Have you checked the issues for a similar suggestions?
Yes
How would you improve Rich?
I am really enjoying using rich and intend to continue using it whenever I am working with large datasets, but I think it would be extremely helpful to natively support the ability to have nested track statements. For example:
for i in track(range(10)):
for j in track(range(10)):
sleep(0.2)
The above example would render a live display with the overall progress like in the gif below but with only two progress bars and the second progress bar would reset to 0 with every iteration of i.
What problem does it solve for you?
I am currently using this to track the training of LLMs which is awesome, but it means processing batches of entire datasets. Processing each batch takes a fair amount of time, so it would be really nice to see how quickly processing is occurring for each batch without having to configure detailed statistics or a more complex configuration with a Live Display.
github-actions[bot]
commented
1 week ago
Have you checked the issues for a similar suggestions?
How would you improve Rich?
I am really enjoying using rich and intend to continue using it whenever I am working with large datasets, but I think it would be extremely helpful to natively support the ability to have nested
trackstatements. For example:The above example would render a live display with the overall progress like in the gif below but with only two progress bars and the second progress bar would reset to 0 with every iteration of i.
What problem does it solve for you?
I am currently using this to track the training of LLMs which is awesome, but it means processing batches of entire datasets. Processing each batch takes a fair amount of time, so it would be really nice to see how quickly processing is occurring for each batch without having to configure detailed statistics or a more complex configuration with a Live Display.
Related Topics
2272