[str(par_name)].is_fittable() |
---|
| 437 | |
---|
| 438 | |
---|
| 439 | def getProfile(self): |
---|
[fa5fd8d] | 440 | # type: () -> (np.ndarray, np.ndarray) |
---|
[ce27e21] | 441 | """ |
---|
| 442 | Get SLD profile |
---|
| 443 | |
---|
| 444 | : return: (z, beta) where z is a list of depth of the transition points |
---|
| 445 | beta is a list of the corresponding SLD values |
---|
| 446 | """ |
---|
[745b7bb] | 447 | args = {} # type: Dict[str, Any] |
---|
[fa5fd8d] | 448 | for p in self._model_info.parameters.kernel_parameters: |
---|
| 449 | if p.id == self.multiplicity_info.control: |
---|
[745b7bb] | 450 | value = float(self.multiplicity) |
---|
[fa5fd8d] | 451 | elif p.length == 1: |
---|
[745b7bb] | 452 | value = self.params.get(p.id, np.NaN) |
---|
[fa5fd8d] | 453 | else: |
---|
[745b7bb] | 454 | value = np.array([self.params.get(p.id+str(k), np.NaN) |
---|
[b32dafd] | 455 | for k in range(1, p.length+1)]) |
---|
[745b7bb] | 456 | args[p.id] = value |
---|
| 457 | |
---|
[e7fe459] | 458 | x, y = self._model_info.profile(**args) |
---|
| 459 | return x, 1e-6*y |
---|
[ce27e21] | 460 | |
---|
| 461 | def setParam(self, name, value): |
---|
[fa5fd8d] | 462 | # type: (str, float) -> None |
---|
[ce27e21] | 463 | """ |
---|
| 464 | Set the value of a model parameter |
---|
| 465 | |
---|
| 466 | :param name: name of the parameter |
---|
| 467 | :param value: value of the parameter |
---|
| 468 | |
---|
| 469 | """ |
---|
| 470 | # Look for dispersion parameters |
---|
| 471 | toks = name.split('.') |
---|
[de0c4ba] | 472 | if len(toks) == 2: |
---|
[ce27e21] | 473 | for item in self.dispersion.keys(): |
---|
[e758662] | 474 | if item == toks[0]: |
---|
[ce27e21] | 475 | for par in self.dispersion[item]: |
---|
[e758662] | 476 | if par == toks[1]: |
---|
[ce27e21] | 477 | self.dispersion[item][par] = value |
---|
| 478 | return |
---|
| 479 | else: |
---|
| 480 | # Look for standard parameter |
---|
| 481 | for item in self.params.keys(): |
---|
[e758662] | 482 | if item == name: |
---|
[ce27e21] | 483 | self.params[item] = value |
---|
| 484 | return |
---|
| 485 | |
---|
[63b32bb] | 486 | raise ValueError("Model does not contain parameter %s" % name) |
---|
[ce27e21] | 487 | |
---|
| 488 | def getParam(self, name): |
---|
[fa5fd8d] | 489 | # type: (str) -> float |
---|
[ce27e21] | 490 | """ |
---|
| 491 | Set the value of a model parameter |
---|
| 492 | |
---|
| 493 | :param name: name of the parameter |
---|
| 494 | |
---|
| 495 | """ |
---|
| 496 | # Look for dispersion parameters |
---|
| 497 | toks = name.split('.') |
---|
[de0c4ba] | 498 | if len(toks) == 2: |
---|
[ce27e21] | 499 | for item in self.dispersion.keys(): |
---|
[e758662] | 500 | if item == toks[0]: |
---|
[ce27e21] | 501 | for par in self.dispersion[item]: |
---|
[e758662] | 502 | if par == toks[1]: |
---|
[ce27e21] | 503 | return self.dispersion[item][par] |
---|
| 504 | else: |
---|
| 505 | # Look for standard parameter |
---|
| 506 | for item in self.params.keys(): |
---|
[e758662] | 507 | if item == name: |
---|
[ce27e21] | 508 | return self.params[item] |
---|
| 509 | |
---|
[63b32bb] | 510 | raise ValueError("Model does not contain parameter %s" % name) |
