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

Add bash script support when creating task via UI (#998)

Browse Source
This commit is contained in:
pollfly 2025-01-12 15:57:06 +02:00 committed by GitHub
parent cd442c989f
commit 7ba665764f
No known key found for this signature in database
GPG Key ID: B5690EEEBB952194
3 changed files with 145 additions and 142 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

2
docs/faq.md
View File

@ -165,7 +165,7 @@ add a [custom column](webapp/webapp_model_table.md#customizing-the-models-table)
that metric column. Available custom column options depend upon the models in the table and the metrics that have been that metric column. Available custom column options depend upon the models in the table and the metrics that have been
attached to them (see [Logging Metrics and Plots](clearml_sdk/model_sdk.md#logging-metrics-and-plots)). attached to them (see [Logging Metrics and Plots](clearml_sdk/model_sdk.md#logging-metrics-and-plots)).
ClearML associates models with the experiments that created them, so you can also add a [custom column](webapp/webapp_exp_table.md#customizing-the-experiments-table) ClearML associates models with the experiments that created them, so you can also add a [custom column](webapp/webapp_exp_table.md#customizing-the-task-table)
in an experiments table and sort by that metric column. in an experiments table and sort by that metric column.
<br/> <br/>

283
docs/webapp/webapp_exp_table.md
View File

@ -1,78 +1,81 @@
--- ---
title: The Experiments Table title: The Task Table
--- ---
The experiments table is a [customizable](#customizing-the-experiments-table) list of experiments associated with a project. From the experiments The task table is a [customizable](#customizing-the-task-table) list of tasks associated with a project. From the tasks
table, view experiment details, and work with experiments (reset, clone, enqueue, create [tracking leaderboards](../guides/ui/building_leader_board.md) table, view task details, and work with tasks (reset, clone, enqueue, create [tracking leaderboards](../guides/ui/building_leader_board.md)
to monitor experimentation, and more). The experiments table's auto-refresh lets users continually monitor experiment progress. to monitor experimentation, and more). The task table's auto-refresh lets users continually monitor task progress.
View the experiments in table view <img src="/docs/latest/icons/ico-table-view.svg" alt="Table view" className="icon size-md space-sm" />, View the tasks in table view <img src="/docs/latest/icons/ico-table-view.svg" alt="Table view" className="icon size-md space-sm" />,
details view <img src="/docs/latest/icons/ico-split-view.svg" alt="Details view" className="icon size-md space-sm" />, or details view <img src="/docs/latest/icons/ico-split-view.svg" alt="Details view" className="icon size-md space-sm" />, or
comparison view <img src="/docs/latest/icons/ico-charts-view.svg" alt="Comparison view" className="icon size-md space-sm" /> comparison view <img src="/docs/latest/icons/ico-charts-view.svg" alt="Comparison view" className="icon size-md space-sm" />
using the buttons on the top left of the page. Use the table view for a comparative view of your experiments according using the buttons on the top left of the page. Use the table view for a comparative view of your tasks according
to columns of interest. Use the details view to access a selected experiment's details, while keeping the experiment list to columns of interest. Use the details view to access a selected task's details, while keeping the task list
in view. Details view can also be accessed by double-clicking a specific experiment in the table view to open its details view. in view. Details view can also be accessed by double-clicking a specific task in the table view to open its details view.
Use the [comparison view](#comparing-experiments) to compare your experiments' scalar and plot results (for a more in Use the [comparison view](#comparing-tasks) to compare your tasks' scalar and plot results (for a more in
depth comparison, see [Comparing Experiments](webapp_exp_comparing.md)). This view compares depth comparison, see [Comparing Tasks](webapp_exp_comparing.md)). This view compares
the scalars/plots of currently selected experiments. If no experiments are selected, the first 100 the scalars/plots of currently selected tasks. If no tasks are selected, the first 100
visible experiments in the table are compared. visible tasks in the table are compared.
You can archive experiments so the experiments table doesn't get too cluttered. Click **OPEN ARCHIVE** on the top of the You can archive tasks so the table doesn't get too cluttered. Click **OPEN ARCHIVE** on the top of the
table to open the archive and view all archived experiments. From the archive, you can restore table to open the archive and view all archived tasks. From the archive, you can restore
experiments to remove them from the archive. You can also permanently delete experiments. tasks to remove them from the archive. You can also permanently delete tasks.
You can download the experiments table as a CSV file by clicking <img src="/docs/latest/icons/ico-download.svg" alt="Download" className="icon size-md space-sm" /> You can download the task table as a CSV file by clicking <img src="/docs/latest/icons/ico-download.svg" alt="Download" className="icon size-md space-sm" />
and choosing one of these options: and choosing one of these options:
* **Download onscreen items** - Download the values for experiments currently visible on screen * **Download onscreen items** - Download the values for tasks currently visible on screen
* **Download all items** - Download the values for all experiments in this project that match the current active filters * **Download all items** - Download the values for all tasks in this project that match the current active filters
The downloaded data consists of the currently displayed table columns. The downloaded data consists of the currently displayed table columns.
![Experiment table](../img/webapp_experiment_table.png) ![Task table](../img/webapp_experiment_table.png)
## Creating Experiments ## Creating Tasks
You can create experiments by: You can create tasks by:
* Running code instrumented with ClearML (see [Task Creation](../clearml_sdk/task_sdk.md#task-creation)) * Running code instrumented with ClearML (see [Task Creation](../clearml_sdk/task_sdk.md#task-creation))
* [Cloning an existing experiment](webapp_exp_reproducing.md) * [Cloning an existing task](webapp_exp_reproducing.md)
* Through the UI interface: Input the experiment's details, including its source code and python requirements, and then * Via CLI using [`clearml-task`](../apps/clearml_task.md)
* Through the UI interface: Input the task's details, including its source code and python requirements, and then
run it through a [ClearML Queue](../fundamentals/agents_and_queues.md#what-is-a-queue) or save it as a *draft*. run it through a [ClearML Queue](../fundamentals/agents_and_queues.md#what-is-a-queue) or save it as a *draft*.
To create an experiment through the UI interface: To create a task through the UI interface:
1. Click `+ New Experiment` 1. Click `+ New Task`
1. In the `Create Experiment` modal, input the following information: 1. In the `Create Task` modal, input the following information:
* **Code** * **Code** - What this task is going to run
* Experiment name * Task name
* Git * Git - Optional fields for checking out the code from a git repository:
* Repository URL * Repository URL
* Version specification - one of the following: * Version specification - one of the following:
* Tag * Tag
* Branch * Branch
* Commit ID * Commit ID
* Execution Entry Point * Entry Point - The code to run
* Working Directory * Working Directory
* One of the following * Script type - Python/Shell
* Script name * Binary - The binary executing the script (e.g. python3, bash etc).
* Module (see [python module specification](https://docs.python.org/3/using/cmdline.html#cmdoption-m)) * Type – How the code is provided
* Add `Task.init` call - If selected, [`Task.init()`](../references/sdk/task.md#taskinit) call is added to the * Script - The name of the file to run using the above specified binary
entry point. Select if it is not already called within your code * Module - The name of a python module to run (Python only, see [Python module specification](https://docs.python.org/3/using/cmdline.html#cmdoption-m))
* Custom code - Directly provide the code to run. Write code, or upload a file:
* File name - The script in which your code is stored. Click `Upload` to upload an existing file.
* Content - The actual code. Click `Edit` to modify the script’s contents.
* Add `Task.init` call (Python only) - If selected, a [`Task.init()`](../references/sdk/task.md#taskinit) call is automatically added to
your script (Use if if you script does not yet make use of ClearML)
* **Arguments** (*optional*) - Add [hyperparameter](../fundamentals/hyperparameters.md) values. * **Arguments** (*optional*) - Add [hyperparameter](../fundamentals/hyperparameters.md) values.
* **Environment** (*optional*) - Set up the experiment’s python execution environment using either of the following * **Environment** (*optional*) - Set up the task’s execution environment
options: * Python - Python environment settings
* Use Poetry specification - Requires specifying a docker image for the experiment to be executed in. * Use Poetry - Force Poetry instead of pip package manager. Disables additional python settings.
* Manually specify the python environment configuration: * Preinstalled venv - The name of a virtual environment available in the task’s execution environment to use when
* Python binary - The python executable to use running the task. Additionally, specify how to use the virtual environment:
* Preinstalled venv - A specific existing virtual environment to use. Requires specifying a docker image for the * Skip - Try to automatically detect an available virtual environment, and use it as is.
experiment to be executed in. * Use `requirements.txt` file - Install packages from a `requirements.txt` file into the specified virtual environment.
* Python package specification: * Specify Packages - Install the specified packages into the specified virtual environment
* Skip - Assume system packages are available. Requires specifying a docker image for the experiment to be * Environment Variables - Set these environment variables when running the task
executed in. * **Container** (*optional*) - Specify container configuration for executing the task
* Use an existing `requirements.txt` file * Image - Image to use for running the task
* Explicitly specify the required packages * Arguments - Add container arguments as a single string
* **Docker** (*optional*) - Specify Docker container configuration for executing the experiment * Startup Script - Add a bash script to be executed inside the container before setting up the task's environment
* Image - Docker image to use for running the experiment
* Arguments - Add Docker arguments as a single string
* Startup Script - Add a bash script to be executed inside the Docker before setting up the experiment's environment
:::important :::important
For a task to run in the specified container, the ClearML Agent executing the task must be running in For a task to run in the specified container, the ClearML Agent executing the task must be running in
@ -85,39 +88,39 @@ To create an experiment through the UI interface:
::: :::
* **Run** * **Run**
* Queue - [ClearML Queue](../fundamentals/agents_and_queues.md#what-is-a-queue) where the experiment should be * Queue - [ClearML Queue](../fundamentals/agents_and_queues.md#what-is-a-queue) where the task should be
enqueued for execution enqueued for execution
* Output Destination - A URI where experiment outputs should be stored (ClearML file server by default). * Output Destination - A URI where task outputs should be stored (ClearML file server by default).
1. Once you have input all the information, click one of the following options 1. Once you have input all the information, click one of the following options
* Save as Draft - Save the experiment as a new draft task. * Save as Draft - Save the task as a new draft.
* Run - Enqueue the experiment for execution in the queue specified in the **Run** tab * Run - Enqueue the task for execution in the queue specified in the **Run** tab
Once you have completed the experiment creation wizard, the experiment will be saved in your current project (where Once you have completed the task creation wizard, the task will be saved in your current project (where
you clicked `+ New Experiment`). See what you can do with your experiment in [Experiment Actions](#experiment-actions). you clicked `+ New Task`). See what you can do with your task in [Task Actions](#task-actions).
## Experiments Table Columns ## Task Table Columns
The experiments table default and customizable columns are described in the following table. The task table default and customizable columns are described in the following table.
| Column | Description | Type | | Column | Description | Type |
|---|---|---| |---|---|---|
| **TYPE** | Type of experiment. ClearML supports multiple [task types](../fundamentals/task.md#task-types) for experimentation, and a variety of workflows and use cases. | Default | | **TYPE** | Type of task. ClearML supports multiple [task types](../fundamentals/task.md#task-types) for experimentation, and a variety of workflows and use cases. | Default |
| **NAME** | Experiment name. | Default | | **NAME** | Task name. | Default |
| **TAGS** | Descriptive, user-defined, color-coded tags assigned to experiments. Use tags to classify experiments, and filter the list. See [tagging experiments](webapp_exp_track_visual.md#tagging-experiments). | Default | | **TAGS** | Descriptive, user-defined, color-coded tags assigned to tasks. Use tags to classify tasks, and filter the list. See [tagging tasks](webapp_exp_track_visual.md#tagging-tasks). | Default |
| **STATUS** | Experiment state (status). See a list of the [task states and state transitions](../fundamentals/task.md#task-states). If you programmatically set task progress values, you will also see a progress indicator for Running, Failed, and Aborted tasks. See [here](../clearml_sdk/task_sdk.md#tracking-task-progress). | Default | | **STATUS** | Task state (status). See a list of the [task states and state transitions](../fundamentals/task.md#task-states). If you programmatically set task progress values, you will also see a progress indicator for Running, Failed, and Aborted tasks. See [here](../clearml_sdk/task_sdk.md#tracking-task-progress). | Default |
| **PROJECT** | Name of experiment's project. | Default | | **PROJECT** | Name of task's project. | Default |
| **USER** | User who created or cloned the experiment. | Default (hidden) | | **USER** | User who created or cloned the task. | Default (hidden) |
| **STARTED** | Elapsed time since the experiment started. To view the date and time of start, hover over the elapsed time. | Default | | **STARTED** | Elapsed time since the task started. To view the date and time of start, hover over the elapsed time. | Default |
| **UPDATED** | Elapsed time since the last update to the experiment. To view the date and time of update, hover over the elapsed time. | Default | | **UPDATED** | Elapsed time since the last update to the task. To view the date and time of update, hover over the elapsed time. | Default |
| **ITERATION** | Last or most recent iteration of the experiment. | Default | | **ITERATION** | Last or most recent iteration of the task. | Default |
| **DESCRIPTION** | A description of the experiment. For cloned experiments, the description indicates it was auto generated with a timestamp. | Default (hidden) | | **DESCRIPTION** | A description of the task. For cloned tasks, the description indicates it was auto generated with a timestamp. | Default (hidden) |
| **RUN TIME** | The current / total running time of the experiment. | Default (hidden) | | **RUN TIME** | The current / total running time of the task. | Default (hidden) |
| **_Metrics_** | Add metrics column (last, minimum, and/or maximum values). The metrics depend upon the experiments in the table. See [adding metrics](#to-add-metrics). | Customizable | | **_Metrics_** | Add metrics column (last, minimum, and/or maximum values). The metrics depend upon the tasks in the table. See [adding metrics](#to-add-metrics). | Customizable |
| **_Hyperparameters_** | Add hyperparameters. The hyperparameters depend upon the experiments in the table. See [adding hyperparameters](#to-add-hyperparameters). | Customizable | | **_Hyperparameters_** | Add hyperparameters. The hyperparameters depend upon the tasks in the table. See [adding hyperparameters](#to-add-hyperparameters). | Customizable |
## Customizing the Experiments Table ## Customizing the Task Table
Customize the table using any of the following: Customize the table using any of the following:
* Dynamic column order - Drag a column title to a different position. * Dynamic column order - Drag a column title to a different position.
@ -130,13 +133,13 @@ Customize the table using any of the following:
main column list. Added columns are by default displayed in the table. You can remove the custom columns from the main column list. Added columns are by default displayed in the table. You can remove the custom columns from the
main column list or the column addition windows. main column list or the column addition windows.
* [Filter columns](#filtering-columns) * [Filter columns](#filtering-columns)
* Sort columns - According to metrics and hyperparameters, type of experiment, experiment name, start and last update elapsed time, and last iteration. * Sort columns - According to metrics and hyperparameters, type of task, task name, start and last update elapsed time, and last iteration.
Use experiments table customization for various use cases, including: Use task table customization for various use cases, including:
* Creating a [leaderboard](#creating-an-experiment-leaderboard) that will update in real time with experiment * Creating a [leaderboard](#creating-a-task-leaderboard) that will update in real time with task
performance, which can be shared and stored. performance, which can be shared and stored.
* Sorting models by metrics - Models are associated with the experiments that created them. For each metric, use the last * Sorting models by metrics - Models are associated with the tasks that created them. For each metric, use the last
value, the minimal value, and/or the maximal value. value, the minimal value, and/or the maximal value.
* Tracking hyperparameters - Track hyperparameters by adding them as columns, and applying filters and sorting. * Tracking hyperparameters - Track hyperparameters by adding them as columns, and applying filters and sorting.
@ -144,25 +147,25 @@ Changes are persistent (cached in the browser), and represented in the URL so cu
bookmark and shared with other ClearML users to collaborate. bookmark and shared with other ClearML users to collaborate.
