Should use mkdir -p .. Otherwise, after waiting for a while, the replication script will end without writing out the results because it will find that the results directory already exists. Naturally, people would try to clone the repository and run the script so the results directory is already there. Perhaps the "saved" results from your work should have a different directory name than the new results being generated by the script.
jodokae
commented
6 years ago
Should use mkdir -p .. Otherwise, after waiting for a while, the replication script will end without writing out the results because it will find that the results directory already exists. Naturally, people would try to clone the repository and run the script so the results directory is already there. Perhaps the "saved" results from your work should have a different directory name than the new results being generated by the script.