---|
[ce27e21] | 511 | |
---|
| 512 | def getParamList(self): |
---|
[04dc697] | 513 | # type: () -> Sequence[str] |
---|
[ce27e21] | 514 | """ |
---|
| 515 | Return a list of all available parameters for the model |
---|
| 516 | """ |
---|
[04dc697] | 517 | param_list = list(self.params.keys()) |
---|
[ce27e21] | 518 | # WARNING: Extending the list with the dispersion parameters |
---|
[de0c4ba] | 519 | param_list.extend(self.getDispParamList()) |
---|
| 520 | return param_list |
---|
[ce27e21] | 521 | |
---|
| 522 | def getDispParamList(self): |
---|
[04dc697] | 523 | # type: () -> Sequence[str] |
---|
[ce27e21] | 524 | """ |
---|
[fb5914f] | 525 | Return a list of polydispersity parameters for the model |
---|
[ce27e21] | 526 | """ |
---|
[1780d59] | 527 | # TODO: fix test so that parameter order doesn't matter |
---|
[3bcb88c] | 528 | ret = ['%s.%s' % (p_name, ext) |
---|
| 529 | for p_name in self.dispersion.keys() |
---|
| 530 | for ext in ('npts', 'nsigmas', 'width')] |
---|
[9404dd3] | 531 | #print(ret) |
---|
[1780d59] | 532 | return ret |
---|
[ce27e21] | 533 | |
---|
| 534 | def clone(self): |
---|
[04dc697] | 535 | # type: () -> "SasviewModel" |
---|
[ce27e21] | 536 | """ Return a identical copy of self """ |
---|
| 537 | return deepcopy(self) |
---|
| 538 | |
---|
| 539 | def run(self, x=0.0): |
---|
[fa5fd8d] | 540 | # type: (Union[float, (float, float), List[float]]) -> float |
---|
[ce27e21] | 541 | """ |
---|
| 542 | Evaluate the model |
---|
| 543 | |
---|
| 544 | :param x: input q, or [q,phi] |
---|
| 545 | |
---|
| 546 | :return: scattering function P(q) |
---|
| 547 | |
---|
| 548 | **DEPRECATED**: use calculate_Iq instead |
---|
| 549 | """ |
---|
[de0c4ba] | 550 | if isinstance(x, (list, tuple)): |
---|
[3c56da87] | 551 | # pylint: disable=unpacking-non-sequence |
---|
[ce27e21] | 552 | q, phi = x |
---|
[60f03de] | 553 | return self.calculate_Iq([q*math.cos(phi)], [q*math.sin(phi)])[0] |
---|
[ce27e21] | 554 | else: |
---|
[60f03de] | 555 | return self.calculate_Iq([x])[0] |
---|
[ce27e21] | 556 | |
---|
| 557 | |
---|
| 558 | def runXY(self, x=0.0): |
---|
[fa5fd8d] | 559 | # type: (Union[float, (float, float), List[float]]) -> float |
---|
[ce27e21] | 560 | """ |
---|
| 561 | Evaluate the model in cartesian coordinates |
---|
| 562 | |
---|
| 563 | :param x: input q, or [qx, qy] |
---|
| 564 | |
---|
| 565 | :return: scattering function P(q) |
---|
| 566 | |
---|
| 567 | **DEPRECATED**: use calculate_Iq instead |
---|
| 568 | """ |
---|
[de0c4ba] | 569 | if isinstance(x, (list, tuple)): |
---|
[60f03de] | 570 | return self.calculate_Iq([x[0]], [x[1]])[0] |
---|
[ce27e21] | 571 | else: |
---|
[60f03de] | 572 | return self.calculate_Iq([x])[0] |
---|
[ce27e21] | 573 | |
---|
| 574 | def evalDistribution(self, qdist): |
---|
[04dc697] | 575 | # type: (Union[np.ndarray, Tuple[np.ndarray, np.ndarray], List[np.ndarray]]) -> np.ndarray |
---|
[d138d43] | 576 | r""" |
---|
[ce27e21] | 577 | Evaluate a distribution of q-values. |
---|
| 578 | |
---|
[d138d43] | 579 | :param qdist: array of q or a list of arrays [qx,qy] |
---|
[ce27e21] | 580 | |
---|
[d138d43] | 581 | * For 1D, a numpy array is expected as input |
