1. Create networkx graphs 1-1. Order the grainID_n.txt/orientations_n.txt properly. 1-2. Make Dream3D files over all the grainID_n.txt/orientations_n.txt files by Dream3D pipeline 1-3. Read 3D np array (grainIDs) from the Dream3D files and duplicate the array by 3x3x3 (np.tile) and re-write the grainID_n.txt 1-4. Re-do the 2. by the re-written grainID_n.txt 4. 1-5. From the result of 5., make the networkx graph.

2. Create grainID-FIP table (max sub-band and elementwise-averaged)

3. Create scalers for all train/validation/test scenarios

4. Create pytorch datalist for all train/validation/test scenarios