:::note :::note
The following experiments-table customizations are saved on a **per-project** basis: The following task-table customizations are saved on a **per-project** basis:
* Columns order * Columns order
* Column width * Column width
* Active sort order * Active sort order
* Active filters * Active filters
* Custom columns * Custom columns
If a project has subprojects, the experiments can be viewed by their subproject groupings or together with If a project has subprojects, the tasks can be viewed by their subproject groupings or together with
all the experiments in the project. The customizations of these two views are saved separately. all the tasks in the project. The customizations of these two views are saved separately.
::: :::
### Adding Metrics and/or Hyperparameters ### Adding Metrics and/or Hyperparameters
![Experiment table customization gif](../img/gif/webapp_exp_table_cust.gif) ![Task table customization gif](../img/gif/webapp_exp_table_cust.gif)
Add metrics and/or hyperparameters columns to the experiments table. The metrics and hyperparameters depend upon the Add metrics and/or hyperparameters columns to the task table. The metrics and hyperparameters depend upon the
experiments in the table. tasks in the table.
#### To Add Metrics: #### To Add Metrics:
@ -175,7 +178,7 @@ experiments in the table.
hyperparameter checkboxes. hyperparameter checkboxes.
:::note Float Values Display :::note Float Values Display
By default, the experiments table displays rounded up float values. Hover over a float to view its precise value in the By default, the task table displays rounded up float values. Hover over a float to view its precise value in the
tooltip that appears. To view all precise values in a column, hover over a float and click <img src="/docs/latest/icons/ico-line-expand.svg" alt="Expand" className="icon size-md" />. tooltip that appears. To view all precise values in a column, hover over a float and click <img src="/docs/latest/icons/ico-line-expand.svg" alt="Expand" className="icon size-md" />.
::: :::
@ -202,38 +205,38 @@ in the top right corner of the table.
## Experiment Actions ## Task Actions
The following table describes the actions that can be done from the experiments table, including the [states](../fundamentals/task.md#task-states) The following table describes the actions that can be done from the task table, including the [states](../fundamentals/task.md#task-states)
that allow each operation. that allow each operation.
Access these actions in any of the following ways: Access these actions in any of the following ways:
* In the experiments table, right-click an experiment or hover over an experiment and click <img src="/docs/latest/icons/ico-dots-v-menu.svg" alt="Dot menu" className="icon size-md space-sm" /> * In the task table, right-click a task or hover over a task and click <img src="/docs/latest/icons/ico-dots-v-menu.svg" alt="Dot menu" className="icon size-md space-sm" />
to open the context menu to open the context menu
* In an experiment info panel, click the menu button <img src="/docs/latest/icons/ico-bars-menu.svg" alt="Bar menu" className="icon size-md space-sm" /> * In a task info panel, click the menu button <img src="/docs/latest/icons/ico-bars-menu.svg" alt="Bar menu" className="icon size-md space-sm" />
* Through the batch action bar: available at screen bottom when multiple experiments are selected * Through the batch action bar: available at screen bottom when multiple tasks are selected
| Action | Description | States Valid for the Action | State Transition | | Action | Description | States Valid for the Action | State Transition |
|---|---|---|---| |---|---|---|-------------------------------------------------|
| Details | Open the experiment's [info panel](webapp_exp_track_visual.md#info-panel) (keeps the experiments list in view). Can also be accessed by double-clicking an experiment in the experiments table. | Any state | None | | Details | Open the task's [info panel](webapp_exp_track_visual.md#info-panel) (keeps the tasks list in view). Can also be accessed by double-clicking a task in the task table. | Any state | None |
| View Full Screen | View experiment details in [full screen](webapp_exp_track_visual.md#full-screen-details-view). | Any state | None | | View Full Screen | View task details in [full screen](webapp_exp_track_visual.md#full-screen-details-view). | Any state | None |
| Manage Queue | If an experiment is *Pending* in a queue, view the utilization of that queue, manage that queue (remove experiments and change the order of experiments), and view information about the worker(s) listening to the queue. See the [Orchestration](webapp_workers_queues.md) page. | *Enqueued* | None | | Manage Queue | If a task is *Pending* in a queue, view the utilization of that queue, manage that queue (remove tasks and change the order of tasks), and view information about the worker(s) listening to the queue. See the [Orchestration](webapp_workers_queues.md) page. | *Enqueued* | None |
| View Worker | If an experiment is *Running*, view resource utilization, worker details, and queues to which a worker is listening. | *Running* | None | | View Worker | If a task is *Running*, view resource utilization, worker details, and queues to which a worker is listening. | *Running* | None |
