> For the complete documentation index, see [llms.txt](https://docs.enginsight.com/docs/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://docs.enginsight.com/docs/manual/english/platform-usage/hosts/tracer/models/model-detail-view.md).

# Model: Detail View

Each Machine Learning (ML)-based model monitors and analyzes a specific metric on a specific host. The metric, how it is aggregated and shown, and which host is monitored are defined through previously created [Tracers](https://docs.enginsight.com/docs/manual/english/platform-usage/hosts/tracer.md).

<i class="fa-computer-mouse">:computer-mouse:</i> In the [model overview](https://docs.enginsight.com/docs/manual/english/platform-usage/hosts/tracer/models.md#models-overview), click the **Name** column of a list item to open the respective ML model.

{% hint style="warning" %}
**Please note**: At least seven days of data are required for an ML model to identify temporal patterns and detect anomalies based on them. We recommend a training period of at least four weeks to establish a meaningful data basis.
{% endhint %}

{% hint style="danger" %}
**Please note**: If the data horizon for your ML models is still shown as insufficient (**Insufficient**) after several days, the configured retention period (Time to Live; TTL) for log entries in the Enginsight component Loggernaut might be too short. In this case, you must extend the TTL accordingly.
{% endhint %}

<figure><img src="https://2423388680-files.gitbook.io/~/files/v0/b/gitbook-x-prod.appspot.com/o/spaces%2FV7OrHgKXA0t1jankQBYf%2Fuploads%2Fead2JHsUZ7bm3OXaNZqf%2Fenginsight-screenshot-hosts-tracer-models-detail%20view.jpg?alt=media&amp;token=49d664e7-c3d5-4bcf-ae79-1caffadb6a24" alt=""><figcaption></figcaption></figure>

The ML model's time series analyses are divided into two views that can be accessed using tabs:

* [Graphs](#tab-graphs)
* [Anomalies](#tab-anomalies)

You can also configure [advanced settings](#advanced-settings) for the time series analysis.

***

## General Settings

Here you can view various basic settings.

{% hint style="warning" %}
**Please note**: Since each ML model is based on a Tracer, these values can only be adjusted in the respective [tracer settings](https://docs.enginsight.com/docs/manual/english/platform-usage/hosts/tracer/tracers/add-tracer.md#general-settings).
{% endhint %}

<figure><img src="https://2423388680-files.gitbook.io/~/files/v0/b/gitbook-x-prod.appspot.com/o/spaces%2FV7OrHgKXA0t1jankQBYf%2Fuploads%2FaSPHKY8rJR4F5xI7fodw%2Fenginsight-screenshot-hosts-tracer-models-detail%20view-general%20settings.jpg?alt=media&amp;token=41dcab2f-3794-4aec-89f6-4c7070cdc4f6" alt=""><figcaption></figcaption></figure>

The following information is shown:

<table><thead><tr><th width="200.47265625">Information</th><th>Description</th></tr></thead><tbody><tr><td>Name</td><td>Shows the name of the Tracer on which the model is based.</td></tr><tr><td>Hosts</td><td>Shows the name of the host that is monitored and analyzed by the model.</td></tr><tr><td>Field Name</td><td><p>Shows the name of the monitored metric, including the aggregation of the value defined in the tracer, in the format <strong>[aggregation abbreviation](field name)</strong>.<br><br>The following aggregation abbreviations can be shown:</p><ul><li><strong>avg</strong>: The average value per measurement is shown.</li><li><strong>min</strong>: The minimum value per measurement is shown.</li><li><strong>max</strong>: The maximum value per measurement is shown.</li><li><strong>sum</strong>: The sum of all measured values is shown.</li></ul></td></tr></tbody></table>

***

## Start and End Time

<i class="fa-computer-mouse">:computer-mouse:</i> Click <i class="fa-calendar-days">:calendar-days:</i> **Start Time** or <i class="fa-calendar-days">:calendar-days:</i> **End Time** in the upper-right corner to define the start or end of the period to be analyzed. The date picker opens.

{% tabs %}
{% tab title="Relative Time Period" %}
Define a **Relative Time Period** manually. You can also select a relative time range from one of the available **Templates**.

