DESCRIPTIONThis manual page documents briefly the pyevolve-graph command, a graphical plotting tool for programs based on Pyevolve. To use this program, you must have python-matplotlib and python-sqlite installed.
GENERAL OPTIONSPyevolve-pygraph supports the following options:
- -h, --help
- show a brief summary of available options and exit
- -f FILENAME, --file=FILENAME
- Database file to read from. default is ./pyevolve.db.
- -i IDENTIFY, --identify=IDENTIFY
- The identify of evolution.
- -o OUTFILE, --outfile=OUTFILE
- Write the graph image to a file. Don't specify an extension, just the filename. The file will be written in PNG format by default, but this can be changed using the --extension or -e options.
- -e EXTENSION, --extension=EXTENSION
- Graph image file format. Supported options (formats) are all those supported by python-matplotlib. Currently, these should be: EMF, EPS, PDF, PNG, PS, RAW, RGBA, SVG, SVGZ. Default is PNG.
- -g GENRANGE, --genrange=GENRANGE
- This is the generation range of the graph. For example, 1:30 would denote an interval between 1 and 30.
- -c COLORMAP, --colormap=COLORMAP
- Sets the Color Map for the graph types -8 and -9. Some options are: summer, bone, gray, hot, jet, cooper, spectral. The default is jet.
- Sets the 'Minimize' mode. Default is the Maximize mode. This option makes sense if you are minimizing your evaluation function.
- Write all standard graphs to file. Equivalent to specifying -1 -2 -3 -4 -5.
- Error bars graph (raw scores).
- Error bars graph (fitness scores).
- Max/min/avg/std. dev. graph (raw scores).
- Max/min/avg graph (fitness scores).
- Raw and Fitness min/max difference graph.
- Compare best raw score of two or more evolutions. You must specify the identify as a comma-separated list with the --identify or -i options. The maximum is 6 items.
- Compare best fitness score of two or more evolutions. You must specify the identify as a comma-separated list with the --identify or -i options. The maximum is 6 items.
- Show a heat map of population raw score distribution between generations.
- Show a heat map of population fitness score distribution between generations.
EXAMPLESThe following are some examples on how to use pyevolve-graph:
1. Output a graph for identify 'ex1' in PNG format (default) to file 'graph_ex1.png':
pyevolve-graph -i ex1 -1 -o graph_ex1
2. Output a graph for identify 'ex1' to the PDF file 'graph_ex1.pdf':
pyevolve-graph -i ex1 -1 -o graph_ex1 -e pdf
3. This command will plot the evolution of the generations between 10 and 20:
pyevolve-graph -i ex1 -1 -g 10:20
4. Use -m to indicate that you have minimized the evaulation function:
pyevolve_graph.py -i ex1 -1 -m
5. Specify an identify list for the comparison output options -6 and -7:
pyevolve_graph.py -i ex1_run1,ex1_run2,ex1_run3 -6
Fore more examples on how to use pyevolve-graph, including sample images, users are recommended to install the python-pyevolve-doc package.