Commit graph

4 commits

Author SHA1 Message Date
Nathan Binkert
c0a4836077 Major improvements in the graph output code. Mostly adding more
options, making existing options more visible and dealing with
holes in data better.

util/stats/barchart.py:
    - move the options for BarChart to a base class ChartOptions so
    they can be more easily set and copied.
    - add an option to set the chart size (so you can adjust the aspect ratio)
    - don't do the add_subplot thing, use add_axes directly so we can
    affect the size of the figure itself to make room for the legend
    - make the initial array bottom floating point so we don't lose precision
    - add an option to set the limits on the y axis
    - use a figure legend instead of an axes legend so we can put the legend
    outside of the actual chart.  Also add an option to set the fontsize of
    the legend.
    - initial hack at outputting csv files
util/stats/db.py:
    don't print out an error when the run is missing from the database
    just return None, the error will be print elsewhere.
util/stats/output.py:
    - make StatOutput derive from ChartOptions so that it's easier to
    set default chart options.
    - make the various output functions (graph, display, etc.) take the
    name of the data as a parameter instead of making it a parameter to
    __init__.  This allows me to create the StatOutput object with
    generic parameters while still being able to specialize the name
    after the fact
    - add support for graph_group and graph_bars to be applied to multiple
    configuration groups.  This results in a cross product of the groups
    to be generated and used.
    - flush the html file output as we go so that we can load the file
    while graphs are still being generated.
    - make the proxy a parameter to the graph function so the proper system's
    data can be graphed
    - for any groups or bars that are completely missing, remove them from
    the graph.  This way, if we decide not to do a set of runs, there won't
    be holes in the data.
    - output eps and ps by default in addition to the png.
util/stats/profile.py:
    - clean up the data structures that are used to store the function
    profile information and try our best to avoid keeping extra data
    around that isn't used.
    - make get() return None if a job is missing so we know it was
    missing rather than the all zeroes thing.
    - make the function profile categorization stuff total up to 100%
    - Fixup the x-axis and y-axis labels.
    - fix the dot file output stuff.
util/stats/stats.py:
    support the new options stuff for StatOutput

--HG--
extra : convert_revision : fae35df8c57a36257ea93bc3e0a0e617edc46bb7
2005-11-22 21:50:34 -05:00
Nathan Binkert
e00237e49e Major cleanup of the statistics handling code
util/stats/db.py:
    Build a result object as the result of a query operation so it is
    easier to populate and contains a bit more information than just
    a big dict.  Also change the next level data into a matrix instead
    of a dict of dicts.
    Move the "get" function into the Database object.  (The get function
    is used by the output parsing function as the interface for accessing
    backend storage, same interface for profile stuff.)
    Change the old get variable to the method variable, it describes how
    the get works, (whether using sum, stdev, etc.)
util/stats/display.py:
    Clean up the display functions, mostly formatting.
    Handle values the way they should be now.
util/stats/info.py:
    Totally re-work how values are accessed from their data store.
    Access individual values on demand instead of calculating everything
    and passing up a huge result from the bottom.
    This impacts the way that proxying works, and in general, everything
    is now esentially a proxy for the lower level database.  Provide new
    operators: unproxy, scalar, vector, value, values, total, and len which
    retrieve the proper result from the object they are called on.
    Move the ProxyGroup stuff (proxies of proxies!) here from the now gone
    proxy.py file and integrate the shared parts of the code.  The ProxyGroup
    stuff allows you to write formulas without specifying the statistics
    until evaluation time.

    Get rid of global variables!
util/stats/output.py:
    Move the dbinfo stuff into the Database itself.  Each source should
    have it's own get() function for accessing it's data.
    This get() function behaves a bit differently than before in that it
    can return vectors as well, deal with these vectors and with no result
    conditions better.
util/stats/stats.py:
    the info module no longer has the source global variable, just
    create the database source and pass it around as necessary

--HG--
extra : convert_revision : 8e5aa228e5d3ae8068ef9c40f65b3a2f9e7c0cff
2005-10-21 16:29:13 -04:00
Nathan Binkert
31d13e9a9b Improvements to the graphing output
util/stats/output.py:
    Create the graph directory if it doesn't exist
    Don't write out a graph if all of the jobs for that graph are missing

--HG--
extra : convert_revision : 7993baf1a4be33a062f86a4f09791f01eaafa43c
2005-10-18 15:01:51 -04:00
Nathan Binkert
02098f8e7b Totally re-work the way that jobfiles are done so there is more
information that can be used for other aspects of sending jobs.
New graphing output stuff with matplotlib.

util/pbs/job.py:
    Shuffle code around and create the JobDir class which encapsulates
    all of the functionality needed for making, organizing, and cleaning
    a job directory.
    Better status output
util/pbs/jobfile.py:
    Majory re-working of the jobfile code.
    A job file now consists of several objects that describe how
    jobs should be run, it includes information about checkpoints,
    and graphing.
util/pbs/send.py:
    use the new jobfile code.
    deal with the 15 character limit of pbs by truncating the name and
    using the raj hack.
util/stats/db.py:
    fix the __str__ function for nodes
    provide __getitem__ for the Database class
util/stats/stats.py:
    use the jobfile stuff to figure out what the proper naming
    and organziation of the graphs should be.
    move all output code to output.py, get rid of ploticus and use
    matplotlib

--HG--
rename : util/categories.py => util/stats/categories.py
extra : convert_revision : 0d793cbf6ad9492290e8ec875ce001c84095e1f7
2005-09-17 16:51:26 -04:00