<div align="left"><figure><img src="https://2423388680-files.gitbook.io/~/files/v0/b/gitbook-x-prod.appspot.com/o/spaces%2FV7OrHgKXA0t1jankQBYf%2Fuploads%2FrrYJOLp7HuEQorpFUCX2%2FEnginsight-EN-Date%20Picker-Relative%20Time%20Period.jpg?alt=media&amp;token=1018f4bd-8137-4bba-a2c8-8fd8202538a4" alt="" width="235"><figcaption></figcaption></figure></div>

<i class="fa-computer-mouse">:computer-mouse:</i> Click the field with the default setting **Days Ago** to expand a list of all available time ranges. Select a time range.

In the number field, enter the number of seconds, minutes, hours, days, or weeks you want to show. The default value is **1**.

<i class="fa-computer-mouse">:computer-mouse:</i> Click **Apply** to confirm your settings.
{% endtab %}

{% tab title="Absolute Time Period" %}
Select a date from the calendar view.

<div align="left"><figure><img src="https://2423388680-files.gitbook.io/~/files/v0/b/gitbook-x-prod.appspot.com/o/spaces%2FV7OrHgKXA0t1jankQBYf%2Fuploads%2Fp2GTtoSJ59r7MIYreBFZ%2FEnginsight-EN-Date%20Picker-Absolute%20Time%20Period.jpg?alt=media&amp;token=e00bb0e2-c7ed-4ddb-9629-e02b34df20e6" alt="" width="232"><figcaption></figcaption></figure></div>

<i class="fa-computer-mouse">:computer-mouse:</i> Optionally, click <i class="fa-clock">:clock:</i> **Time** to set an exact time in the 24-hour format using the dial. First, select an hour. The dial then switches to the minute view. Select the required minutes.

<i class="fa-computer-mouse">:computer-mouse:</i> Click **Apply** to confirm your settings.
{% endtab %}
{% endtabs %}

<i class="fa-computer-mouse">:computer-mouse:</i> Click <i class="fa-arrow-rotate-left">:arrow-rotate-left:</i> **Reset** in the upper-right corner to reset the settings for the selected time range.

***

## Tab: Graphs

The graphs show deviations between the actual measured value ([Time Series Progression](#time-series-progression)) and the calculated variance value ([Time Series Variation](#time-series-variation)) and the calculated baseline range of the monitored metric.

<figure><img src="https://2423388680-files.gitbook.io/~/files/v0/b/gitbook-x-prod.appspot.com/o/spaces%2FV7OrHgKXA0t1jankQBYf%2Fuploads%2F4GxprMLa8eTxJWSzYVXN%2Fenginsight-screenshot-siem-tracers-models-detail%20view-tab%20graphs.jpg?alt=media&amp;token=06f210f2-d944-4f42-ba80-60c2882bb64b" alt=""><figcaption></figcaption></figure>

### Time Series Progression

The time series progression shows the actual measured values of the monitored metric within the selected period. It is visualized using three graphs, with the [Time Series](#time-series) graph visible by default.

<i class="fa-eye">:eye:</i> The [Standard Deviation](#standard-deviation) and [Score](#score) graphs become visible only after you click **Show Details** below the first graph.

All graphs mark the ML model's **Retrain** time using a dashed black line. At this point, the model updates its value calculation to improve detection accuracy.

{% tabs %}
{% tab title="Time Series" %}

<figure><img src="https://2423388680-files.gitbook.io/~/files/v0/b/gitbook-x-prod.appspot.com/o/spaces%2FV7OrHgKXA0t1jankQBYf%2Fuploads%2FgL9wDYIToy5w6tWCnzkR%2Fenginsight-screenshot-siem-tracers-models-detail%20view-tab%20graphs-time%20series%20progression-time%20series.jpg?alt=media&amp;token=111f056c-5ff0-46d6-b2c9-68028f89f412" alt=""><figcaption></figcaption></figure>

The line chart shows the progression of the actual measured values within the selected period.\
\
The gray shaded area represents the expected value range calculated by the model. All values measured outside this range are detected as anomalies and [color-coded](https://docs.enginsight.com/docs/manual/english/appendix/color-coding-of-severity-levels.md) accordingly.\
\ <i class="fa-arrow-pointer">:arrow-pointer:</i> Hover over the line to show the actual measured value compared with the model's prediction at a specific point in time.
{% endtab %}

{% tab title="Standard Deviation" %} <i class="fa-eye">:eye:</i> This graph is visible only if you click **Show Details** below the **Time Series** graph.

