When running the parse-evaluator in sequential or random mode, the parameter -t specifies where the sequential/random parses will be written. There is a bug and a theoretical problem with this:
Bug: if the file doesn't already exists, the evaluation fails and exits. If we're creating the sequential/random parses for the same time, the file won't exist. There should be an extra condition in the corresponding FileNotFoundError
Problem: if I am trying to evaluate a parse file that took hours to create (like GC MST-parses) and by mistake I activate the flag -s, the test file will be overwritten. Hopefully you have a backup, but this would be at least annoying.
I suggest that the "-t" option is not required when using sequential/random, and that the outfile is specified with the -o flag
When running the parse-evaluator in sequential or random mode, the parameter -t specifies where the sequential/random parses will be written. There is a bug and a theoretical problem with this:
Bug: if the file doesn't already exists, the evaluation fails and exits. If we're creating the sequential/random parses for the same time, the file won't exist. There should be an extra condition in the corresponding FileNotFoundError
Problem: if I am trying to evaluate a parse file that took hours to create (like GC MST-parses) and by mistake I activate the flag -s, the test file will be overwritten. Hopefully you have a backup, but this would be at least annoying.
I suggest that the "-t" option is not required when using sequential/random, and that the outfile is specified with the -o flag