Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Visualize task graphs using ipycytoscape #5822

Closed
bnaul opened this issue May 24, 2022 · 7 comments
Closed

Visualize task graphs using ipycytoscape #5822

bnaul opened this issue May 24, 2022 · 7 comments
Labels
enhancement An improvement of an existing feature

Comments

@bnaul
Copy link

bnaul commented May 24, 2022

Inspired by dask/dask#9091; the same reasons given there make sense to me here as well:

  1. The ability to run in pyodide without compiling graphviz to WASM (cf. Dask in pyodide dask/dask#9053)
  2. More interactivity, especially with different coloring schemes and layouts.
  3. The possibility to add richer tooltips/popups (cf. Clearer colourmap / labels when showing task order in visualize dask/dask#9069)

I actually don't know if Orion has a .visualize implementation yet, could be easier to add it there if starting from scratch?

@bnaul bnaul added the enhancement An improvement of an existing feature label May 24, 2022
@zanieb
Copy link
Contributor

zanieb commented May 24, 2022

This looks really cool!

We don't have any visualize implementation in v2 yet. Since the flow is discovered dynamically it'll be quite the undertaking :) We should be able to do local visualizations after a run easily though.

It's very unlikely that we'd take this on for v1.

@bnaul
Copy link
Author

bnaul commented May 24, 2022

Yeah it seems like a whole lot of work unless you can hijack the existing dask visualize functionality somehow, which seems doable to me...?

Since the flow is discovered dynamically

I know this is true in the general case but it seems like there are still cases where the flow is all declared statically, right? I know at least I can't easily break out of the DAG mindset so even my Orion flows still have all their tasks declared up front. So maybe "visualize whatever you already know about" would at least partly cover both cases?

@zanieb
Copy link
Contributor

zanieb commented May 24, 2022

Yeah it seems like a whole lot of work unless you can hijack the existing dask visualize functionality somehow, which seems doable to me...?

For v2? That could be interesting. We are planning on dropping dask as a top-level dependency though.

I know this is true in the general case but it seems like there are still cases where the flow is all declared statically, right?

Yep, that's the case we intend to address first. We'll have to run the flow function with all of the actual execution mocked to build the graph. We're also planning to explore walking the AST to capture dynamic flows.

@zareami10
Copy link

zareami10 commented Jun 20, 2022

I would also find this very useful.

Also see the Kedro case for a very well done example of pipeline visualization with interactive graphs.

@felix-ht
Copy link

felix-ht commented Oct 6, 2022

I love this idea as well.

It would even be better to have it integrated into orion (with https://js.cytoscape.org/) - i really dislike the current radar view.

@github-actions
Copy link
Contributor

This issue is stale because it has been open 30 days with no activity. To keep this issue open remove stale label or comment.

@github-actions
Copy link
Contributor

This issue was closed because it has been stale for 14 days with no activity. If this issue is important or you have more to add feel free to re-open it.

@github-actions github-actions bot closed this as not planned Won't fix, can't repro, duplicate, stale Apr 13, 2023
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
enhancement An improvement of an existing feature
Projects
None yet
Development

No branches or pull requests

4 participants