You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
Is your feature request related to a problem? Please describe.
Imagine I am experimenting with some things and I have a command line call that generates stuff, and it's slow: ! generate_stuff > stuff
Now imagine I've tested this a few times and so I have another line: ! rm stuff
And then I have another code block that does something. Like... do_stuff_with_stuff(stuff)
So, intending to click the first and third one, I brain fart and click all three. With the brain fart continuing, let's say it doesn't dawn on me that I did this until a few minutes in. So generate_stuff has been running for three or four minutes and I don't want to waste that time spent.
Colab is happily churning along with the "generate_stuff" block, and is cued up to then execute both of the other two blocks in turn. But the second block will make the third fail (because "stuff" is missing), and will also make the time spent on generate_stuff useless.
I can, of course, interrupt execution of all three in turn, but I'll have to start the WHOLE THING over.
I cannot say "wait, belay execution of number 2, I changed my mind!"
What I would like to see happen: Stop button appears immediately after click of play button, with the waiting spinner until execution started. If clicked before cell execution has started, button will remove cell from execution queue and restore play button without any spinners.
As I see it, the only other option is cancelling everything, including the one you want to keep running, to get Colab to execute the later blocks allowing them to be cancelled (if it can even be clicked that fast)
The text was updated successfully, but these errors were encountered:
This would be a very helpful feature (I came here to see if there was a way to do this in colab). Here's a work around you can use however that is somewhat inconvenient:
You can copy the contents of the cell you don't want to run, delete it and place them in a new cell. That way you still have the cells, and the deleted cells are removed from the execution queue
Is your feature request related to a problem? Please describe.
Imagine I am experimenting with some things and I have a command line call that generates stuff, and it's slow:
! generate_stuff > stuff
Now imagine I've tested this a few times and so I have another line:
! rm stuff
And then I have another code block that does something. Like...
do_stuff_with_stuff(stuff)
So, intending to click the first and third one, I brain fart and click all three. With the brain fart continuing, let's say it doesn't dawn on me that I did this until a few minutes in. So generate_stuff has been running for three or four minutes and I don't want to waste that time spent.
Colab is happily churning along with the "generate_stuff" block, and is cued up to then execute both of the other two blocks in turn. But the second block will make the third fail (because "stuff" is missing), and will also make the time spent on generate_stuff useless.
I can, of course, interrupt execution of all three in turn, but I'll have to start the WHOLE THING over.
I cannot say "wait, belay execution of number 2, I changed my mind!"
What I would like to see happen: Stop button appears immediately after click of play button, with the waiting spinner until execution started. If clicked before cell execution has started, button will remove cell from execution queue and restore play button without any spinners.
As I see it, the only other option is cancelling everything, including the one you want to keep running, to get Colab to execute the later blocks allowing them to be cancelled (if it can even be clicked that fast)
The text was updated successfully, but these errors were encountered: