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
Hello, I Am Working With A Pytorch Model, And I Would Like To Save It As An Outputmodel. I Read That Clearml Has Pytorch Auto-Logging Ebabled Automatically. My Question Is: What Path Should I Specify To Torch.Save() In Order For It To Be Uploaded Automati

Hello, I am working with a pytorch model, and I would like to save it as an Outputmodel. I read that Clearml has pytorch auto-logging ebabled automatically.
My question is: What path should I specify to torch.save() in order for it to be uploaded automatically as an artifact?

With tensorflow I save my model via a checkpoint to os.path.join(tempfile.gettempdir(), 'trained_models') but it doesn't seems to work with pytorch torch.save() :
FileNotFoundError: [Errno 2] No such file or directory: '/tmp/trained_models

  
  
Posted 2 years ago
Votes Newest

Answers 5


ClearML logs the paths as they are when torch.save() is run. Your problem currently is getting torch.save() to work, no?

  
  
Posted 2 years ago

The problem is with the path I am trying to save the model to.Therefore my question How do I extract paths inside ClearML?

  
  
Posted 2 years ago

As it seems torch.save() just saves to the disk, so it seems there is no need for (server) parent path, just the name of the file , in my case : ' http://my_model.pt ' .
Thanks for you help CostlyOstrich36

  
  
Posted 2 years ago

I think you need to specify some pythonic object to work with torch.save() - as it appears in their documentation:
https://pytorch.org/docs/stable/generated/torch.save.html

  
  
Posted 2 years ago

yes of course I specified it with :
torch.save(state, f_path)
where f_path = os.path.join(tempfile.gettempdir(), 'trained_models') + '\ http://my_model.pt '

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