Pipeline additions

All the additions can run after any of the steps and will run in parallel to the next step. They are run the same way as the steps.

Creating plots

Plots can be created after each step with the data available from the previous step. Hence, each corresponding csv file is called step_?_plots.csv, where the question mark will be the number of the step the plots take the final PSyGrid object from. All the csv files for plotting have the same structure:

path_to_grid,grid_type,quantities_to_plot,path_to_plot,plot_extension

Beside the grid, it states the type and takes a list of quantities to plot. All final quantities supported for a 2D plot, as third dimension can be specified. Additionally, you can put a LOG10_ in front of each of them to switch on plotting in log-scale. Beside that there are predefined plots. Finally, the path to the directory, where the plots should get stored, and the extension of the image files (those need to be valid extension for mathplotlib) are given. There is one additional extension multipage-pdf, which will create a PDF, where several plots are stored as pages in a single PDF.

Basic predefined plots

quantities_to_plot

‘term_flag’

‘zvar’

‘zmin’

‘zmax’

‘zlog’

‘combined_TF12’

‘combined_TF12’

None

None

None

False

‘termination_flag_1’

‘termination_flag_1’

‘lg_mtransfer_rate’

-8

-1

False

‘termination_flag_2’

‘termination_flag_2’

None

None

None

False

‘termination_flag_3’

‘termination_flag_3’

None

None

None

False

‘termination_flag_4’

‘termination_flag_4’

None

None

None

False

‘rl_relative_overflow_1’

‘debug’

‘rl_relative_overflow_1’

-0.5

0.5

False

‘rl_relative_overflow_2’

‘debug’

‘rl_relative_overflow_2’

-0.5

0.5

False

‘lg_mtransfer_rate’

‘debug’

‘lg_mtransfer_rate’

-8

-1

False

After Step3: calculating extra values from detailed data, the supernova model quantities get available, too.

Supernova predefined plots

quantities_to_plot

‘term_flag’

‘zvar’

‘zmin’

‘zmax’

‘zlog’

‘S1_MODEL??_CO_type’

‘S1_MODEL01_CO_type’

‘S1_MODEL??_CO_type’

None

None

False

‘S1_MODEL??_SN_type’

‘S1_MODEL01_SN_type’

‘S1_MODEL??_SN_type’

None

None

False

‘S1_MODEL??_mass’

‘termination_flag_1’

‘S1_MODEL??_mass’

True

‘S1_MODEL??_spin’

‘termination_flag_1’

‘S1_MODEL??_spin’

False

‘S1_MODEL??_m_disk_radiated’

‘termination_flag_1’

‘S1_MODEL??_m_disk_radiated’

False

After Step4: training of the interpolators, the interpolators can be used.

Predefined plots with variable QUANTITY

quantities_to_plot

‘term_flag’

‘zvar’

‘zmin’

‘zmax’

‘zlog’

QUANTITY

‘termination_flag_1’

QUANTITY

None

None

False

‘LOG10_ QUANTITY

‘termination_flag_1’

QUANTITY

None

None

True

‘INTERP_ERROR_ QUANTITY

None

QUANTITY

0

0.1

False

Doing checks

After each step one can perform some checks with the data from that step.

Currently supported checks

Check

Description

‘failure_rate’

calculates the failure rate of the grid

‘CO_type’

gets counts of compact object types

‘SN_type’

gets counts of supernova types