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
Answered
Is It Possible To Link Independent Training Experiments.. For Example.. I Have An Ensemble Of 2 Models (A & B) Each Models Are Trained Under Their Own Training Task In Trains Now I Will Run Another Script Which Will Use These Models To Create An Ensemble

is it possible to link independent training experiments.. for example.. i have an ensemble of 2 models (A & B) each models are trained under their own training task in Trains now i will run another script which will use these models to create an ensemble model.. i can have it registered under Trains as another run ... but wondering if i can link from ensemble model as which underlying training task run created the models used in this ensemble.. a bit like what you have currently for models. As from any given model, i can go back to experiment which created it.. here i am looking for the same but for ensemble model which will have 1 to N relationship where as current model as 1 to 1 relationship with experiments

  
  
Posted 4 years ago
Votes Newest

Answers 7


while you guys gonna work on it.. just a small feature addition to it.. it would be cool to have a DAG figure which shows how models are linked under this task and ability to just click a circle in that DAG figure to navigate to given task... i think it will be very useful UX 🙂

  
  
Posted 4 years ago

look forward to the new job workflow part in 0.16 then 🙂

  
  
Posted 4 years ago

Month +- 🙂

  
  
Posted 4 years ago

yeah that would solve it i think.. so what is the normal cadence for release.. every month or quarter ?

  
  
Posted 4 years ago

Hmm I see what you mean. It is on the roadmap (ETA the next version 0.17, 0.16 is due in a week or so) to add multiple models per Task so it is easier to see the connections in the UI. I'm assuming this will solve the problem?

  
  
Posted 4 years ago

i mean linking more in UI.. as when i go to model detail page, i can see that a given experiment created this model and click on that to see its detail... so something similar to that for ensemble models

  
  
Posted 4 years ago

Hi PompousParrot44
So do you mean something like:
` task_model_a = Task.get('id_a')
task_model_b = Task.get('id_b')

model_a_file = task_model_a.models['output][-1].get_local_copy()

model_b_file = task_model_b.models['output][-1].get_local_copy() `

  
  
Posted 4 years ago