Changes in src/sas/sasgui/perspectives/fitting/media/fitting_help.rst [8b89396:6d7b252b] in sasview
- File:
-
- 1 edited
Legend:
- Unmodified
- Added
- Removed
-
src/sas/sasgui/perspectives/fitting/media/fitting_help.rst
r8b89396 r6d7b252b 42 42 * *Ellipsoid* - ellipsoidal shapes (oblate,prolate, core shell, etc) 43 43 * *Parellelepiped* - as the name implies 44 * *Sphere* - s heroidal shapes (sphere, core multishell, vesicle, etc)44 * *Sphere* - spheroidal shapes (sphere, core multishell, vesicle, etc) 45 45 * *Lamellae* - lamellar shapes (lamellar, core shell lamellar, stacked 46 46 lamellar, etc) … … 61 61 on the *Description* button to the right. 62 62 63 Product Models 64 ^^^^^^^^^^^^^^ 65 66 .. figure:: p_and_s_buttons.png 67 68 S(Q) models can be combined with models in the other categories to generate 69 what SasView calls "product models". See :ref:`Product_Models` for more 70 information. 71 63 72 Show 1D/2D 64 73 ^^^^^^^^^^ … … 119 128 120 129 For a complete list of all the library models available in SasView, see 121 the `Model Documentation <../../../ index.html>`_ .130 the `Model Documentation <../../../sasgui/perspectives/fitting/models/index.html>`_ . 122 131 123 132 It is also possible to add your own models. … … 341 350 .. ZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZ 342 351 352 Fitting Integer Parameters 353 -------------------------- 354 355 Most of the parameters in SasView models will naturally take floating point (decimal) 356 values, but there are some parameters which can only have integer values. Examples 357 include, but are not limited to, the number of shells in a multilayer vesicle, the 358 number of beads in a pearl necklace, the number of arms of a star polymer, and so on. 359 Wherever possible/recognised, the integer nature of a parameter is specified in the 360 respective model documentation and/or parameter table, so read the documentation 361 carefully! 362 363 Integer parameters must be fitted with care. 364 365 Start with your best possible guess for the value of the parameter. And using 366 *a priori* knowledge, fix as many of the other parameters as possible. 367 368 The SasView optimizers treat integer parameters internally as floating point 369 numbers, but the values presented to the user are truncated or rounded, as 370 appropriate. 371 372 In most instances integer parameters will probably be greater than zero. A good 373 policy in such cases is to use a constraint to enforce this. 374 375 Because an integer parameter should, by definition, only move in integer steps, 376 problems may be encountered if the optimizer step size is too small. Similarly, 377 be **very careful** about applying polydispersity to integer parameters. 378 379 The Levenberg-Marquardt and Quasi-Newton BFGS (and other derivative-based) 380 optimizers are probably best avoided for fitting models with integer parameters. 381 382 .. ZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZ 383 343 384 Fitting Limits 344 385 -------------- … … 426 467 See :ref:`Assessing_Fit_Quality`. 427 468 428 The objective of model-fitting is to find a *physically-plausible* model, and set429 of model parameters, that generate a theory that reproduces the experimental data 430 and gives residual values as close to zero as possible.469 The objective of model-fitting is to find a *physically-plausible* model, and 470 set of model parameters, that generate a theory that reproduces the experimental 471 data and minimizes the values of the residuals. 431 472 432 473 Change the default values of the model parameters by hand until the theory line 433 starts to represent the experimental data. Then uncheck the tick boxes alongside 434 all parameters *except* the 'background' and the 'scale'. Click the *Fit* button. 435 SasView will optimise the values of the 'background' and 'scale' and also display 436 the corresponding uncertainties on the optimised values. 437 438 *NB: If no uncertainty is shown it generally means that the model is not very* 439 *dependent on the corresponding parameter (or that one or more parameters are* 440 *'correlated').* 441 442 In the bottom left corner of the *Fit Page* is a box displaying the normalised value 443 of the statistical $\chi^2$ parameter returned by the optimiser. 474 starts to represent the experimental data. Then check the tick boxes alongside 475 the 'background' and 'scale' parameters. Click the *Fit* button. SasView 476 will optimise the values of the 'background' and 'scale' and also display the 477 corresponding uncertainties on the optimised values. 478 479 .. note:: 480 If the uncertainty on a fitted parameter is unrealistically large, or if it 481 displays as NaN, the model is most likely a poor representation of the data, 482 the parameter in question is highly correlated with one or more of the other 483 fitted parameters, or the model is relatively insensitive to the value of 484 that particular parameter. 485 486 In the bottom left corner of the *Fit Page* is a box displaying a normalised 487 value of the statistical $\chi^2$ parameter (the reduced $\chi^2$, 488 See :ref:`Assessing_Fit_Quality`) returned by the optimiser. 