# Performance Metrics

The NLU service uses various metrics to show the performance and characteristics of the model. The correct interpretation of these metrics is vital to utilize the full power of NLU.

The overall metrics we use in NLU are 

1. Precision
2. Recall
3. F1-score
4. Accuracy. 

{% hint style="info" %}
For the definition of each metric, please refer to [Term](/nlu-api-dashboard/terms.md)s page
{% endhint %}

## How to Measure Performance

You can train a model by clicking the 'Train Model' button on the top right corner of the Dashboard. The training process ends with measuring the model's performance using the samples you marked as 'Test' in the Dashboard tab page \[Image 1].

<figure><img src="/files/oYzAxYEXRWsJK0jfcTAo" alt=""><figcaption><p>[Image 1] Start model training with samples to measure performance</p></figcaption></figure>

## Where you can check Performance

The 'History' tap page lists up models you trained with the measured performances in order of most recent creation. \[Image 2]

<figure><img src="/files/xiX5AfXttLRfLq8WrOHq" alt=""><figcaption><p>[Image 2] Model history with each model measured its performance metrics</p></figcaption></figure>

If you press the 'details' button provided for each model, you can check the performance metric for each category \[Image 3]

<figure><img src="/files/UP6ezO561pPTgxIwEb1n" alt=""><figcaption><p>[Image 3] Detailed performance metric</p></figcaption></figure>


---

# 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.allganize.ai/nlu-api-dashboard/model-management/performance-metrics.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.