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, I'M Trying To Access/Use Experiment'S Model+Data+Params My Model And Data Are Stored In S3 And I'M Not Sure What Is The Practice Of Getting Them Reddy To Use. When I Use

Hi, I'm trying to access/use experiment's model+data+params
My model and data are stored in s3 and I'm not sure what is the practice of getting them reddy to use.
When I use local_csv = a_task.artifacts['train_data'].get_local_copy()
and cls = a_task.models['output'] I don't get the actual data and model.
The params dict I can get using get_parameters() but for some reason I get in my dict key string an addition of 'general/' followed by my param string 'max_depth'
I wish to understand what is the easiest way to reproduce my experiment using clearml SDK and have it ready to run again from my python interperter.

Posted one year ago
Votes Newest

Answers 4

IIRC, get_local_copy() downloads a local copy and returns the path to the downloaded file. So you might be interested in e.g.
local_csv = pd.read_csv(a_task.artifacts['train_data'].get_local_copy())
With the models, you're looking for get_weights() . It acts the same as get_local_copy() , so it returns a path.

EDIT: I think also get_local_copy() for a model should work 👍

Posted one year ago

HappyDove3 , Hi 🙂

I'm not sure I understand what you want to do. You want to get a local copy of your model file or you just want to play with the experiment and reproduce it 1:1 ?

Posted one year ago

I want to play with the experiment and reproduce it 1:1.
I managed to it with the advice of UnevenDolphin73 is there a different way to do it?

Posted one year ago

You can clone it via the UI, enqueue it to a queue that has a worker running against that queue. You should get a perfect 1:1 reproduction

Posted one year ago
4 Answers
one year ago
4 months ago