444 489 445 490 Now check the box for another model parameter and click *Fit* again. Repeat this 446 process until most or all parameters are checked and have been optimised. As the447 fit of the theory to the experimental data improves the value of 'chi2/Npts' will 448 decrease. A good model fit should easily produce values of 'chi2/Npts' that are 449 close to one, and certainly < 100. See :ref:`Assessing_Fit_Quality`.491 process until all relevant parameters are checked and have been optimised. As 492 the fit of the theory to the experimental data improves, the value of 'Reduced 493 Chi2' will decrease. A good model fit should produce values of Reduced Chi2 494 close to one, and certainly << 100. See :ref:`Assessing_Fit_Quality`. 450 495 451 496 SasView has a number of different optimisers (see the section :ref:`Fitting_Options`). … … 462 507 *the Data Explorer is checked (see the section* :ref:`Loading_data` *).* 463 508 464 This mode is an extension of the :ref:`Single_Fit_Mode` that fits two or more data 465 sets *to the same model* simultaneously. If necessary it is possible to constrain 466 fit parameters between data sets (eg, to fix a background level, or radius, etc). 509 This mode is an extension of the :ref:`Single_Fit_Mode` that allows for some 510 relatively extensive constraints between fitted parameters in a single *FitPage* 511 or between several *FitPage*'s (eg, to constrain all fitted parameters to be the 512 same in a contrast series of *FitPages* except for the solvent sld parameter, 513 constrain the length to be twice that of the radius in a single *FitPage*, 514 fix the radius of the sphere in one *FitPage* to be the same as the radius of 515 the cylinder in a second *FitPage*, etc). 467 516 468 517 If the data to be fit are in multiple files, load each file, then select each file … … 501 550 next to *Add Constraint?* in the *Fit Constraints* box. 502 551 552 To constrain all identically named parameters to fit *simultaneously* to the 553 same value across all the *Fitpages* use the *Easy Setup* drop-down buttons in 554 the *Const & Simul Fit* page. 555 503 556 *NB: You can only constrain parameters that are set to refine.* 557 558 Constraints will generally be of the form 559 560 Mi Parameter1 = Mj.Parameter1 561 562 however the text box after the '=' sign can be used to adjust this 563 relationship; for example 564 565 Mi Parameter1 = scalar \* Mj.Parameter1 566 567 A 'free-form' constraint box is also provided. 568 569 Many constraints can be entered for a single fit. 504 570 505 571 When ready, click the *Fit* button on the *Const & Simul Fit* page, NOT the *Fit* 506 572 button on the individual *FitPage*'s. 507 573 508 Simultaneous Fits without Constraints509 ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^510 511 574 The results of the model-fitting will be returned to each of the individual 512 575 *FitPage*'s. 513 576 514 Note that the chi2/Npts value returned is the SUM of the chi2/Npts of each fit. To 515 see the chi2/Npts value for a specific *FitPage*, click the *Compute* button at the 516 bottom of that *FitPage* to recalculate. Also see :ref:`Assessing_Fit_Quality`. 517 518 Simultaneous Fits with Constraints 519 ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ 520 521 Use the *Easy Setup* drop-down buttons in the *Const & Simul Fit* page to set 522 up constraints between *FitPage*'s. 523 524 Constraints will generally be of the form 525 526 Mi Parameter1 = Mj.Parameter1 527 528 however the text box after the '=' sign can be used to adjust this 529 relationship; for example 530 531 Mi Parameter1 = scalar \* Mj.Parameter1 532 533 A 'free-form' constraint box is also provided. 534 535 Many constraints can be entered for a single fit. 536 537 The results of the model-fitting will be returned to each of the individual 538 *FitPage*'s. 539 540 Note that the chi2/Npts value returned is the SUM of the chi2/Npts of each fit. To 541 see the chi2/Npts value for a specific *FitPage*, click the *Compute* button at the 542 bottom of that *FitPage* to recalculate. Also see :ref:`Assessing_Fit_Quality`. 577 Note that the Reduced Chi2 value returned is the SUM of the Reduced Chi2 of 578 each fit. To see the Reduced Chi2 value for a specific *FitPage*, click the 579 *Compute* button at the bottom of that *FitPage* to recalculate. Note that in 580 doing so the degrees of freedom will be set to Npts. 581 See :ref:`Assessing_Fit_Quality`. Moreover in the case of constraints the 582 degrees of freedom are less than one might think due to those constraints. 543 583 544 584 .. ZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZ … … 766 806 .. ZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZ 767 807 768 .. note:: This help document was last changed by Paul Butler, 10 September 769 2017 808 .*Document History* 809 810 | 2017-09-10 Paul Butler 811 | 2017-09-15 Steve King 812 | 2018-03-05 Paul Butler
Note: See TracChangeset
for help on using the changeset viewer.
