Friendlier output for sweep_summary.txt
I suggest a more parser-friendly format for the sweep_summary.txt
.
The current output is cumbersome to parse with automated tools, due to a variable number of header lines to ignore indicated by a #
and the fact that the actual column header is also prefixed with a #
.
To alleviate this issue, I suggest to split it into two files:
sweep_header.csv
:
Date Tue Oct 16 2018
Time 13:59:01
Jobs 6
Params 2
P1 MorpheusModel/CellTypes/CellType[name=cells]/System/Constant[symbol=n]/@value
P2 MorpheusModel/CellTypes/CellType[name=cells]/System/Constant[symbol=K]/@value
sweep_data.csv
:
Folder P1 P2
Example-CellCycle_sweep_82/Example-CellCycle_8481 3 0.2
Example-CellCycle_sweep_82/Example-CellCycle_8482 3 0.5
Example-CellCycle_sweep_82/Example-CellCycle_8483 4 0.2
Example-CellCycle_sweep_82/Example-CellCycle_8484 4 0.5
Example-CellCycle_sweep_82/Example-CellCycle_8485 5 0.2
Example-CellCycle_sweep_82/Example-CellCycle_8486 5 0.5
Further improvements could be:
- To have a column in the QTreeView with a convenience name (e.g.
n
orK
) - Skip the root folder in the
sweep_data.csv