Storage¶
ClearML automatically logs the storage of models and debug samples locally. You can also configure ClearML for the storage you use.
If your script loads other data, that data must be accessible from the machine where the script runs. If, for example, you develop a script on your local machine, and then reproduce it by executing it on a remote machine, the data must be accessible from the remote machine.
The supported storage includes:
local and shared folders
S3 buckets
Google Cloud Storage
Azure Storage
http(s)
Configuring ClearML for artifact storage¶
Configure ClearML for model checkpoint (snapshot) storage in any of the following ways:
In the configuration file, set default_output_uri.
In code, when initializing a Task, use the
output_uri
parameter.In the ClearML Web UI, when modifying an experiment.
Configuring ClearML for debug samples storage¶
Configure ClearML for debug sample storage by calling the Logger.set_default_upload_destination before calling explicit reporting methods (see the explicit report examples).
Next Steps
Learn about workers and queues.