Containers of the Nomad Remote Tools Hub (NORTH) are configured such that they have access to the data in the uploads section of the Oasis instance.<br>
Individual examples which exemplify how to use the tools in the apmtools container may have to be unpacked/decompressed before their first use.
Computational requirements: Examples with dataset with a few million ions like most used below should be processable even on a computer with a<br>
single core and four GB main memory for docker tasks. Having multiple CPU cores can be useful as the tools of the paraprobe-toolbox use<br>
multi-threading for most of the numerical and geometrical analyses.<br>
Making guarantees about the maximum data set sizes (in terms of number of ions) is difficult as it strongly depends on which type of analyses<br>
are performed and how these are parameterized. Noteworthy to mention is that even the largest examples at the time of writing this cheatsheet<br>
which are available in the paraprobe-toolbox were processable with a laptop with 32GB main memory in total. Examples with a few million ions<br>
consumed not more than one to eight GB.<br>
</div>
%% Cell type:markdown id:ec70f493 tags:
The paraprobe-toolbox is a collection of software tools for applying computational geometry tasks on point cloud data<br>
such as tomographic reconstructions of atom probe data to extract and characterize microstructural features.<br>
The tool is developed by <ahref="https://arxiv.org/abs/2205.13510">Markus Kühbach et al.</a><br>
The tool is instructed via a jupyter notebook which documents how to chain and script different<br>
analysis steps into a workflow using Python. This can be useful especially for batch processing<br>
on computer clusters.<br>
Tools of the paraprobe-toolbox are chained into computational workflows. Each step uses a different scientific<br>
speciality tool. All these tools have *paraprobe* as a prefix in their executable name.<br>
The tools use CPU parallelization and specific libraries of the computational geometry or other specialists'<br>
communities. The jupyter notebooks enable users to achieve a complete automation of their<br>
data analyses (if this is desired). Internally, each tools keeps track of input and output files via<br>
hashes and time stamps to enable provenance tracking and support repeatable and reproducible research.<br>
All results are openly accessible and documented via so-called <ahref="https://fairmat-experimental.github.io/nexus-fairmat-proposal">NeXus application definitions (see the NORTH/apmtools pages).</a><br>
Such workflows can include parameter studies, mesh processing, writing of reports, and creation of plots.<br>
%% Cell type:markdown id:57dd96fe tags:
### Step 1: Enter the paraprobe-toolbox/teaching/example_analyses sub-directory via the explorer panel to the left.
APAV (Atom Probe Analysis and Visualization) is a Python package for analysis and visualization of atom probe tomography datasets.<br>
The tool is developed by <ahref="https://joss.theoj.org/papers/10.21105/joss.04862">Jesse Smith et al.</a>. Complementary to the design of the paraprobe-toolbox functionalities,<br>
APAV can be chained into workflows via e.g. a jupyter notebook. A particular functional strength and focus of APAV<br>
has been ranging and handling of so-called multi-hit events via a graphical user interface via Python.<br>
APAV has a detailed documentation https://apav.readthedocs.io/en/latest/index.html.