---|
[ce27e21] | 582 | |
---|
[d138d43] | 583 | :: |
---|
[ce27e21] | 584 | |
---|
[d138d43] | 585 | evalDistribution(q) |
---|
[ce27e21] | 586 | |
---|
[d138d43] | 587 | where *q* is a numpy array. |
---|
[ce27e21] | 588 | |
---|
[d138d43] | 589 | * For 2D, a list of *[qx,qy]* is expected with 1D arrays as input |
---|
[ce27e21] | 590 | |
---|
[d138d43] | 591 | :: |
---|
[ce27e21] | 592 | |
---|
[d138d43] | 593 | qx = [ qx[0], qx[1], qx[2], ....] |
---|
| 594 | qy = [ qy[0], qy[1], qy[2], ....] |
---|
[ce27e21] | 595 | |
---|
[d138d43] | 596 | If the model is 1D only, then |
---|
[ce27e21] | 597 | |
---|
[d138d43] | 598 | .. math:: |
---|
[ce27e21] | 599 | |
---|
[d138d43] | 600 | q = \sqrt{q_x^2+q_y^2} |
---|
[ce27e21] | 601 | |
---|
| 602 | """ |
---|
[de0c4ba] | 603 | if isinstance(qdist, (list, tuple)): |
---|
[ce27e21] | 604 | # Check whether we have a list of ndarrays [qx,qy] |
---|
| 605 | qx, qy = qdist |
---|
[05df1de] | 606 | return self.calculate_Iq(qx, qy) |
---|
[ce27e21] | 607 | |
---|
| 608 | elif isinstance(qdist, np.ndarray): |
---|
| 609 | # We have a simple 1D distribution of q-values |
---|
| 610 | return self.calculate_Iq(qdist) |
---|
| 611 | |
---|
| 612 | else: |
---|
[3c56da87] | 613 | raise TypeError("evalDistribution expects q or [qx, qy], not %r" |
---|
| 614 | % type(qdist)) |
---|
[ce27e21] | 615 | |
---|
[9dcb21d] | 616 | def calc_composition_models(self, qx): |
---|
[64614ad] | 617 | """ |
---|
[9dcb21d] | 618 | returns parts of the composition model or None if not a composition |
---|
| 619 | model. |
---|
[64614ad] | 620 | """ |
---|
[946c8d27] | 621 | # TODO: have calculate_Iq return the intermediates. |
---|
| 622 | # |
---|
| 623 | # The current interface causes calculate_Iq() to be called twice, |
---|
| 624 | # once to get the combined result and again to get the intermediate |
---|
| 625 | # results. This is necessary for now. |
---|
| 626 | # Long term, the solution is to change the interface to calculate_Iq |
---|
| 627 | # so that it returns a results object containing all the bits: |
---|
[9644b5a] | 628 | # the A, B, C, ... of the composition model (and any subcomponents?) |
---|
[946c8d27] | 629 | # the P and S of the product model, |
---|
| 630 | # the combined model before resolution smearing, |
---|
| 631 | # the sasmodel before sesans conversion, |
---|
| 632 | # the oriented 2D model used to fit oriented usans data, |
---|
| 633 | # the final I(q), |
---|
| 634 | # ... |
---|
[9644b5a] | 635 | # |
---|
[946c8d27] | 636 | # Have the model calculator add all of these blindly to the data |
---|
| 637 | # tree, and update the graphs which contain them. The fitter |
---|
| 638 | # needs to be updated to use the I(q) value only, ignoring the rest. |
---|
| 639 | # |
---|
| 640 | # The simple fix of returning the existing intermediate results |
---|
| 641 | # will not work for a couple of reasons: (1) another thread may |
---|
| 642 | # sneak in to compute its own results before calc_composition_models |
---|
| 643 | # is called, and (2) calculate_Iq is currently called three times: |
---|
| 644 | # once with q, once with q values before qmin and once with q values |