<figure><img src="https://2423388680-files.gitbook.io/~/files/v0/b/gitbook-x-prod.appspot.com/o/spaces%2FV7OrHgKXA0t1jankQBYf%2Fuploads%2FpawrXwYaJFxx5EcN1uwR%2Fenginsight-screenshot-siem-tracers-models-detail%20view-tab%20graphs-time%20series%20progression-standard%20deviation.jpg?alt=media&amp;token=72660093-bcdb-4f70-8b8b-6c6834b643c3" alt=""><figcaption></figcaption></figure>

The line chart shows how strongly the actual measured values deviate from the expected average calculated by the model. The average corresponds to the value **0** on the y-axis and is visualized by a gray dashed line.

A high standard deviation indicates irregular or unusual behavior of the monitored metric.

<i class="fa-arrow-pointer">:arrow-pointer:</i> Hover over the line to show the calculated standard deviation at a specific point in time.
{% endtab %}

{% tab title="Score" %} <i class="fa-eye">:eye:</i> This graph is visible only if you click **Show Details** below the **Time Series** graph.

<figure><img src="https://2423388680-files.gitbook.io/~/files/v0/b/gitbook-x-prod.appspot.com/o/spaces%2FV7OrHgKXA0t1jankQBYf%2Fuploads%2Fp0qNZzcLrd33GDsbwr0E%2Fenginsight-screenshot-siem-tracers-models-detail%20view-tab%20graphs-time%20series%20progression-score.jpg?alt=media&amp;token=8e5e3a42-16ff-438a-9764-ee33ad172ed5" alt=""><figcaption></figcaption></figure>

The line chart shows the anomaly score (**Score**) calculated by the model based on the actual measured values in relation to thresholds with different severity levels. The thresholds indicate the anomaly score from which an anomaly is considered relevant.\
\
The following thresholds can be shown:

* **Critical Threshold**: If the value exceeds or falls below this threshold, the anomaly has a very high severity.
* **High Threshold**: If the value exceeds or falls below this threshold, the anomaly has a high severity.
* **Medium Threshold**: If the value exceeds or falls below this threshold, the anomaly has a medium severity.
* **Low Threshold**: If the value exceeds or falls below this threshold, the anomaly has a low severity.

<i class="fa-computer-mouse">:computer-mouse:</i> Click the name of a value to show or hide the respective line in the chart.

{% hint style="info" %}
By default, the score, the critical threshold, and the thresholds that have been exceeded or fallen below are shown.
{% endhint %}

<i class="fa-arrow-pointer">:arrow-pointer:</i> Hover over the line to show the calculated anomaly score and the critical threshold at a specific point in time.
{% endtab %}
{% endtabs %}

### Time Series Variation

The time series variation shows the calculated variance value of the monitored metric within the selected period. This is an abstract value that indicates how much the measured data varies within a time interval.

The time series variation is shown using three graphs, with the [Time Series](#time-series-1) graph visible by default.

<i class="fa-eye">:eye:</i> The [Standard Deviation](#standard-deviation-1) and [Score](#score-1) graphs become visible only after you click **Show Details** below the first graph.

{% tabs %}
{% tab title="Time Series" %}

<figure><img src="https://2423388680-files.gitbook.io/~/files/v0/b/gitbook-x-prod.appspot.com/o/spaces%2FV7OrHgKXA0t1jankQBYf%2Fuploads%2F9Z8yFQBR4kI7KLmt2Sew%2Fenginsight-screenshot-siem-tracers-models-detail%20view-tab%20graphs-time%20series%20variation-time%20series.jpg?alt=media&amp;token=5830bd76-0691-4030-80c8-b67ec7d3d07b" alt=""><figcaption></figcaption></figure>

The line chart shows the progression of the calculated variance within the selected period. These are not actual measured values, but an abstract value that indicates how much the measured data varies within a time interval. The greater the variation in the data, the higher the displayed variance value.\
\
The gray shaded area represents the expected value range calculated by the model. All values measured outside this range are detected as anomalies and [color-coded](https://docs.enginsight.com/docs/manual/english/appendix/color-coding-of-severity-levels.md) accordingly.\
\ <i class="fa-arrow-pointer">:arrow-pointer:</i> Hover over the line to show the calculated variance value compared with the model's prediction at a specific point in time.
{% endtab %}

{% tab title="Standard Deviation" %} <i class="fa-eye">:eye:</i> This graph is visible only if you click **Show Details** below the **Time Series** graph.

