Cluster expansion does not finish

Alloy Theoretic Automated Toolkit (ATAT)
1

Dear Axel,

I have the problem that the cluster expansion is not complete after all the structures in predstr.out have been calculated. There are only 2 structures left in predstr.out:

siaufung@iwv2901:~/atat_workspace/LiCoxB1-xO2$ more predstr.out
0.250000 0.000000 0.000000 0.500000 0.250000 0.474922 ? u
0.041667 0.208333 0.125000 0.500000 0.125000 -0.056663 48 eg
0.041667 0.208333 0.125000 0.500000 0.125000 -0.044407 49 e

Structures 48 and 49 had relaxed too much, so I excluded them. But maps does not populate the predstr.out with new structures and when I type touch ready, I keep waiting for structures to calculate. What can I do?

My maps.log says: The internal database of structures extends at least up to 0 atoms/unit cell, see predstr.out
Among structures of known energy, true and predicted ground states agree
New ground states with at most 12 atoms/unit cell predicted , see predstr.out
Crossvalidation score: 0.0567012

2

Sorry, sometimes the "next" larger volume takes some time to generate.

I wrote a faster algorithm which you can turn on in the make file. Change

CXXFLAGS=$(PATCHCXXFLAGS) -DSLOWENUMALGO

to

CXXFLAGS=$(PATCHCXXFLAGS)

You can also try some optimization

CXXFLAGS=$(PATCHCXXFLAGS) -O

And then

make
make install

When you restart maps, you force populating the internal database up to where you were with the -g=?? option (here probably -g=12)