2018-04-08 18:01

6 files changed, 98 insertions, 3 deletions

diff --git a/processing/chapter.tex b/processing/chapter.tex
index e430e1d..1ec2c37 100644
--- a/processing/chapter.tex
+++ b/processing/chapter.tex
@@ -1004,12 +1004,83 @@ Guess...
 

 Can be used directly...

 

+[USERS CAN WRITE THEIR OWN FUNCTION OBJECTS]

+

+\begin{figure}

+  \includegraphics[width=0.5\textwidth]{"processing/fit_function"}

+  \includepython{"processing/fit_function.py"}

+  \caption[CAPTION TODO]{

+    CAPTION TODO

+  }

+\end{figure}

+

 \subsection{Fitter}  % ----------------------------------------------------------------------------

 

-Loops through...

-Returns model and outs...

+The Fitter class is specially made to work seamlessly with data objects.  %

+

+WrightTools is especially good at dimensionality reduction through fitting.  %

+This concept is best demonstrated through an example.  %

+

+Let’s load in some test data.  %

+\begin{codefragment}{python}

+#import

+import WrightTools as wt

+from WrightTools import datasets

+# create

+ps = datasets.COLORS.v2p1_MoS2_TrEE_movie

+data = wt.data.from_COLORS(ps)

+# cleanup

+data.level('ai0', 'd2', -3)

+data.scale()

+data.convert('eV')

+data.name = 'MoS2'

+\end{codefragment}

+This is a three-dimensional dataset:  %

+\begin{codefragment}{python}

+>>> data.axis_names

+['w2', 'w1', 'd2']

+>>> data.shape

+(41, 41, 23)

+\end{codefragment}

+We could create an animation to see every single pixel, but we can't see everything at once that

+way.  %

+Instead we could imagine fitting every decay ($\tau_{21}$ trace) to an exponential.  %

+Then we could plot the amplitude and time constant of that exponential decay.  %

+This helps us get at subtle questions about the data.  %

+Do the lineshapes narrow with time? Does the redder feature decay slower than the bluer feature?  %

+Faster?  %

+

+Using the \python{Fitter} class, it is easy to perform an exponential fit along each TAU21 trace at

+each OMEGA1, OMEGA2 coordinate.  %

+\begin{codefragment}{python}

+# isolate only relevant data

+data = data.split('w1', 1.75)[1].split('d2', 0)[0]

+# prepare a function

+function = wt.fit.Exponential()

+function.limits['amplitude'] = [0, 1]

+function.limits['offset'] = [0, 0]

+function.limits['tau'] = [0, 2000]

+# do the fit

+fitter = wt.fit.Fitter(function, data, 'd2')

+outs = fitter.run()

+\end{codefragment}

+When we call fitter.run(), every slice of the data object will be fit according to the given

+function object. Fitter automatically creates two new data objects when this happens. outs contains

+the fit parameters, in this case amplitude, tau, and offset. Accordingly, outs is lower-dimensional

+than the original data object. model contains the fit evaluated at each coordinate of the original

+dataset—it’s really useful for inspecting the quality of your fit procedure.

+

+[ALSO GOOD FOR WORKUP OF TUNING DATA: SEE SECTION ...]

+

+\begin{figure}

+  \includegraphics[width=0.4\textwidth]{"processing/fit_amplitude"}

+  \includegraphics[width=0.4\textwidth]{"processing/fit_tau"}

+  \caption[CAPTION TODO]{

+    CAPTION TODO

+  }

+  \label{pro:fig:fitted_movie}

+\end{figure}

 

-\clearpage

 \section{Construction, maintenance, and distribution}  % ==========================================

 

 While WrightTools has already been useful to the work done in the WrightGroup over the last 3

@@ -1098,6 +1169,8 @@ WrightTools uses semantic versioning.  %
 

 Git...

 

+As of 2018-04-08, WrightTools has ........ commits from ..... developers

+

 \subsection{Unit tests}  % ------------------------------------------------------------------------

 

 Unit testing...

diff --git a/processing/fit_amplitude.png b/processing/fit_amplitude.png
new file mode 100644
index 0000000..85e3f02
--- /dev/null
+++ b/processing/fit_amplitude.png
diff --git a/processing/fit_function.png b/processing/fit_function.png
new file mode 100644
index 0000000..ae10662
--- /dev/null
+++ b/processing/fit_function.png
diff --git a/processing/fit_function.py b/processing/fit_function.py
new file mode 100644
index 0000000..247b476
--- /dev/null
+++ b/processing/fit_function.py
@@ -0,0 +1,21 @@
+# import
+import os
+import numpy as np
+import matplotlib.pyplot as plt
+import WrightTools as wt
+from WrightTools import fit
+# define
+here = os.path.abspath(os.path.dirname(__file__))
+# noisey gaussian
+xi = np.linspace(-100, 100, 25)
+yi = 20*np.exp(-0.5*((xi-5)/20.)**2)
+yi = np.random.poisson(yi)
+plt.scatter(xi, yi)
+# fitted
+g = wt.fit.Gaussian()
+ps = g.fit(yi, xi)
+xi = np.linspace(-100, 100, 101)
+model = g.evaluate(ps, xi)
+# plot
+plt.plot(xi, model)
+wt.artists.savefig(os.path.join(here, 'fit_function.png'))
diff --git a/processing/fit_tau.png b/processing/fit_tau.png
new file mode 100644
index 0000000..b68fd76
--- /dev/null
+++ b/processing/fit_tau.png
diff --git a/todo.org b/todo.org
index 1ea787e..443356b 100644
--- a/todo.org
+++ b/todo.org
@@ -25,6 +25,7 @@
 ** TODO figure
 ** TODO arguments
 * TODO insert content                                                  :PbSe:
+* TODO cite important CMDS discoveries, reviews                :introduction:
 * TODO summarize PbSe chapter                                  :introduction:
 * TODO summarize MX2 chapter                                   :introduction:
 * TODO summarize PEDOT:PSS chapter                             :introduction: