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
Hi All, I’Ve Been Running Into An Issue Lately When Using

Hi all, I’ve been running into an issue lately when using get_local_copy . The way I use it is: I use get_tasks to find the previous relevant experiments on my project, I then iterate over them and download the output models using get_local_copy .
This is the error I get:
File "...", line 69, in fetch_previous_models download_path = model.get_local_copy() File "...\site-packages\clearml\model.py", line 483, in get_local_copy return self.get_weights(raise_on_error=raise_on_error) File "...\site-packages\clearml\model.py", line 338, in get_weights return self._get_base_model().download_model_weights(raise_on_error=raise_on_error) File "...\site-packages\clearml\backend_interface\model.py", line 406, in download_model_weights if Path(dl_file).exists(): File "...\site-packages\pathlib2\__init__.py", line 1718, in exists self.stat() File "...\site-packages\pathlib2\__init__.py", line 1523, in stat return self._accessor.stat(self) File "...\site-packages\pathlib2\__init__.py", line 646, in wrapped return strfunc(str(pathobj), *args) OSError: [WinError 123] The filename, directory name, or volume label syntax is incorrect: "('<path_to_temp_folder>', '<model_name>'): <model_name>_0.pt"When I check the <path_to_temp_folder> , I noticed that the model available there was called <model_name>_<num_epochs_model_was_trained_for>.pt (it is looking for _ http://0.pt instead of the model from the last epoch)
I checked the web UI and the name is displayed correctly there (… _<num_epochs_model_was_trained_for>.pt )
Also, I set the experiment to keep a single checkpoint, so there is a single output model in the task ( {'input': [<clearml.model.Model object at 0x00000170A789C2B0>], 'output': [<clearml.model.Model object at 0x00000170A788C850>]} )
Any ideas what could be causing this?

(This is on a self-hosted server 1.4.0 running on Ubuntu - the script that causes that error is running on Windows using clearml 1.4.1)

  
  
Posted one year ago
Votes Newest

Answers 3


Hi SteepDeer88 , in general, this should not be affected by the server version, as the fileserver implementation rarely changes - I would expect this to be more related to the SDK version which is in charge of composing the filename.

  
  
Posted one year ago

Were the experiments generated with the same SDK version?

  
  
Posted one year ago

hi SteepDeer88
did you managed to get rid of your issue ?

  
  
Posted one year ago
532 Views
3 Answers
one year ago
one year ago
Tags
Similar posts