<figure><img src="https://2423388680-files.gitbook.io/~/files/v0/b/gitbook-x-prod.appspot.com/o/spaces%2FV7OrHgKXA0t1jankQBYf%2Fuploads%2FNaiwfZsOxUTk1QxPAoDR%2Fenginsight-screenshot-siem-tracers-models-detail%20view-tab%20graphs-time%20series%20variation-standard%20deviation.jpg?alt=media&amp;token=ff7f1102-2c18-4833-982c-16e3b47c7e5a" alt=""><figcaption></figcaption></figure>

The line chart shows how strongly the calculated variance value deviates from the expected average calculated by the model. The average corresponds to the value **0** on the y-axis and is visualized by a gray dashed line.\
\ <i class="fa-arrow-pointer">:arrow-pointer:</i> Hover over the line to show the calculated standard deviation at a specific point in time.
{% endtab %}

{% tab title="Score" %} <i class="fa-eye">:eye:</i> This graph is visible only if you click **Show Details** below the **Time Series** graph.

<figure><img src="https://2423388680-files.gitbook.io/~/files/v0/b/gitbook-x-prod.appspot.com/o/spaces%2FV7OrHgKXA0t1jankQBYf%2Fuploads%2Ft8kQip5CBFibuBzKvfLX%2Fenginsight-screenshot-siem-tracers-models-detail%20view-tab%20graphs-time%20series%20variation-score.jpg?alt=media&amp;token=cc43c331-7a4d-4145-85e3-49d84e548c18" alt=""><figcaption></figcaption></figure>

The line chart shows the anomaly score (**Score**) calculated by the model based on the variance value in relation to thresholds with different severity levels. The thresholds indicate the anomaly score from which an anomaly is considered relevant.\
\
The following thresholds can be shown:

* **Critical Threshold**: If the value exceeds or falls below this threshold, the anomaly has a very high severity.
* **High Threshold**: If the value exceeds or falls below this threshold, the anomaly has a high severity.
* **Medium Threshold**: If the value exceeds or falls below this threshold, the anomaly has a medium severity.
* **Low Threshold**: If the value exceeds or falls below this threshold, the anomaly has a low severity.

<i class="fa-computer-mouse">:computer-mouse:</i> Click the name of a value to show or hide the respective line in the chart.

{% hint style="info" %}
By default, the score, the critical threshold, and the thresholds that have been exceeded or fallen below are shown.
{% endhint %}

<i class="fa-arrow-pointer">:arrow-pointer:</i> Hover over the line to show the calculated anomaly score and the critical threshold at a specific point in time.
{% endtab %}
{% endtabs %}

***

## Tab: Anomalies

Navigate to the **Anomalies** tab to show a list of all detected deviations from the baseline. If no anomaly has been detected yet, this view is empty.

<figure><img src="https://2423388680-files.gitbook.io/~/files/v0/b/gitbook-x-prod.appspot.com/o/spaces%2FV7OrHgKXA0t1jankQBYf%2Fuploads%2Fk54wB76vnlWpb3bDmkar%2Fenginsight-screenshot-siem-tracers-models-detail%20view-tab%20anomalies.jpg?alt=media&amp;token=8dbd6660-b446-4deb-9f4d-48d81ab1da5f" alt=""><figcaption></figcaption></figure>

### Navigation

<i class="fa-magnifying-glass">:magnifying-glass:</i> Click the **Text Search** field above the list and enter a search term of your choice to find specific list items more quickly.

<i class="fa-computer-mouse">:computer-mouse:</i> Click the **Refresh** icon <i class="fa-arrows-rotate-reverse">:arrows-rotate-reverse:</i> above the list to update all list items.

