ClearML/clearml-docs
1
0
Fork 0
mirror of https://github.com/clearml/clearml-docs synced 2025-06-26 18:17:44 +00:00
Code Issues Actions Packages Projects Releases Wiki Activity

Edit transformer integration

Browse Source
This commit is contained in:
revital 2025-03-05 14:12:32 +02:00
parent af33fc1477
commit 658991c644
1 changed files with 20 additions and 8 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

28
docs/integrations/transformers.md
View File

@ -76,19 +76,31 @@ cloud of your choice (AWS, GCP, Azure) and automatically deploy ClearML agents:
and shuts down instances as needed, according to a resource budget that you set.
### Cloning, Editing, and Enqueuing
### Reproducing Tasks
![Cloning, editing, enqueuing gif](../img/gif/integrations_yolov5.gif#light-mode-only)
![Cloning, editing, enqueuing gif](../img/gif/integrations_yolov5_dark.gif#dark-mode-only)
Use ClearML's web interface to edit task details, like configuration parameters or input models, then execute the task
with the new configuration on a remote machine:
Use ClearML's web interface to reproduce tasks and edit their details, like hyperparameters or input models, then execute the task
with the new configuration on a remote machine.
When ClearML is integrated into a script, it captures and stores configurations, such as hyperparameters
and model settings. Typically, when a ClearML Agent re-executes a task, the agent users the values stored in the task,
overriding any execution-time settings. ignores configuration in the execution parameters, and
uses the values stored on the task. If any of these configurations are later modified in the UI, ClearML prioritizes the
stored values and overrides any execution-time settings.
However, for tasks using Transformers, an exception to this behavior is sometimes needed—ensuring that
execution-time parameters (such as environment variables) take precedence instead
of the values stored in the task. For ClearML Agent to run as it typically does, using values stored on the task
and allowing UI overrides, you need to set the `_ignore_hparams_ui_overrides_` and `_ignore_model_config_ui_overrides_` variables
to `False` in the task's **CONFIGURATION > HYPERPARAMETERS > Transformers** section..
To reproduce a ClearML task with Hugging Face Transformers while allowing UI overrides:
1. Clone the task
1. Edit the hyperparameters and/or other details
* Ensure that `_ignore_hparams_ui_overrides` is set to `False` in the **CONFIGURATION > HYPERPARAMETERS > Transformers** section.
This allows the task to use the new hyperparameter values during execution.
* Ensure that `_ignore_model_config_ui_overrides` is set to `False` in the **CONFIGURATION > CONFIGURATION OBJECTS > Model Configuration**
section. This allows the task to use the new model configuration values during execution.
1. Optionally, edit the hyperparameters and/or other details. Set both `_ignore_hparams_ui_overrides_` and `_ignore_model_config_ui_overrides_`
to `False` in the **CONFIGURATION > HYPERPARAMETERS > Transformers** section. This allows the task to use the new hyperparameter and model
configuration values respectively during execution.
1. Enqueue the task
The ClearML Agent executing the task will use the new values to [override any hard coded values](../clearml_agent.md).