Examples: query, "exact match", wildcard*, wild?ard, wild*rd
Fuzzy search: cake~ (finds cakes, bake)
Term boost: "red velvet"^4, chocolate^2
Field grouping: tags:(+work -"fun-stuff")
Escaping: Escape characters +-&|!(){}[]^"~*?:\ with \, e.g. \+
Range search: properties.timestamp:[1587729413488 TO *] (inclusive), properties.title:{A TO Z}(excluding A and Z)
Combinations: chocolate AND vanilla, chocolate OR vanilla, (chocolate OR vanilla) NOT "vanilla pudding"
Field search: properties.title:"The Title" AND text
Hi, There! After Running A Task In Clearml, How Can I Use The Api To Get The Best Performing Output Model For That Task? I Came Across This

Hi, there! After running a Task in ClearML, how can I use the API to get the best performing output model for that task?

I came across this Stack Overflow post , where the answer gets the best performing Task by using get_last_scalar_metrics() and then it gets the first model in that Task's model list.

However, in my case, I know the ID of the specific Task that I want to work with. But I want to get the local copy of the output model which had the lowest validation accuracy (for example). Anyone have any idea how to do that, please? Cheers!

Posted 2 months ago
Votes Newest

Answers 5

well that depends on you, what did you write there to know it is the best one ? file name ? added some metric ?

Posted 2 months ago

Thanks, @<1523701205467926528:profile|AgitatedDove14> ! And how do I identify which one is the best model in the ClearML web app?

Posted 2 months ago

Ah, got you 👍 I'll check that again - I think it's training accuracy

Posted 2 months ago

Hi @<1614069770586427392:profile|FlutteringFrog26>
So since you have the Task id. you do:

task = Task.get_task("task id here")

Then to get the models

models = task.models["output]

the models is a list And a dict, if you want the lats one you do last_model = models[-1] if you know the best model name you do model = models["best model"] (notice the model name is the exact one you see in the UI. Once you have the model object you can get a copy with model.get_local_copy()

Posted 2 months ago


Posted 2 months ago
5 Answers
2 months ago
2 months ago
Similar posts