trappy.wa.results module

Parse the results from a Workload Automation run and show it in a “pretty” table

class trappy.wa.results.Result(*args, **kwargs)

Bases: pandas.core.frame.DataFrame

A DataFrame-like class for storing benchmark results

enlarge_axis(data)

Make sure that the axis don’t clobber some of the data

init_fig()

plot_results()

plot_results_benchmark(benchmark, title=None)

Plot the results of the execution of a given benchmark

A title is added to the plot if title is not supplied

trappy.wa.results.combine_results(data)

Combine two DataFrame results into one

The data should be an array of results like the ones returned by get_results() or have the same structure. The returned DataFrame has two column indexes. The first one is the benchmark and the second one is the key for the result.

trappy.wa.results.get_results(path='.', name=None)

Return a pd.DataFrame with the results

The DataFrame’s rows are the scores. The first column is the benchmark name and the second the id within it. For benchmarks that have a score result, that’s what’s used. For benchmarks with FPS_* result, that’s the score. E.g. glbenchmarks “score” is it’s fps.

An optional name argument can be passed. If supplied, it overrides the name in the results file.

trappy.wa.results.get_run_number(metric)