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

Small edits (#1046)
Some checks failed
CI / build (push) Has been cancelled
Details

Browse Source
This commit is contained in:
pollfly
2025-02-20 09:32:26 +02:00
committed by GitHub
parent 60638ce467
commit 15416bb43d
8 changed files with 40 additions and 40 deletions
Download Patch File Download Diff File Expand all files Collapse all files

8
docs/webapp/applications/apps_dashboard.md
View File

@@ -28,13 +28,13 @@ of the chosen metric over time.
* Monitored Metric - Series - Metric series (variant) to track
* Monitored Metric - Trend - Choose whether to track the monitored metric's highest or lowest values
* **Slack Notification** (optional) - Set up Slack integration for notifications of task failure. Select the
`Alert on completed experiments` under `Additional options` to set up alerts for task completions.
`Alert on completed tasks` under `Additional options` to set up alerts for task completions.
* API Token - Slack workspace access token
* Channel Name - Slack channel to which task failure alerts will be posted
* Alert Iteration Threshold - Minimum number of task iterations to trigger Slack alerts (tasks that fail prior to the threshold will be ignored)
* **Additional options**
* Track manual (non agent-run) experiments as well - Select to include in the dashboard tasks that were not executed by an agent
* Alert on completed experiments - Select to include completed tasks in alerts: in the dashboard's Task Alerts section and in Slack Alerts.
* Track manual (non agent-run) tasks as well - Select to include in the dashboard tasks that were not executed by an agent
* Alert on completed tasks - Select to include completed tasks in alerts: in the dashboard's Task Alerts section and in Slack Alerts.
* **Export Configuration** - Export the app instance configuration as a JSON file, which you can later import to create
a new instance with the same configuration.
@@ -50,7 +50,7 @@ of the chosen metric over time.
Once a project dashboard instance is launched, its dashboard displays the following information about a project:
* Task Status Summary - Percentages of Tasks by status
* Task Type Summary - Percentages of local tasks vs. agent tasks
* Experiments Summary - Number of tasks by status over time
* Task Summary - Number of tasks by status over time
* Monitoring - GPU utilization and GPU memory usage
* Metric Monitoring - An aggregated view of the values of a metric over time
* Project's Active Workers - Number of workers currently executing tasks in the monitored project

12
docs/webapp/applications/apps_hpo.md
View File

@@ -56,18 +56,18 @@ limits.
**CONFIGURATION > HYPERPARAMETERS > Hydra**).
:::
* **Optimization Job Title** (optional) - Name for the HPO instance. This will appear in the instance list
* **Optimization Experiments Destination Project** (optional) - The project where optimization tasks will be saved.
* **Optimization Tasks Destination Project** (optional) - The project where optimization tasks will be saved.
Leave empty to use the same project as the Initial task.
* **Maximum Concurrent Tasks** - The maximum number of simultaneously running optimization tasks
* **Advanced Configuration** (optional)
* Limit Total HPO Experiments - Maximum total number of optimization tasks
* Number of Top Experiments to Save - Number of best performing tasks to save (the rest are archived)
* Limit Single Experiment Running Time (Minutes) - Time limit per optimization task. Tasks will be
* Limit Total HPO Tasks - Maximum total number of optimization tasks
* Number of Top Tasks to Save - Number of best performing tasks to save (the rest are archived)
* Limit Single Task Running Time (Minutes) - Time limit per optimization task. Tasks will be
stopped after the specified time elapsed
* Minimal Number of Iterations Per Single Experiment - Some search methods, such as Optuna, prune underperforming
* Minimal Number of Iterations Per Single Task - Some search methods, such as Optuna, prune underperforming
tasks. This is the minimum number of iterations per task before it can be stopped. Iterations are
based on the tasks' own reporting (for example, if tasks report every epoch, then iterations=epochs)
* Maximum Number of Iterations Per Single Experiment - Maximum iterations per task after which it will be
* Maximum Number of Iterations Per Single Task - Maximum iterations per task after which it will be
stopped. Iterations are based on the tasks' own reporting (for example, if tasks report every epoch,
then iterations=epochs)
* Limit Total Optimization Instance Time (Minutes) - Time limit for the whole optimization process (in minutes)