Getting started
The majority of APAV is split into 2 imports. The core import loads the main classes that represent data in APAV:
>>> import apav
Or more succinctly:
>>> import apav as ap
This brings in core containers such as Roi, Range, RangeCollection, and others. The second import loads most
of the analysis processes, such as mass spectrum quantification routines, correlation histograms, density maps, etc:
>>> import apav.analysis
or shortened:
>>> import apav.analysis as anl
Importing data
Positional data
Positional data is stored and accessed in the Roi class. See the Roi page for detailed information.
Loading positional data into Roi objects is done through its alternate constructors (i.e. @classmethod )
of the Roi object. For example, loading a POS file into a new Roi instance is simply:
>>> import apav as ap
>>> pos_roi = ap.Roi.from_pos(r"/path/to/experimental/data.pos")
Similarly for other file types:
>>> import apav as ap
>>> epos_roi = ap.Roi.from_epos(r"/path/to/experimental/data.epos")
>>> ato_roi = ap.Roi.from_ato(r"/path/to/experimental/data.ato")
The Roi instance can then be used directly, or passed to other analysis routines.
Ranging data
Mass ranging uses the dedicated container RangeCollection, which can read RNG or RRNG files in the same
manor as the Roi class with positional data. That is, through @classmethod alternate constructors. For
example, loading a RRNG file of a Si material:
>>> import apav as ap
>>> rng = ap.RangeCollection.from_rng(r"/path/to/ranges.rng")
>>> rrng = ap.RangeCollection.from_rrng(r"/path/to/ranges.rrng")
Now print a summary of the RRNG file:
>>> print(rrng)
Range data set
Number of ranges: 25
Mass range: 5.896 - 86.555
Number of unique elements: 5
Elements: ('Si', 'Cu', 'C', 'O', 'Cr')
Composition Min (Da) Max (Da) Volume Color (RGB 0-1)
------------- ---------- ---------- -------- -----------------
C 5.896 6.193 0.00878 (0.4, 0.0, 0.2)
C 11.866 12.198 0.00878 (0.4, 0.0, 0.2)
Si 13.8745 14.241 0.02003 (0.8, 0.8, 0.8)
Si 14.407 14.643 0.02003 (0.8, 0.8, 0.8)
Si 14.912 15.171 0.02003 (0.8, 0.8, 0.8)
O2 15.858 16.48 0.02883 (0.0, 0.8, 1.0)
O2 17.838 18.304 0.02883 (0.0, 0.8, 1.0)
Cr 24.895 25.445 0.01201 (1.0, 0.2, 0.8)
Cr 25.771 27.211 0.01201 (1.0, 0.2, 0.8)
Si 27.856 28.595 0.02003 (0.8, 0.8, 0.8)
Si 28.826 29.255 0.02003 (0.8, 0.8, 0.8)
Si 29.783 30.252 0.02003 (0.8, 0.8, 0.8)
CrO 32.892 33.275 0.04083 (1.0, 0.0, 0.0)
CrO 33.898 34.802 0.04083 (1.0, 0.0, 0.0)
CrO 34.917 35.231 0.04083 (1.0, 0.0, 0.0)
CrO2 41.869 43.181 0.06966 (0.0, 1.0, 0.0)
Cr 49.612 50.526 0.01201 (1.0, 0.2, 0.8)
Cr 51.699 54.243 0.01201 (1.0, 0.2, 0.8)
Cr2O 57.819 61.159 0.05284 (0.0, 0.0, 1.0)
Cu 62.567 63.496 0.01181 (1.0, 0.4, 0.0)
Cu 64.619 65.548 0.01181 (1.0, 0.4, 0.0)
CrO 65.76 66.264 0.04083 (1.0, 0.0, 0.0)
CrO 67.622 69.574 0.04083 (1.0, 0.0, 0.0)
CrO 69.781 70.156 0.04083 (1.0, 0.0, 0.0)
CrO2 83.595 86.555 0.06966 (0.0, 1.0, 0.0)
Jupyter Notebooks
APAV works in any IDE/code editor, however, Jupyter Lab is recommended due to the ability to segment code
execution into “cells”. This is most advantageous when dealing with datasets with large number of counts where loading
data could take longer than the analysis itself. The
ability to separate data import from analysis can save much time when code needs to be adjusted and re-executed.
See http://jupyterlab.io/ for more information about Jupyter Lab, it is installable straight from pip or conda.
See this example notebook:
Interactive plotting
Many classes in APAV can produce interactive plots. These plots are a combination of the pyqtgraph library and
custom Qt widgets written just for APAV. All of these integrated plots are interactive with the mouse. By default
“right-click” and drag will scale/zoom the plot and “single-click” and drag will move the plot. This can
be changed in the right-click menu to be more usable on trackpads. Many interactive plots in APAV also include
edit fields where the analysis can be adjusted in real-time.
Important
When APAV is being used interactively, such as in a Jupyter Notebook or IPython session, the interpreter must
be told to integrate itself with the Qt event loop. Otherwise none of the interactive plots will work. This
is as simple as including %gui qt at the top of the script.
As an example, here is the interactive correlation histogram window. Note that the %gui qt statements is only
required in interactive sessions such as notebooks or ipython sessions:
>>> %gui qt
>>> import apav as ap
>>> import apav.analysis as anl
>>> roi = ap.Roi.from_epos
>>> corr_hist = anl.CorrelationHistogram(roi)
>>> plot = corr_hist.plot()
>>> plot.show()
Note
Each interactive plot in APAV follows the same template. That is, the plot function returns a Qt object which
must be assigned to an object or it will be garbage collected. Then the plot is shown using
some_plot.show().
For example, this is valid:
plot = some_analysis.some_plot_function()
plot.show()
This is not:
some_analysis.some_plot_function().show()