---|
| 645 | # after q max. Both of these should be addressed before |
---|
| 646 | # replacing this code. |
---|
[9644b5a] | 647 | composition = self._model_info.composition |
---|
| 648 | if composition and composition[0] == 'product': # only P*S for now |
---|
| 649 | with calculation_lock: |
---|
| 650 | self._calculate_Iq(qx) |
---|
| 651 | return self._intermediate_results |
---|
| 652 | else: |
---|
| 653 | return None |
---|
[bf8c271] | 654 | |
---|
[fa5fd8d] | 655 | def calculate_Iq(self, qx, qy=None): |
---|
| 656 | # type: (Sequence[float], Optional[Sequence[float]]) -> np.ndarray |
---|
[ff7119b] | 657 | """ |
---|
| 658 | Calculate Iq for one set of q with the current parameters. |
---|
| 659 | |
---|
| 660 | If the model is 1D, use *q*. If 2D, use *qx*, *qy*. |
---|
| 661 | |
---|
| 662 | This should NOT be used for fitting since it copies the *q* vectors |
---|
| 663 | to the card for each evaluation. |
---|
| 664 | """ |
---|
[a38b065] | 665 | ## uncomment the following when trying to debug the uncoordinated calls |
---|
| 666 | ## to calculate_Iq |
---|
| 667 | #if calculation_lock.locked(): |
---|
[724257c] | 668 | # logger.info("calculation waiting for another thread to complete") |
---|
| 669 | # logger.info("\n".join(traceback.format_stack())) |
---|
[a38b065] | 670 | |
---|
| 671 | with calculation_lock: |
---|
| 672 | return self._calculate_Iq(qx, qy) |
---|
| 673 | |
---|
| 674 | def _calculate_Iq(self, qx, qy=None): |
---|
[fb5914f] | 675 | if self._model is None: |
---|
[d2bb604] | 676 | self._model = core.build_model(self._model_info) |
---|
[fa5fd8d] | 677 | if qy is not None: |
---|
| 678 | q_vectors = [np.asarray(qx), np.asarray(qy)] |
---|
| 679 | else: |
---|
| 680 | q_vectors = [np.asarray(qx)] |
---|
[a738209] | 681 | calculator = self._model.make_kernel(q_vectors) |
---|
[6a0d6aa] | 682 | parameters = self._model_info.parameters |
---|
| 683 | pairs = [self._get_weights(p) for p in parameters.call_parameters] |
---|
[9c1a59c] | 684 | #weights.plot_weights(self._model_info, pairs) |
---|
[bde38b5] | 685 | call_details, values, is_magnetic = make_kernel_args(calculator, pairs) |
---|
[4edec6f] | 686 | #call_details.show() |
---|
[05df1de] | 687 | #print("================ parameters ==================") |
---|
| 688 | #for p, v in zip(parameters.call_parameters, pairs): print(p.name, v[0]) |
---|
[ce99754] | 689 | #for k, p in enumerate(self._model_info.parameters.call_parameters): |
---|
| 690 | # print(k, p.name, *pairs[k]) |
---|
[4edec6f] | 691 | #print("params", self.params) |
---|
| 692 | #print("values", values) |
---|
| 693 | #print("is_mag", is_magnetic) |
---|
[6a0d6aa] | 694 | result = calculator(call_details, values, cutoff=self.cutoff, |
---|
[9eb3632] | 695 | magnetic=is_magnetic) |
---|
[ce99754] | 696 | #print("result", result) |
---|
[bf8c271] | 697 | self._intermediate_results = getattr(calculator, 'results', None) |
---|
[a738209] | 698 | calculator.release() |
---|
[d533590] | 699 | #self._model.release() |
---|
[ce27e21] | 700 | return result |
---|
| 701 | |
---|
| 702 | def calculate_ER(self): |
---|
[fa5fd8d] | 703 | # type: () -> float |
---|
[ce27e21] | 704 | """ |
---|
| 705 | Calculate the effective radius for P(q)*S(q) |
---|
| 706 | |
---|
| 707 | :return: the value of the effective radius |
---|
| 708 | """ |
---|
[4bfd277] | 709 | if self._model_info.ER is None: |
---|
[ce27e21] | 710 | return 1.0 |
---|
| 711 | else: |
---|
[4bfd277] | 712 | value, weight = self._dispersion_mesh() |
---|
| 713 | fv = self._model_info.ER(*value) |
---|
[9404dd3] | 714 | #print(values[0].shape, weights.shape, fv.shape) |
---|
[4bfd277] | 715 | return np.sum(weight * fv) / np.sum(weight) |
---|
[ce27e21] | 716 | |
---|
| 717 | def calculate_VR(self): |
---|
[fa5fd8d] | 718 | # type: () -> float |
---|
[ce27e21] | 719 | """ |
---|
| 720 | Calculate the volf ratio for P(q)*S(q) |
---|
| 721 | |
---|
| 722 | :return: the value of the volf ratio |
---|
| 723 | """ |
---|
[4bfd277] | 724 | if self._model_info.VR is None: |
---|
[ce27e21] | 725 | return 1.0 |
---|
| 726 | else: |
---|
[4bfd277] | 727 | value, weight = self._dispersion_mesh() |
---|
| 728 | whole, part = self._model_info.VR(*value) |
---|
| 729 | return np.sum(weight * part) / np.sum(weight * whole) |
---|
[ce27e21] | 730 | |
---|
| 731 | def set_dispersion(self, parameter, dispersion): |
---|
[7c3fb15] | 732 | # type: (str, weights.Dispersion) -> None |
---|
[ce27e21] | 733 | """ |
---|
| 734 | Set the dispersion object for a model parameter |
---|
| 735 | |
---|
| 736 | :param parameter: name of the parameter [string] |
---|
| 737 | :param dispersion: dispersion object of type Dispersion |
---|
| 738 | """ |
---|
[fa800e72] | 739 | if parameter in self.params: |
---|
[1780d59] | 740 | # TODO: Store the disperser object directly in the model. |
---|
[56b2687] | 741 | # The current method of relying on the sasview GUI to |
---|
[fa800e72] | 742 | # remember them is kind of funky. |
---|
[1780d59] | 743 | # Note: can't seem to get disperser parameters from sasview |
---|
[9c1a59c] | 744 | # (1) Could create a sasview model that has not yet been |
---|
[1780d59] | 745 | # converted, assign the disperser to one of its polydisperse |
---|
| 746 | # parameters, then retrieve the disperser parameters from the |
---|
[9c1a59c] | 747 | # sasview model. |
---|
| 748 | # (2) Could write a disperser parameter retriever in sasview. |
---|
| 749 | # (3) Could modify sasview to use sasmodels.weights dispersers. |
---|
[1780d59] | 750 | # For now, rely on the fact that the sasview only ever uses |
---|
| 751 | # new dispersers in the set_dispersion call and create a new |
---|
| 752 | # one instead of trying to assign parameters. |
---|
[ce27e21] | 753 | self.dispersion[parameter] = dispersion.get_pars() |
---|
| 754 | else: |
---|
[7c3fb15] | 755 | raise ValueError("%r is not a dispersity or orientation parameter" |
---|
| 756 | % parameter) |
---|
[ce27e21] | 757 | |
---|
[aa4946b] | 758 | def _dispersion_mesh(self): |
---|
[fa5fd8d] | 759 | # type: () -> List[Tuple[np.ndarray, np.ndarray]] |
---|
[ce27e21] | 760 | """ |
---|
| 761 | Create a mesh grid of dispersion parameters and weights. |
---|
| 762 | |
---|
| 763 | Returns [p1,p2,...],w where pj is a vector of values for parameter j |
---|
| 764 | and w is a vector containing the products for weights for each |
---|
| 765 | parameter set in the vector. |
---|
| 766 | """ |
---|
[4bfd277] | 767 | pars = [self._get_weights(p) |
---|
| 768 | for p in self._model_info.parameters.call_parameters |
---|
| 769 | if p.type == 'volume'] |
---|
[9eb3632] | 770 | return dispersion_mesh(self._model_info, pars) |
---|
[ce27e21] | 771 | |
---|
| 772 | def _get_weights(self, par): |
---|
[fa5fd8d] | 773 | # type: (Parameter) -> Tuple[np.ndarray, np.ndarray] |
---|
[de0c4ba] | 774 | """ |
---|
[fb5914f] | 775 | Return dispersion weights for parameter |
---|
[de0c4ba] | 776 | """ |
---|
[fa5fd8d] | 777 | if par.name not in self.params: |
---|
| 778 | if par.name == self.multiplicity_info.control: |
---|
[32f87a5] | 779 | return self.multiplicity, [self.multiplicity], [1.0] |
---|
[fa5fd8d] | 780 | else: |
---|
[17db833] | 781 | # For hidden parameters use default values. This sets |
---|
| 782 | # scale=1 and background=0 for structure factors |
---|
| 783 | default = self._model_info.parameters.defaults.get(par.name, np.NaN) |
---|
| 784 | return default, [default], [1.0] |
---|
[fa5fd8d] | 785 | elif par.polydisperse: |
---|
[32f87a5] | 786 | value = self.params[par.name] |
---|
[fb5914f] | 787 | dis = self.dispersion[par.name] |
---|
[9c1a59c] | 788 | if dis['type'] == 'array': |
---|
[32f87a5] | 789 | dispersity, weight = dis['values'], dis['weights'] |
---|
[9c1a59c] | 790 | else: |
---|
[32f87a5] | 791 | dispersity, weight = weights.get_weights( |
---|
[9c1a59c] | 792 | dis['type'], dis['npts'], dis['width'], dis['nsigmas'], |
---|
[32f87a5] | 793 | value, par.limits, par.relative_pd) |
---|
| 794 | return value, dispersity, weight |
---|
[fb5914f] | 795 | else: |
---|
[32f87a5] | 796 | value = self.params[par.name] |
---|
[ce99754] | 797 | return value, [value], [1.0] |
---|
[ce27e21] | 798 | |
---|
[749a7d4] | 799 | def test_cylinder(): |
---|
[fa5fd8d] | 800 | # type: () -> float |
---|
[4d76711] | 801 | """ |
---|
[749a7d4] | 802 | Test that the cylinder model runs, returning the value at [0.1,0.1]. |
---|
[4d76711] | 803 | """ |
---|
| 804 | Cylinder = _make_standard_model('cylinder') |
---|
[fb5914f] | 805 | cylinder = Cylinder() |
---|
[b32dafd] | 806 | return cylinder.evalDistribution([0.1, 0.1]) |
---|
[de97440] | 807 | |
---|
[8f93522] | 808 | def test_structure_factor(): |
---|
| 809 | # type: () -> float |
---|
| 810 | """ |
---|
[749a7d4] | 811 | Test that 2-D hardsphere model runs and doesn't produce NaN. |
---|
[8f93522] | 812 | """ |
---|
| 813 | Model = _make_standard_model('hardsphere') |
---|
| 814 | model = Model() |
---|
[17db833] | 815 | value2d = model.evalDistribution([0.1, 0.1]) |
---|
| 816 | value1d = model.evalDistribution(np.array([0.1*np.sqrt(2)])) |
---|
| 817 | #print("hardsphere", value1d, value2d) |
---|
| 818 | if np.isnan(value1d) or np.isnan(value2d): |
---|
| 819 | raise ValueError("hardsphere returns nan") |
---|
[8f93522] | 820 | |
---|
[ce99754] | 821 | def test_product(): |
---|
| 822 | # type: () -> float |
---|
| 823 | """ |
---|
| 824 | Test that 2-D hardsphere model runs and doesn't produce NaN. |
---|
| 825 | """ |
---|
| 826 | S = _make_standard_model('hayter_msa')() |
---|
