Can not compare TimeMetric across runs

[bl1zzardx]

Hi,

it is not possible to compare two benchmark runs based on TimeMetric, memory and cpu metrics work fine though. As there is no interval for the TimeMetric, it results in a simple bar chart. These are not plotted side-by-side, as opposed to the overlaying of line charts of cpu and memory.

> umlaut-cli benchmark.db
? Please select one or more uuids.  done (2 selections)
? Please select measurement types corresponding to uuids.  done (2 selections)
? Please select descriptions corresponding to uuids and types.  done (2 selections)
Traceback (most recent call last):
  File "/home/laberand/.pyenv/versions/3.9.1/envs/pyenv3.9/bin/umlaut-cli", line 33, in <module>
    sys.exit(load_entry_point('umlaut', 'console_scripts', 'umlaut-cli')())
  File "/home/laberand/daphne_work/End-to-end-ML-System-Benchmark/umlaut/umlaut/visualization_cli.py", line 164, in main
    figs.append(visualizer.plot())
  File "/home/laberand/daphne_work/End-to-end-ML-System-Benchmark/umlaut/umlaut/visualization.py", line 24, in plot
    return self.plot_with_matplotlib()
  File "/home/laberand/daphne_work/End-to-end-ML-System-Benchmark/umlaut/umlaut/visualization.py", line 240, in plot_with_matplotlib
    self.df["measurement_data"] = float(self.df["measurement_data"])
  File "/home/laberand/.pyenv/versions/3.9.1/envs/pyenv3.9/lib/python3.9/site-packages/pandas/core/series.py", line 248, in wrapper
    raise TypeError(f"cannot convert the series to {converter}")
TypeError: cannot convert the series to <class 'float'>

Is there a way to plot TimeMetric for multiple benchmarks together?