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

clip generator with mutlithreading progress visualization support #134

Open
linzuzeng opened this issue Mar 12, 2020 · 7 comments
Open

clip generator with mutlithreading progress visualization support #134

linzuzeng opened this issue Mar 12, 2020 · 7 comments
Assignees
Labels
GUI Features related to the GUI help wanted Extra attention is needed recipe issue related to a built-in recipe

Comments

@linzuzeng
Copy link
Member

Currently we show percentage of file reading in clip generators, but it is super noisy and doesn't make much sense when used together with muli-threading.

I would suggest to have a block diagram to visualize the analysis progress. I have an example from AiraNg
189869

We just need to send the starting recordid and the ending record id of a file to the Web GUI, and ask Web GUI to put green on to the corresponding block, indicating that block of file is read already.

@linzuzeng linzuzeng added recipe issue related to a built-in recipe GUI Features related to the GUI labels Mar 12, 2020
@linzuzeng linzuzeng added the help wanted Extra attention is needed label Mar 12, 2020
@linzuzeng
Copy link
Member Author

We could have a special logger for progress visulization, or add some javascript to parse the progress in the GUI side.
https://github.com/timetag/ETA/blob/master/etabackend/clip.py#L162

@gyger
Copy link
Contributor

gyger commented Mar 18, 2021

There is a blog entry that has a similar structure: https://towardsdatascience.com/stunning-tables-using-bokeh-and-svg-e612a6f1c541

@linzuzeng
Copy link
Member Author

COOL!
Can bokeh render to a given div or we use iframe with a url pointing to a bokeh dashboard?

@Cyberfly100
Copy link
Contributor

Cyberfly100 commented Mar 19, 2021 via email

@linzuzeng
Copy link
Member Author

Great!
Now I realize maybe it's better to render the progress on the display panel (which is already bokeh) instead of embedding it on the main GUI.
It will allow users to see the progress on file with the the growing histograms on the same screen during an analysis. This may be handy for diagnosing some problem.
Also this may save one server for bokeh and probably prevents the headaches of running two bokeh dashboard simultaneously :)

@linzuzeng
Copy link
Member Author

The only drawback is that the progress is not enabled by default for every analysis and the user need to manually write progress bar rendering in the display panel Python script.
But I think it's fine as the users don't always need progress bar (progress bar for realtime is meaningless ; progress bars for batch process in parallel requires extra rendering and typesetting ) and we can put something in eta.tk to minimize users' efforts.

@Cyberfly100
Copy link
Contributor

Cyberfly100 commented Mar 20, 2021 via email

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
GUI Features related to the GUI help wanted Extra attention is needed recipe issue related to a built-in recipe
Projects
None yet
Development

No branches or pull requests

3 participants