crystal-vis

Python code for visualizing crystals given a .cfl

All software used for the visualization is internal to NIST's ITL Laboratory. All code written in a centOS 7 terminal.

Important shell commands

Write into shell:

$ source /usr/local/HEV/.bashhev
$ hev

This sets up the HEV environment for .osg.

Prerequisite work:

For x,y,z coordinates

Highlight the desired rows with the mouse and copy.

Write into shell:

$ cat > coordinates.txt

Then paste into the shell and hit Ctrl + D. This copies the rows into a text file, which can be plugged into a python read() function.

Simply run crystal-vis.py for the desired visualization.

For hkl Rewards

Highlight the desired rows with the mouse and copy.

Write into shell:

$ cat > hkl.txt

Then paste into the shell and hit Ctrl + D. This copies the rows into a text file, which can be plugged into a python read() function.

There exist two files to run, one for writing the data (hkl-vis.py) and one for running the visualization (run.sh). First run the write code (hkl-rewards-vis.py) until it terminates, then write into shell:

$ chmod +x run_count.sh
$ ./run_count.sh

This will output the desired visualization.

For hkl Counts

Copy and Paste the folder containing data for a single simulation inside of Github/crystal-vis/hkl. Make sure the .txt data files within the folder are named "epGreedyResults" + + ".txt".

Use any text editor to edit hkl-count-vis.py by changing variable 'foldername' to your data folder's name.

Run the write code (hkl-counts-vis.py) until it terminates, then write into shell:

$ chmod +x run_count.sh
$ ./run_count.sh

This is currently hard coded to work for epGreedyResults data and the axes have a fixed size, but can be easily changed to it with actor-critic and q-learning data.