> For the complete documentation index, see [llms.txt](https://docs.cloudeka.ai/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://docs.cloudeka.ai/guidance-for-enterprise/deka-gpu-mlops/experiments-kfp.md).

# Experiments (KFP)

The **Experiments (KFP)** menu is used to group a collection of experiments or pipelines that are executed to achieve specific objectives. Each experiment in KFP acts as a container that manages a set of pipeline runs related to a particular experiment or goal. This helps organize, compare, and analyze the results of multiple pipeline runs. On the Cloudeka Deka MLOps Service Portal page, select the **Experiments (KFP)** menu.

<figure><img src="/files/LRM73DsMVBla8JYvszIm" alt=""><figcaption></figcaption></figure>

In the **Experiments (KFP)** menu, several submenus are available, as described below.

1. The **Active** submenu is designed to help you manage and monitor ongoing machine learning experiments in the Cloudeka Deka MLOps Service Portal.

   <figure><img src="/files/IjJHOv4My27tRtutnsyK" alt=""><figcaption></figcaption></figure>

   * **Refresh**\
     This option allows you to refresh the page and display the latest information about ongoing experiments. It helps ensure that the displayed data remains accurate and up to date.
   * **Archive**\
     This feature is used to store completed or inactive experiment runs. Archiving runs helps keep the interface organized while retaining the data for future reference.
   * **Clone Run**\
     This option allows you to duplicate an existing experiment run. It is useful when you want to create variations of an experiment without reconfiguring all parameters.
   * **Compare Runs**\
     This feature is used to compare the results of multiple experiment runs. By comparing runs, you can analyze the performance of different models and identify the most effective one.
   * **Create Experiment**\
     This option allows you to create a new experiment. You can define the parameters and configurations required to run the experiment.
2. The **Archived** submenu contains experiments that have been archived. Although archived experiments are no longer active, they can still be accessed for reference and analysis. This helps keep the workspace organized by separating active experiments from completed or inactive ones.

   <figure><img src="/files/wyMcfEYa6SpS7ULGc5R1" alt=""><figcaption></figcaption></figure>

   \
   **Refresh**, this option allows you to refresh the page and display the latest information about archived experiments.<br>