Right now what graphs to create for intra- and inter-batch experiments are
hardcoded into .py files as dictionaries. This should be changed so that the
dictionaries are created from .yaml files which are read in during stage4 from
config/ (or specified via command line).
Similarly, in all the non-linegraphs in which I need to read from a specific
.csv file and access specific columns (e.g. when calculating scalability),
entries should also be added in .yaml file(s) to specify those.
By doing both of these, sierra should then be capable of more general purpose
graph generation from .csv files, moving it closer to being code that could be
profitably used by other researchers.
Right now what graphs to create for intra- and inter-batch experiments are hardcoded into .py files as dictionaries. This should be changed so that the dictionaries are created from .yaml files which are read in during stage4 from
config/(or specified via command line).Similarly, in all the non-linegraphs in which I need to read from a specific .csv file and access specific columns (e.g. when calculating scalability), entries should also be added in .yaml file(s) to specify those.
By doing both of these, sierra should then be capable of more general purpose graph generation from .csv files, moving it closer to being code that could be profitably used by other researchers.