| Share | For **ClearML Hosted Service** users only, [share](webapp_exp_sharing.md) an experiment and its model with a **ClearML Hosted Service** user in another workspace. | Any state | None | | Share | For **ClearML Hosted Service** users only, [share](webapp_exp_sharing.md) a task and its model with a **ClearML Hosted Service** user in another workspace. | Any state | None |
| Archive | Move experiment to the project's archive. If it is shared (ClearML Hosted Service only), the experiment becomes private. | Any state | *Pending* to *Draft* | | Archive | Move task to the project's archive. If it is shared (ClearML Hosted Service only), the task becomes private. | Any state | *Pending* to *Draft* |
| Restore |Action available in the archive. Restore an experiment to the active experiments table.| Any State | None | | Restore |Action available in the archive. Restore a task to the active task table.| Any State | None |
| Delete | Action available in the archive. Delete an experiment, which will also remove all their logs, results, artifacts and debug samples. | Any State | N/A | | Delete | Action available in the archive. Delete a task, which will also remove all their logs, results, artifacts and debug samples. | Any State | N/A |
| Enqueue | Add an experiment to a queue for a worker or workers (listening to the queue) to execute. | *Draft*, *Aborted* | *Pending* | | Enqueue | Add a task to a queue for a worker or workers (listening to the queue) to execute. | *Draft*, *Aborted* | *Pending* |
| Dequeue | Remove an experiment from a queue. | *Pending* | *Draft* | | Dequeue | Remove a task from a queue. | *Pending* | *Draft* |
| Reset | Delete the log and output from a previous run of an experiment (for example, before rerunning it). | *Completed*, *Aborted*, or *Failed* | *Draft* | | Reset | Delete the log and output from a previous run of a task (for example, before rerunning it). | *Completed*, *Aborted*, or *Failed* | *Draft* |
| Abort | Manually terminate a *Running* experiment. | *Running* | *Aborted* | | Abort | Manually terminate a *Running* task. | *Running* | *Aborted* |
| Abort All Children | Manually terminate all *Running* experiments which have this task as a parent | *Running* or *Aborted* | None for parent experiment, *Aborted* for child experiments | | Abort All Children | Manually terminate all *Running* tasks which have this task as a parent | *Running* or *Aborted* | None for parent task, *Aborted* for child tasks |
| Retry | Enqueue a failed experiment in order to rerun it. Make sure you have resolved the external problem which previously prevented the experiment’s completion. | *Failed* | *Pending* | | Retry | Enqueue a failed task in order to rerun it. Make sure you have resolved the external problem which previously prevented the task’s completion. | *Failed* | *Pending* |
| Publish | Publish an experiment to prevent changes to its tracking data, inputs, and outputs. Published experiments and their models are read-only. *Published* experiments cannot be enqueued, but they can be cloned, and their clones can be edited, tuned, and enqueued. | *Completed*, *Aborted*, or *Failed*. | *Published* | | Publish | Publish a task to prevent changes to its tracking data, inputs, and outputs. Published tasks and their models are read-only. *Published* tasks cannot be enqueued, but they can be cloned, and their clones can be edited, tuned, and enqueued. | *Completed*, *Aborted*, or *Failed*. | *Published* |
| Add Tag | Tag experiments with color-coded labels to assist you in organizing your work. See [tagging experiments](webapp_exp_track_visual.md#tagging-experiments). | Any state | None | | Add Tag | Tag tasks with color-coded labels to assist you in organizing your work. See [tagging tasks](webapp_exp_track_visual.md#tagging-experiments). | Any state | None |
| Clone | Make an exact, editable copy of an experiment (for example, to reproduce an experiment, but keep the original). | *Draft* | Newly Cloned Experiment is *Draft* | | Clone | Make an exact, editable copy of a task (for example, to reproduce a task, but keep the original). | *Draft* | Newly cloned task is *Draft* |
| Move to Project | Move an experiment to another project. | Any state | None | | Move to Project | Move a task to another project. | Any state | None |
| Compare | Compare selected experiments (see [Comparing Experiments](webapp_exp_comparing.md)) | Any state | None | | Compare | Compare selected tasks (see [Comparing Tasks](webapp_exp_comparing.md)) | Any state | None |
:::important Enterprise Feature :::important Enterprise Feature
The ClearML Enterprise Server provides a mechanism to define your own custom actions, which will The ClearML Enterprise Server provides a mechanism to define your own custom actions, which will
@ -241,43 +244,43 @@ appear in the context menu. Create a custom action by defining an HTTP request t
action. For more information see [Custom UI Context Menu Actions](../deploying_clearml/clearml_server_config.md#custom-ui-context-menu-actions). action. For more information see [Custom UI Context Menu Actions](../deploying_clearml/clearml_server_config.md#custom-ui-context-menu-actions).
