Small edits

This commit is contained in:
revital 2025-04-15 08:24:47 +03:00
parent 586d0a7362
commit bd06c98927
10 changed files with 16 additions and 16 deletions

View File

@ -49,11 +49,11 @@ optimization.
documentation.
* **BOHB** - [`automation.hpbandster.OptimizerBOHB`](../references/sdk/hpo_hpbandster_bandster_optimizerbohb.md). BOHB performs robust and efficient hyperparameter optimization
at scale by combining the speed of Hyperband searches with the guidance and guarantees of convergence of Bayesian Optimization.
For more information about HpBandSter BOHB, see the [HpBandSter](../https://automl.github.io/HpBandSter/build/html/index.html)
For more information about HpBandSter BOHB, see the [HpBandSter](https://automl.github.io/HpBandSter/build/html/index.html)
documentation and a [code example](../guides/frameworks/pytorch/notebooks/image/hyperparameter_search.md).
* **Random** uniform sampling of hyperparameters - [`automation.RandomSearch`](../references/sdk/hpo_optimization_randomsearch.md).
* **Full grid** sampling strategy of every hyperparameter combination - [`automation.GridSearch`](../references/sdk/hpo_optimization_gridsearch.md).
* **Custom** - [`automation.optimization.SearchStrategy`](../https://github.com/clearml/clearml/blob/master/clearml/automation/optimization.py#L268) - Use a custom class and inherit from the ClearML automation base strategy class.
* **Custom** - [`automation.optimization.SearchStrategy`](https://github.com/clearml/clearml/blob/master/clearml/automation/optimization.py#L295) - Use a custom class and inherit from the ClearML automation base strategy class.
## Defining a Hyperparameter Optimization Search Example
@ -131,7 +131,7 @@ optimization.
## Optimizer Execution Options
The `HyperParameterOptimizer` provides options to launch the optimization tasks locally or through a ClearML [queue](../fundamentals/agents_and_queues.md#what-is-a-queue).
Start a `HyperParameterOptimizer` instance using either [`HyperParameterOptimizer.start()`](../references/sdk/hpo_optimization_hyperparameteroptimizer.md#start)
or [`HyperParameterOptimizer.start_locally()`](references/sdk/hpo_optimization_hyperparameteroptimizer.md#start_locally).
or [`HyperParameterOptimizer.start_locally()`](../references/sdk/hpo_optimization_hyperparameteroptimizer.md#start_locally).
Both methods run the optimizer controller locally. `start()` launches the base task clones through a queue
specified when instantiating the controller, while `start_locally()` runs the tasks locally.

