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
Hey All, I Started To Use Clearml Couple Of Days Ago And It’S Great So Far. I’M Working On A Project That Uses Pytorch Lightning. I Saw The Code Snippet About How To Integrate Clearml With Pytorch Lightning, However I’M Not Sure I Understand How The Integ

Hey all, I started to use ClearML couple of days ago and it’s great so far.
I’m working on a project that uses Pytorch Lightning. I saw the code snippet about how to integrate ClearML with Pytorch Lightning, however I’m not sure I understand how the integration works. This is the code snippet I found:

In the code snippet, we don’t seem to supply PL a logger or something. My question is, how PL knows to use ClearMLLogger , instead of the default one?

For example, when I do self.log inside my LightningModule - how does it know to use ClearML here?

Posted 2 years ago
Votes Newest

Answers 6


Posted 2 years ago

New to lightning too, but I'm suspecting that since your args don't mention a specific logger, the pl trainer will instantiate the default one. Excerpt from the trainer docstring:
logger: Logger (or iterable collection of loggers) for experiment tracking. A ``True`` value uses the default ``TensorBoardLogger``. ``False`` will disable logging. If multiple loggers are provided and thesave_dirproperty of that logger is not set, local files (checkpoints, profiler traces, etc.) are saved in ``default_root_dir`` rather than in the ``log_dir`` of any of the individual loggers.

Posted 2 years ago

Hi MinuteCamel2 ,

It’s all part of the auto-magic 🪄 after Task.init , the pytorch lightning functions are bind with ClearML, so it will report all the outputs to your ClearML app 🙂

Posted 2 years ago

TimelyPenguin76 Okay, nice! Will try that. Thanks 🙂

Posted 2 years ago

You can get your task logger any time with get_logger() function, like in this example:

Posted 2 years ago

TimelyPenguin76 I see. But when I try to fetch the underline logger from Pytorch Lightning, I see its type is SummaryWriter . Do you know how can I “extract” ClearMLLogger from PL, after a task is initiated?

Posted 2 years ago
6 Answers
2 years ago
one year ago
Similar posts