::: :::
Most of the actions mentioned in the chart above can be performed on multiple experiments at once. Most of the actions mentioned in the chart above can be performed on multiple tasks at once.
[Select multiple experiments](#selecting-multiple-experiments), then use either the context menu, or the batch action bar [Select multiple tasks](#selecting-multiple-tasks), then use either the context menu, or the batch action bar
that appears at the bottom of the page, to perform that appears at the bottom of the page, to perform
operations on the selected experiments. Actions can be performed only on the experiments that match the action criteria operations on the selected tasks. Actions can be performed only on the tasks that match the action criteria
(for example, only *Running* experiments can be aborted). The context menu shows the number (for example, only *Running* tasks can be aborted). The context menu shows the number
of experiments that can be affected by each action. The same information can be found in the batch action bar, in a tooltip that of tasks that can be affected by each action. The same information can be found in the batch action bar, in a tooltip that
appears when hovering over an action icon. appears when hovering over an action icon.
![Experiment table batch operations](../img/webapp_experiment_table_context_menu.png) ![Task table batch operations](../img/webapp_experiment_table_context_menu.png)
## Selecting Multiple Experiments ## Selecting Multiple Tasks
Select multiple experiments by clicking the checkbox on the left of each relevant experiment. Clear any existing selection Select multiple tasks by clicking the checkbox on the left of each relevant task. Clear any existing selection
by clicking the checkbox in the top left corner of the table. by clicking the checkbox in the top left corner of the table.
Click the checkbox in the top left corner of the table to select all items currently visible. Click the checkbox in the top left corner of the table to select all items currently visible.
An extended bulk selection tool is available through the down arrow next to the checkbox in the top left corner, enabling An extended bulk selection tool is available through the down arrow next to the checkbox in the top left corner, enabling
selecting items beyond the items currently on-screen: selecting items beyond the items currently on-screen:
* **All** - Select all experiments in the project * **All** - Select all tasks in the project
* **None** - Clear selection * **None** - Clear selection
* **Filtered** - Select **all experiments in the project** that match the current active filters in the project * **Filtered** - Select **all tasks in the project** that match the current active filters in the project
## Comparing Experiments ## Comparing Tasks
The comparison view compares experiment scalar and plot results (for a more in depth comparison, see [Comparing Experiments](webapp_exp_comparing.md)). The comparison view compares task scalar and plot results (for a more in depth comparison, see [Comparing Experiments](webapp_exp_comparing.md)).
When selected, the view presents a comparison of all [selected experiments](#selecting-multiple-experiments). If no When selected, the view presents a comparison of all [selected tasks](#selecting-multiple-tasks). If no
experiments are selected, the first 100 visible experiments in the table are displayed in the comparison. tasks are selected, the first 100 visible tasks in the table are displayed in the comparison.
In the dropdown menu, select to view **Scalars** or **Plots**. In the dropdown menu, select to view **Scalars** or **Plots**.
**Scalars** shows experiment scalar results as time series line graphs. **Scalars** shows task scalar results as time series line graphs.