View File

@ -127,7 +127,7 @@ clearmlApplications:
```
The credentials specified in `<SUPERVISOR_USER_KEY>` and `<SUPERVISOR_USER_SECRET>` can be used to login as the
The credentials specified in `<SUPERVISOR_USER_KEY>` and `<SUPERVISOR_USER_SECRET>` can be used to log in as the
supervisor user from the ClearML Web UI accessible using the URL `app.<BASE_DOMAIN>`.

View File

@ -402,7 +402,7 @@ appear in a tooltip and its usage percentage will appear at the center of the ch
![Version label statistics](../../img/hyperdatasets/dataset_version_statistics.png#light-mode-only)
![Version label statistics](../../img/hyperdatasets/dataset_version_statistics_dark.png#dark-mode-only)
## Metadata
## Metadata
The **Metadata** tab presents any additional metadata that has been attached to the dataset version.
**To edit a version's metadata,**
@ -414,7 +414,7 @@ The **Metadata** tab presents any additional metadata that has been attached to
![Version metadata](../../img/hyperdatasets/dataset_version_metadata.png#light-mode-only)
![Version metadata](../../img/hyperdatasets/dataset_version_metadata_dark.png#dark-mode-only)
## Info
## Info
The **Info** tab presents a version's general information:
* Version ID

View File

@ -6,7 +6,7 @@ title: Version 1.2
**New Features and Improvements**
* Improve GPU Performance, 50%-300% improvement over vanilla Triton
* Improve performance on CPU, optimize uvloop + multi-processing
* Improve performance on CPU, optimize uvloop + multiprocessing
* Add Hugging Face Transformer example
* Add binary input support ([ClearML Serving PR #37](https://github.com/clearml/clearml-serving/pull/37) )

View File

@ -48,11 +48,11 @@ title: Version 1.0
- Fix `default_output_uri` for Dataset creation ([ClearML GitHub issue 371](https://github.com/clearml/clearml/issues/371))
- Fix `clearml-task` failing without a docker script ([ClearML GitHub issue 378](https://github.com/clearml/clearml/issues/378))
- Fix PyTorch DDP sub-process spawn multi-process
- Fix PyTorch DDP sub-process spawn multiprocess
- Fix `Task.execute_remotely()` on created Task (not initialized Task)
- Fix auto scaler custom bash script should be called last before starting agent
- Fix auto scaler spins too many instances at once then kills the idle ones (spin time is longer than poll time)
- Fix multi-process spawn context using `ProcessFork` kills sub-process before parent process ends
- Fix multiprocess spawn context using `ProcessFork` kills sub-process before parent process ends
### ClearML 1.0.3

View File

@ -74,7 +74,7 @@ title: Version 1.1
* Fix `matplotlib` auto-magic detect bar graph series name ([ClearML GitHub issue #518](https://github.com/clearml/clearml/issues/518))
* Fix path limitation on storage services (posix, object storage) when storing target artifacts by limiting length of
project name (full path) and task name used for object path ([ClearML GitHub issue #516](https://github.com/clearml/clearml/issues/516))
* Fix multi-processing context block catching exception
* Fix multiprocessing context block catching exception
* Fix Google Cloud Storage with no default project causes a crash
* Fix main process's reporting subprocess lost, switch back to thread mode
* Fix forked `StorageHelper` should use its own `ThreadExecuter`
@ -239,7 +239,7 @@ title: Version 1.1
- Fix Python2 compatibility
- Fix *clearml-task* exit with error when failing to verify `output_uri` (output warning instead)
- Fix unsafe Google Storage delete object
- Fix multi-process spawning wait-for-uploads can create a deadlock in very rare cases
- Fix multiprocess spawning wait-for-uploads can create a deadlock in very rare cases
- Fix `task.set_parent()` fails when passing Task object
- Fix `PipelineController` skipping queued Tasks
- Remove `humanfriendly` dependency (unused)

View File

@ -49,7 +49,7 @@ title: Version 1.13
**New Features**
* Add support for recursive list, dict, and tuple ref parsing for `PipelineController.add step()` parameter overrides ([ClearML GitHub issue #1089](https://github.com/clearml/clearml/issues/1089))
* Update PyNVML to the latest NVIDIA version for better GPU monitoring
* Add `force_single_script_file` argument to `Task.create()` to avoid git repository auto detection
* Add `force_single_script_file` argument to `Task.create()` to avoid git repository auto-detection
* Use `os.register_at_fork` instead of monkey patching fork for `python>3.6`
* Add support to programmatically archive and unarchive models from the model registry using the `Model.archive()` and
`Model.unarchive()` methods ([ClearML GitHub issue #1096](https://github.com/clearml/clearml/issues/1096))

View File

@ -26,7 +26,7 @@ Manage administrator vaults in ClearML UI, under **SETTINGS \> Administrator Vau
* **Create** vaults
* **Edit** vault contents.
For more information see [Administrator Vaults](../webapp/settings/webapp_settings_admin_vaults) settings page.
For more information see [Administrator Vaults](../webapp/settings/webapp_settings_admin_vaults.md) settings page.
## Managing Vaults via the REST API
@ -116,7 +116,7 @@ curl -s -XPUT $CLEARML_SERVER/users.add_or_update_vault \
```
In the example above, the group ID `"30795571-a470-4717-a80d-e8705fc776bf"` refers to the Users group (to which all users
belong). You can define other groups in the [Users & Groups](../webapp/settings/webapp_settings_users#user-groups)
belong). You can define other groups in the [Users & Groups](../webapp/settings/webapp_settings_users.md#user-groups)
settings page and create vaults for these specific groups.
The command returns the ID of the newly created vault. For example:

View File

@ -148,7 +148,7 @@ resources from the H100 pool.
All available resources having been assigned - 2 jobs of each team will remain pending until some of the currently
running jobs finish and resources become available.
## Applying Resource Configuration
## Applying Resource Configuration
Administrators can globally activate/deactivate resource policy management. To enable the currently provisioned
configuration, click on the `Enable resource management` toggle. Enabling resource management will service the policy
queues according to the provisioned resource profile and pool assignments. Disabling the resource management will stop

View File

@ -89,7 +89,7 @@ and [datasets](datasets/webapp_dataset_page.md)) provide column filters to easil
Click the help menu button <img src="/docs/latest/icons/ico-help-outlined.svg" alt="Help menu" className="icon size-md space-sm" />
in the top right corner of the web UI screen to access the self-help resources including:
* ClearML Python Package setup - Instruction to get started with the `clearml` Python package
* [ClearML on YouTube](https://www.youtube.com/c/ClearML/featured) <img src="/docs/latest/icons/ico-youtube.svg" alt="Youtube" className="icon size-md space-sm" /> - Instructional videos on integrating ClearML into your workflow
* [ClearML on YouTube](https://www.youtube.com/c/ClearML/featured) <img src="/docs/latest/icons/ico-youtube.svg" alt="YouTube" className="icon size-md space-sm" /> - Instructional videos on integrating ClearML into your workflow
* Online Documentation
* Pro Tips - Tips for working with ClearML efficiently
* [Contact Us](https://clear.ml/contact-us) - Quick access to ClearML contact form