Have a wrapper over Task to ensure S3 usage, tags, version number etc and project name can be skipped and it picks from the env var
Cool. Notice that when you clone the Task and the agents executes it, the project is already defined, so this env variable is meaningless, no ?
I get other things from the project like the dataset
Is there a good way to get the project of a task?
This worked well:
if project_name is None and Task.current_task() is not None: project_name = Task.current_task().get_project_name()
AgitatedDove14 - the actual replication failed. When we run a task by cloning and enqueueing, there is a current task even if I am yet to do a Task.init right?
AS in clone the task from the UI and run it
The first line of it is getting a dataset and it’s failing with no project name
if project_name is None and Task.current_task() is not None: project_name = Task.current_task().get_project_name()
This should have fixed it, no?
yeah i was trying in local and it worked as expected. But in local I was creating a Task first and then seeing if it’s able to get project name from it
Looks like Task.current_task() is indeed None in this case. Bit of log below where I print(Task.current_task()) as first step in the script
Environment setup completed successfully Starting Task Execution: None
TrickySheep9 can I see your code? Or at least the relevant part?
It’s essentially this now:
from clearml import Task print(Task.current_task())
And exact output after the package install and stuff:
Environment setup completed successfully Starting Task Execution: None
Thoughts AgitatedDove14 SuccessfulKoala55 ? Some help would be appreciated.
TrickySheep9 , this script would indeed not have a current task - such a task is created using Tasl.init()
- if you didn't call it, it won't be there
But that itself is running in a task right?
As in I am cloning a task and running it and in that takes without doing any Task.init i am trying to get the task that is running
Again, I'm not sure I understand your meaning. You're cloning a task and running it using an agent? That task has task = Task.init()
in it?
SuccessfulKoala55 AgitatedDove14 - this is the example I recreated in demo server - https://demoapp.trains.allegro.ai/projects/ce59885cb4df4cd1afa34aa5b593bb54/experiments/58e02b02d1714f58a590e2335a0f2ab7/execution?columns=selected&columns=type&columns=name&columns=tags&columns=status&columns=project.name&columns=users&columns=started&columns=last_update&columns=last_iteration&columns=parent.name&order=-last_update
Essentially - 1. run a task normally. 2. clone 3. edit to have only those two lines.
Question - since this is a task, why is Task.currnet_task() None?
Well, the SDK is only aware it's a task if you call Task.init()
. If not, you can fetch tasks using Task.get_task()
, but none of them will be the current.
Task.init()
will know if you're running locally (and so a new task should be created) or remotely (in case the current task is the one being executed remotely)
How can a task running like this know its own project name?
It’s a task, it’s running in context of a project, but I don’t have a way to get the project name
(I need this because I refer to datasets in the same project but without specifying the project name)