![Merged comparison plots](../img/webapp_compare_view_1.png) ![Merged comparison plots](../img/webapp_compare_view_1.png)
All single value scalars are plotted into a single clustered bar chart under the "Summary" title, where each cluster All single value scalars are plotted into a single clustered bar chart under the "Summary" title, where each cluster
represents a reported metric, and each bar in the cluster represents an experiment. represents a reported metric, and each bar in the cluster represents a task.
![Single scalar comparison](../img/webapp_compare_view_3.png) ![Single scalar comparison](../img/webapp_compare_view_3.png)
@ -288,34 +291,34 @@ Click <img src="/docs/latest/icons/ico-tune.svg" alt="Tune" className="icon size
[scalar plot tools](webapp_exp_track_visual.md#scalar-plot-tools). [scalar plot tools](webapp_exp_track_visual.md#scalar-plot-tools).
**Plots** shows the last reported iteration sample of each metric/variant combination per compared **Plots** shows the last reported iteration sample of each metric/variant combination per compared
experiment. task.
Line, scatter, box, and bar graphs are compared by overlaying each metric/variant from all compared experiments' into a Line, scatter, box, and bar graphs are compared by overlaying each metric/variant from all compared tasks' into a
single comparative plot. single comparative plot.
![Merged comparison plots](../img/webapp_compare_view_4.png) ![Merged comparison plots](../img/webapp_compare_view_4.png)
Other plot types are displayed separately for each experiment. Other plot types are displayed separately for each task.
![Separate comparison plots](../img/webapp_compare_view_2.png) ![Separate comparison plots](../img/webapp_compare_view_2.png)
## Creating an Experiment Leaderboard ## Creating a Task Leaderboard
Filter and sort the experiments of any project to create a leaderboard that can be shared and stored. This leaderboard Filter and sort the tasks of any project to create a leaderboard that can be shared and stored. This leaderboard
updates in real time with experiment performance and outputs. updates in real time with task performance and outputs.
Modify the experiments table in the following ways to create a customized leaderboard: Modify the task table in the following ways to create a customized leaderboard:
* Add experiment configuration ([hyperparameters](#to-add-hyperparameters)) * Add task configuration ([hyperparameters](#to-add-hyperparameters))
* Edit and add experiments [properties](webapp_exp_track_visual.md#user-properties) * Edit and add task [properties](webapp_exp_track_visual.md#user-properties)
* Add reported [metrics](#to-add-metrics), any time series reported metric can be selected, then select the last reported * Add reported [metrics](#to-add-metrics), any time series reported metric can be selected, then select the last reported
value, or the minimum / maximum reported value. value, or the minimum / maximum reported value.
* Filter based on user (dropdown and select) or [experiment types](../fundamentals/task.md#task-types) * Filter based on user (dropdown and select) or [task types](../fundamentals/task.md#task-types)
* Add specific [tags](webapp_exp_track_visual.md#tagging-experiments) and filter based on them. * Add specific [tags](webapp_exp_track_visual.md#tagging-tasks) and filter based on them.
Now the table can be sorted based on any of the columns (probably one of the performance metrics). Select to filter experiments Now the table can be sorted based on any of the columns (probably one of the performance metrics). Select to filter tasks
based on their name by using the search bar. based on their name by using the search bar.
The final dashboard can be shared by copying the URL from the address bar, this address will replicate the exact same dashboard on any browser. The final dashboard can be shared by copying the URL from the address bar, this address will replicate the exact same dashboard on any browser.
The dashboard can also be bookmarked for later use. The dashboard can also be bookmarked for later use.
![Experiment table sharing](../img/webapp_exp_table_sharing.png) ![Task table sharing](../img/webapp_exp_table_sharing.png)

2
docs/webapp/webapp_exp_track_visual.md
View File

@ -29,7 +29,7 @@ table / full screen**.
### Info Panel ### Info Panel
The info panel keeps the experiment table in view so that [experiment actions](webapp_exp_table.md#experiment-actions) The info panel keeps the experiment table in view so that [experiment actions](webapp_exp_table.md#task-actions)
can be performed from the table (as well as the menu in the info panel). can be performed from the table (as well as the menu in the info panel).
![Info panel](../img/webapp_tracking_40.png) ![Info panel](../img/webapp_tracking_40.png)