On 12/03/2012 10:21 PM, alepulver wrote:
On 12/03/2012 01:24 PM, Tiago Peixoto [via Main discussion list for the graph-tool project] wrote:
On 12/01/2012 12:54 AM, alepulver wrote:
Does it also happen to you: pos = gt.random_layout(polysemy) print "here" gt.graph_draw(polysemy, pos=pos, output='ejercicio_5_gv.png') This took only about 2 seconds on my machine... I have no idea how it could take two hours.
I made a mistake. My program was also using "output_size=(10000, 10000)". With the default value it takes a few seconds.
But is it supposed to take so much when increasing output size? The most expensive part should be calculating node positions, as you said.
If you use the PNG format it has to construct a huge 10000x10000 matrix, and paint he pixels inside the vertices, edges, etc. Everything should scale roughly with the area of the figure. This does not happen of course with vector formats, such as SVG and PDF, where the size of the figure is just a number. Cheers, Tiago -- Tiago de Paula Peixoto <tiago@skewed.de>