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
Hey! Did Anyone Try Hpo On Yolov5 Model According To The Following Tutorial:

Hey! Did anyone try HPO on yolov5 model according to the following tutorial: https://github.com/ultralytics/yolov5/tree/master/utils/loggers/clearml in ClearML Hosted Service?

I am getting some weird behavior (or maybe I am doing something wrong), but firstly wanted to check weather the HPO (according to the tutorial) is even available for free accounts?

  
  
Posted 2 years ago
Votes Newest

Answers 5


CheekyFox58 what do you have in the plots Tab?

  
  
Posted 2 years ago

AgitatedDove14 It works fine now. In that reported case nothing has been plotted. I've been making some conflict situations with tasks, but then I saw that the task is already being created in hpo.py script, so I removed my previously made one

  
  
Posted 2 years ago

Hi CheekyFox58
If you are running the HPO+training on your own machine, it should work just fine in the Free tier

The HPO with the UI and everything, is designed to run the actual training on remote machines, and I think this makes it a Pro feature.

  
  
Posted 2 years ago

Hey CheekyFox58 like Martin said, it should at least work locally. If not, can you give us some more details on what exactly the werid behaviour is?

  
  
Posted 2 years ago

AgitatedDove14 ExasperatedCrab78 Yes, I trained the initial model and tracked it in my clearml project. When it comes to the HPO, I just run the hpo.py script, changed the id, and got such log, shown in the image (there are all failed trials)

Also, according to the tutorial, the Optimization Objective should be plotted, but I don't get anything in the plot section, and not sure weather the optimization is completed successfully, because I am not sure if the idea is that model will be automatically updated after finished HPO (and where) or somewhere will be written optimal hyperparameters, so I have to train the model again based on new hparams?

  
  
Posted 2 years ago
888 Views
5 Answers
2 years ago
one year ago
Tags