ArXiv preprint 1903.11815: Supplementary material

This repo contains the data analysis code in the form of Jupyter notebooks and conventional Python scripts.

Data files

Because of Github upload limits, the data sets are available at the publicly visible Google Drive folder here. These are the BCDI data, after pre-processing, background reduction and removal of stray scattering. They are in the .mat format, and can be opened in Matlab, Octave or Pyhon (using the scipy.io module).

File 111-A.mat

BCDI data set corresponding to the $\left[111\right]$ reflection of the Bragg peak Friedel pair.

File 111-B.mat

BCDI data set corresponding to the $\left[\bar{1}\bar{1}\bar{1}\right]$ reflection of the Bragg peak Friedel pair.

File 200-D.mat

BCDI data set corresponding to the $\left[200\right]$ reflection of the second grain.

File stdSample.mat

BCDI data set corresponding to the $\left[111\right]$ reflection of the dewetted gold nanoparticle.

Miscellaneous (misc/*)

Contains the calculated bases of sampling vectors for each reconstruction (real- and reciprocal-space), estimates of spatially varying strain, and isosurface data.

Computation scripts

Notebooks

Jupyter notebooks were used for the phase retrieval reconstructions and and a few subsequent computations for results seen in the manuscript. Dependencies include Tensorflow 1.5 for the GPU reconstruction code. The notebooks are named correspondingly to the data files described above.

Python scripts

These were used for computations other than those included above.