I have some large data sets that I need to evaluate graphically. I
find the graph navigation tools (left-click and drag to pan,
right-click and drag to zoom) in matplotlib absolutely superb and I'd
like to be able to better use them with the large data sets.
At the moment, I'm achieving this by plotting subsets of the data to
browse and then when I've completed the assessment, plotting the next
subset of data. This is fairly laborious...
What I'd really like to do is have a matplotlib plot figure appear on
the screen and as I browse it would pass the limits to some callback
function which would provide the data to plot within those limits.
The callback function could then manage the memory a bit better and
only read a subset of the data out of the file at any one time.
Ideally, the figure would hold the data it had while you browse and
(maybe) when you release the mouse button it would get the new data.
The data might not look perfect when you zoom in, but it would quickly
refresh and therefore improve. The callback function could provide a
bit of extra data either side of the visible area on the screen to
make panning work.
Is this a mammoth task involving writing a rather complex application
that reimplements a lot of the matplotlib plotting stuff, or is it
something that can be done relatively easily with the API?
Thanks in advance,