| 827 | P = _make_standard_model('cylinder')() |
---|
| 828 | model = MultiplicationModel(P, S) |
---|
| 829 | value = model.evalDistribution([0.1, 0.1]) |
---|
| 830 | if np.isnan(value): |
---|
| 831 | raise ValueError("cylinder*hatyer_msa returns null") |
---|
| 832 | |
---|
[04045f4] | 833 | def test_rpa(): |
---|
| 834 | # type: () -> float |
---|
| 835 | """ |
---|
[749a7d4] | 836 | Test that the 2-D RPA model runs |
---|
[04045f4] | 837 | """ |
---|
| 838 | RPA = _make_standard_model('rpa') |
---|
| 839 | rpa = RPA(3) |
---|
[b32dafd] | 840 | return rpa.evalDistribution([0.1, 0.1]) |
---|
[04045f4] | 841 | |
---|
[749a7d4] | 842 | def test_empty_distribution(): |
---|
| 843 | # type: () -> None |
---|
| 844 | """ |
---|
| 845 | Make sure that sasmodels returns NaN when there are no polydispersity points |
---|
| 846 | """ |
---|
| 847 | Cylinder = _make_standard_model('cylinder') |
---|
| 848 | cylinder = Cylinder() |
---|
| 849 | cylinder.setParam('radius', -1.0) |
---|
| 850 | cylinder.setParam('background', 0.) |
---|
| 851 | Iq = cylinder.evalDistribution(np.asarray([0.1])) |
---|
[2d81cfe] | 852 | assert Iq[0] == 0., "empty distribution fails" |
---|
[4d76711] | 853 | |
---|
| 854 | def test_model_list(): |
---|
[fa5fd8d] | 855 | # type: () -> None |
---|
[4d76711] | 856 | """ |
---|
[749a7d4] | 857 | Make sure that all models build as sasview models |
---|
[4d76711] | 858 | """ |
---|
| 859 | from .exception import annotate_exception |
---|
| 860 | for name in core.list_models(): |
---|
| 861 | try: |
---|
| 862 | _make_standard_model(name) |
---|
| 863 | except: |
---|
| 864 | annotate_exception("when loading "+name) |
---|
| 865 | raise |
---|
| 866 | |
---|
[c95dfc63] | 867 | def test_old_name(): |
---|
| 868 | # type: () -> None |
---|
| 869 | """ |
---|
[a69d8cd] | 870 | Load and run cylinder model as sas-models-CylinderModel |
---|
[c95dfc63] | 871 | """ |
---|
| 872 | if not SUPPORT_OLD_STYLE_PLUGINS: |
---|
| 873 | return |
---|
| 874 | try: |
---|
| 875 | # if sasview is not on the path then don't try to test it |
---|
| 876 | import sas |
---|
| 877 | except ImportError: |
---|
| 878 | return |
---|
| 879 | load_standard_models() |
---|
| 880 | from sas.models.CylinderModel import CylinderModel |
---|
| 881 | CylinderModel().evalDistribution([0.1, 0.1]) |
---|
| 882 | |
---|
[05df1de] | 883 | def magnetic_demo(): |
---|
| 884 | Model = _make_standard_model('sphere') |
---|
| 885 | model = Model() |
---|
| 886 | model.setParam('M0:sld', 8) |
---|
| 887 | q = np.linspace(-0.35, 0.35, 500) |
---|
| 888 | qx, qy = np.meshgrid(q, q) |
---|
| 889 | result = model.calculate_Iq(qx.flatten(), qy.flatten()) |
---|
| 890 | result = result.reshape(qx.shape) |
---|
| 891 | |
---|
| 892 | import pylab |
---|
| 893 | pylab.imshow(np.log(result + 0.001)) |
---|
| 894 | pylab.show() |
---|
| 895 | |
---|
[fb5914f] | 896 | if __name__ == "__main__": |
---|
[749a7d4] | 897 | print("cylinder(0.1,0.1)=%g"%test_cylinder()) |
---|
[05df1de] | 898 | #magnetic_demo() |
---|
[ce99754] | 899 | #test_product() |
---|
[17db833] | 900 | #test_structure_factor() |
---|
| 901 | #print("rpa:", test_rpa()) |
---|
[749a7d4] | 902 | #test_empty_distribution() |
---|