diff options
Diffstat (limited to 'doc/fitting.rst')
| -rw-r--r-- | doc/fitting.rst | 17 |
1 files changed, 8 insertions, 9 deletions
diff --git a/doc/fitting.rst b/doc/fitting.rst index 1045653..50a05c8 100644 --- a/doc/fitting.rst +++ b/doc/fitting.rst @@ -129,7 +129,7 @@ Choosing Different Fitting Methods By default, the `Levenberg-Marquardt <https://en.wikipedia.org/wiki/Levenberg-Marquardt_algorithm>`_ algorithm is used for fitting. While often criticized, including the fact it finds a -*local* minima, this approach has some distinct advantages. These include +*local* minimum, this approach has some distinct advantages. These include being fast, and well-behaved for most curve-fitting needs, and making it easy to estimate uncertainties for and correlations between pairs of fit variables, as discussed in :ref:`fit-results-label`. @@ -449,7 +449,7 @@ and standard errors could be done as print('-------------------------------') print('Parameter Value Stderr') for name, param in out.params.items(): - print('{:7s} {:11.5f} {:11.5f}'.format(name, param.value, param.stderr)) + print(f'{name:7s} {param.value:11.5f} {param.stderr:11.5f}') .. _fit-itercb-label: @@ -476,7 +476,7 @@ be used to abort a fit. :type resid: numpy.ndarray :param args: Positional arguments. Must match ``args`` argument to :func:`minimize` :param kws: Keyword arguments. Must match ``kws`` argument to :func:`minimize` - :return: Residual array (generally ``data-model``) to be minimized in the least-squares sense. + :return: Iteration abort flag. :rtype: None for normal behavior, any value like ``True`` to abort the fit. @@ -575,8 +575,6 @@ parameters, which is a similar goal to the one here. x = np.linspace(1, 10, 250) np.random.seed(0) y = 3.0 * np.exp(-x / 2) - 5.0 * np.exp(-(x - 0.1) / 10.) + 0.1 * np.random.randn(x.size) - plt.plot(x, y, 'b') - plt.show() Create a Parameter set for the initial guesses: @@ -603,9 +601,9 @@ and plotting the fit using the Maximum Likelihood solution gives the graph below .. jupyter-execute:: - plt.plot(x, y, 'b') - plt.plot(x, residual(mi.params) + y, 'r', label='best fit') - plt.legend(loc='best') + plt.plot(x, y, 'o') + plt.plot(x, residual(mi.params) + y, label='best fit') + plt.legend() plt.show() Note that the fit here (for which the ``numdifftools`` package is installed) @@ -656,9 +654,10 @@ worked as intended (as a rule of thumb the value should be between 0.2 and .. jupyter-execute:: - plt.plot(res.acceptance_fraction, 'b') + plt.plot(res.acceptance_fraction, 'o') plt.xlabel('walker') plt.ylabel('acceptance fraction') + plt.show() With the results from ``emcee``, we can visualize the posterior distributions for the parameters using the ``corner`` package: |