# Datasets

Datasets reflect the dataset metadata logged to Vectice during model development.

{% hint style="info" %}
**Please note that Vectice does not store your actual datasets.** 

Vectice documents dataset metadata like origin, version, and artifacts from various sources, while automatically managing dataset versions for lineage tracking.
{% endhint %}

The Vectice API enables you to log all dataset artifacts used during development to the Vectice app, including your **origin**, **cleaned**, and **modeling datasets.**

| Dataset type      | Description                                                                                                             |
| ----------------- | ----------------------------------------------------------------------------------------------------------------------- |
| Origin datasets   | Origin datasets refer to your datasets containing **raw data**.                                                         |
| Cleaned datasets  | Cleaned datasets refer to your datasets that have been **cleaned and prepared for data modeling** or **data analysis**. |
| Modeling datasets | Modeling datasets combine **training**, **testing**, and **validating** data in a single dataset.                       |

## Things you should know

* Data scientists can capture basic statistics when logging their datasets' artifacts to Vectice. These statistics include the mean, median, variance, quartiles, and more. Learn more by viewing our [Capturing Dataset Statistics](/24.4/log-and-manage-assets-with-vectice-api/log-assets-to-vectice/log-datasets.md#capturing-dataset-statistics) section.

## Datasets best practices

* Logged datasets should have the dataset types **Origin**, **Cleaned**, and **Modeling** appended to the end of the corresponding dataset name for easy identification in the Vectice app.
* Create a phase with the **primary objective** of origin datasets registration. This is usually included in the **Data Preparation** phase of a CRISP-DM project. 
* Project phases that aim to clean and process raw data can have a requirement that ask members to log the **cleaned datasets**. This is typically completed in the **Data Preparation** phase of a CRISP-DM project.
* Mark your **most valuable iterations and assets** in the Vectice app by selecting the **star** next to the corresponding iteration and assets **before** beginning the phase review. This will make it easier for stakeholders and subject matter experts to identify the iterations and assets in review.

### Advanced Dataset guides

* [Dataset resources](/24.4/glossary/concepts/datasets/dataset-resources.md)
* [Dataset properties](/24.4/glossary/concepts/datasets/dataset-properties.md)
* [Dataset lineage and versions](/24.4/glossary/concepts/datasets/dataset-lineage-and-versions.md)

{% hint style="info" %}
Select the **star** next to the **valuable iterations** before completing a phase, even without review. This allows you to identify the **most valuable** **iterations** for that phase.
{% endhint %}

## How to log datasets to Vectice?

Logging datasets is your first step to knowledge capture in Vectice. You can log all datasets used during development, including your **origin datasets** (raw datasets), **cleaned datasets**, and **modeling datasets**. 

View our [How to log datasets](/24.4/log-and-manage-assets-with-vectice-api/log-assets-to-vectice/log-datasets.md) guide for more information on how to log datasets during iterative development.


---

# Agent Instructions: Querying This Documentation

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

Perform an HTTP GET request on the current page URL with the `ask` query parameter:

```
GET https://docs.vectice.com/24.4/glossary/concepts/datasets.md?ask=<question>
```

The question should be specific, self-contained, and written in natural language.
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.