1

Given I have a really large task graph, my_delayed.visualize() either is impossible to generate or is too dense to be visually useful. If I have key for a particular task, can I specify a particular depth or x number of parents and children to visualize?

For example:

from dask import delayed
@delayed
def inc(x):
    return x + 1

def inc_list(x):
    return [inc(n) for n in x]

task = delayed(sum)(inc_list(inc_list([1,2,3])))
task.visualize(n_parents=1)

Would generate the below graph without the crossed out nodes.

enter image description here

postelrich
  • 3,274
  • 5
  • 38
  • 65

1 Answers1

0

Short answer

No, this is not currently supported

Longer answer

The task graph is just a dictionary, so you could manipulate it however you like to visualize sub-pieces of it. You might look at the dask.core functions get_dependencies, and get_deps to help manipulate the graph, as well as normal Python for loops.

MRocklin
  • 55,641
  • 23
  • 163
  • 235