Home

Welcome to the MIDAS wiki!

MIDAS allows for reduction of Near-Field and Far-Field High Energy Diffraction Data to obtain microstructural information. MIDAS can be run both locally and on a HPC server. All required packages are downloaded automatically during installation.

Installation

To install on MACOS: go to MAC (ARM) Install and follow instructions before continuing.

MIDAS has hard-coded paths, requiring it to be installed in the ${USER.HOME}/opt directory. The following steps may be followed for installation (steps in parentheses are optional):

(mkdir opt;) cd ~/opt

git clone https://github.com/marinerhemant/MIDAS.git

cd MIDAS/FF_HEDM

pip install -r requirements.txt

make local

cd ../NF_HEDM

make local

cd ../TOMO

make all

In case it hangs during compilation of external libraries, run make clean and then try make local again.

Execution

TEST FF: Go to the ~/opt/MIDAS/FF_HEDM/Example folder and first generate a simulated data. To do this (minimum RAM size 20 GB) using the following command:

~/opt/MIDAS/FF_HEDM/bin/ForwardSimulationCompressed Parameters.txt

In case of missing libraries error, run source ~/.MIDAS/paths to locate the libraries.

Then generate the input file required to run a reconstruction using the following command:

python ~/opt/MIDAS/utils/ffGenerateZip.py -resultFolder . -paramFN Parameters.txt -dataFN Au_FF_000001_pf_scanNr_0.zip

Then run the reconstruction using the following command:

python ~/opt/MIDAS/FF_HEDM/v7/ff_MIDAS.py -dataFN Au_FF_000001_pf_scanNr_0.zip.analysis.MIDAS.zip -nCPUs 50 -convertFiles 0

Use as many CPU cores as you want. This will run locally and save the results in a subfolder named LayerNr_1. You can compare your results to example results in ~/opt/MIDAS/FF_HEDM/Example/GrainsReconstructed.csv

The following might not work!!!

FF Stitching manual: FF-HEDM including stitching

FF GrainTracking manual: FF-Tracking

NF manual: NF-HEDM

FF DL Trial: Once MIDAS is installed in ~/opt/MIDAS using (local), ~/opt/MIDAS/utils/DL2FF.py code can be used to run an example analysis. Only variables are peaksFN, paramFile and nCPUs to use.

Find missing grains FF-seeding for NF: Seeding

HDF files generation: HDF

FF Integration Manual: WAXS-Integration