> For the complete documentation index, see [llms.txt](https://docs.tensorleap.ai/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://docs.tensorleap.ai/user-interface/dashboards/dashboard-actions/fetch-similar.md).

# Fetch Similar

This section describes what is "Fetch Similar" within the platform

It is often required to get similar samples to a given selection to complete an analysis, or test an hypothesis within the platform. For instance, if a subset of the reviewed samples is consistently failing and appears to share a common characteristic - such as a specific lighting condition or the presence of a particular object -retrieving similar samples can help validate whether that characteristic is indeed contributing to the failures. 

The "Fetch Similar" provides similar samples to selected samples [from the population exploration](/user-interface/dashboards/dashlets/population-exploration.md#selecting-and-reviewing-samples) or the [sample analysis](/user-interface/dashboards/dashlets/sample-analysis.md). The similar samples could be taken from a slice of the data, defined by [filters](/user-interface/dashboards/dashboard-actions/dashboard-filtering.md). 

### Fetch similar from population exploration

<figure><img src="https://3509361326-files.gitbook.io/~/files/v0/b/gitbook-x-prod.appspot.com/o/spaces%2F9UXeOlFqlw8pl79U2HGU%2Fuploads%2FbHboV9n27jcnGGN3TRwX%2FScreenshot%202026-07-30%20at%2012.55.23.png?alt=media&amp;token=d702cbf9-a261-40ff-a03e-384f562767a7" alt=""><figcaption><p>Running fetch similar from the population exploration to get the most similar samples given a specific filter (samples should be in the validation set)</p></figcaption></figure>

Once samples has been selected from the population exploration, we could use the "Fetch Similar" button to retrieve similar samples. The computed result would be added as a special [filter](/user-interface/dashboards/dashboard-actions/dashboard-filtering.md) to the population exploration, resulting in only the most similar samples presented.

### Fetch Similar from sample analysis

<figure><img src="https://3509361326-files.gitbook.io/~/files/v0/b/gitbook-x-prod.appspot.com/o/spaces%2F9UXeOlFqlw8pl79U2HGU%2Fuploads%2FSGBzjLtOZjmMVWit21Im%2FScreenshot%202026-07-30%20at%2012.57.48.png?alt=media&amp;token=d79c859c-121b-4423-a983-70617a7cfb7d" alt=""><figcaption><p>Fetch similar from sample analysis by clicking the magnet icon</p></figcaption></figure>

Given a selection of samples in the [sample analysis](/user-interface/dashboards/dashlets/sample-analysis.md) dashlet, similar samples could also be retrieved. This would create a special global filter, applying the entire dashboard to only contain these similar samples.

{% hint style="warning" %}
Notice that it is sometimes required to examine only one slice (i.e. train), but to fetch similars from an disjoint slice (i.e. unlabeled). Such an action would disable all pre-existing filters from the dashlet, to ensure the fetch similar filter is not in contradiction with existing filters.
{% endhint %}


---

# Agent Instructions
This documentation is published with GitBook. GitBook is the documentation platform designed so that both humans and AI agents can read, navigate, and reason over technical content effectively. Learn more at gitbook.com.

## Querying This Documentation
If you need additional information that is not directly available in this page, you can query the documentation by asking a question.

Perform an HTTP GET request on the following URL with the `ask` and `goal` query parameters:

```
GET https://docs.tensorleap.ai/user-interface/dashboards/dashboard-actions/fetch-similar.md?ask=<question>&goal=<user_goal>
```

`ask` is the immediate question: it should be specific, self-contained, and written in natural language.
`goal` is what the user is ultimately trying to achieve, the reason they need the answer. Sharing it helps GitBook give you a better, more relevant answer. A goal is most helpful when it describes the outcome the user wants rather than restating the question. For example, with `ask=how do I create an API token`, a goal like `automate deployments from our CI pipeline` lets GitBook tailor the answer to that use case.

The response will contain a direct answer to the question and relevant excerpts and sources from the documentation.

Use this mechanism when the answer is not explicitly present in the current page, you need clarification or additional context, or you want to retrieve related documentation sections.