<i class="fa-computer-mouse-scrollwheel">:computer-mouse-scrollwheel:</i> Scroll down the list or adjust the number of results shown per page ![](https://2423388680-files.gitbook.io/~/files/v0/b/gitbook-x-prod.appspot.com/o/spaces%2FV7OrHgKXA0t1jankQBYf%2Fuploads%2FfjKACCt2SKXcFd3S7Xb5%2FEnginsight-EN-Results%20per%20Page.jpg?alt=media\&token=e1650160-7a41-4965-aaca-6975db758dbe) to view additional entries. Use the page navigation ![](https://content.gitbook.com/content/V7OrHgKXA0t1jankQBYf/blobs/wnxVw06QGoSsYjhwMJRF/Enginsight-Seitennavigation-2.jpg) to switch between pages.&#x20;

#### Sorting List Items

<i class="fa-computer-mouse">:computer-mouse:</i> Click the **Sort** icon <i class="fa-sort">:sort:</i> next to a column name in the table header to sort by that column in ascending <i class="fa-angle-up">:angle-up:</i> or descending order <i class="fa-chevron-down">:chevron-down:</i> .

<i class="fa-computer-mouse">:computer-mouse:</i> Click the sort icon <i class="fa-bars-sort">:bars-sort:</i> above the list to adjust the priority of the applied sorting. A menu opens that shows all columns by which the list is currently sorted. Drag and drop the shown items to change the sorting priority.

#### Filtering the List

<i class="fa-computer-mouse">:computer-mouse:</i> Click the **Filter** icon <i class="fa-filter">:filter:</i> above the list to filter the list by predefined parameters.

<details>

<summary>Available Filters</summary>

<table><thead><tr><th width="206.60546875">Parameter</th><th width="116.515625">Type</th><th>Description</th></tr></thead><tbody><tr><td>Resolved</td><td>Boolean</td><td><p>Filter by whether the cause of the anomaly has already been resolved.<br><br>The following values are available:</p><ul><li><strong>True</strong>: The cause of the anomaly has already been resolved.</li><li><strong>False</strong>: The cause of the anomaly has not yet been resolved.</li></ul></td></tr><tr><td>Source</td><td>String</td><td>Filter by whether the anomaly was detected in the time series progression (<strong>interval</strong>) or through a time series variation (<strong>variance</strong>).</td></tr><tr><td>Severity</td><td>String</td><td>Filter by a specific severity or exclude it from the results. All available severities are shown.</td></tr></tbody></table>

</details>

### Anomalies: Overview

The Anomalies overview is structured as a table. The following information is shown:

<table><thead><tr><th width="199.99609375">Column</th><th>Description</th></tr></thead><tbody><tr><td>Severity</td><td><p>Shows the severity of the anomaly.</p><p></p><p>The following severity levels can be shown:</p><ul><li><strong>Critical</strong>: The anomaly has a very high criticality.</li><li><strong>High</strong>: The anomaly has a high criticality.</li><li><strong>Medium</strong>: The anomaly has a medium criticality.</li><li><strong>Low</strong>: The anomaly has a low criticality.</li></ul></td></tr><tr><td>Name</td><td>Shows whether the anomaly is associated with a time series progression or a time series variation.</td></tr><tr><td>Started At</td><td>Shows the exact date and time when the anomaly began.</td></tr><tr><td>Ended At</td><td>Shows the exact date and time when the anomaly ended.</td></tr></tbody></table>

***

## Advanced Settings

<i class="fa-computer-mouse">:computer-mouse:</i> Click <i class="fa-sliders">:sliders:</i> **Advanced Settings** in the upper-right corner to enable or disable the ML model and configure additional anomaly detection settings.

<div align="left"><figure><img src="https://2423388680-files.gitbook.io/~/files/v0/b/gitbook-x-prod.appspot.com/o/spaces%2FV7OrHgKXA0t1jankQBYf%2Fuploads%2FTULWBqVluFQTh3uq6Qkm%2Fenginsight-screenshot-siem-tracers-models-add%20model-advanced%20settings.jpg?alt=media&amp;token=bcc179b4-4804-4487-9f5b-96fc0bada9ec" alt="" width="563"><figcaption></figcaption></figure></div>

The following options are available:

<table><thead><tr><th width="199.9296875">Option</th><th>Description</th></tr></thead><tbody><tr><td>Enabled</td><td><p>Enable the ML model to calculate baseline values and potential anomalies.<br><br><i class="fa-computer-mouse">:computer-mouse:</i> Select or deselect the checkbox <i class="fa-square">:square:</i> next to <strong>Enabled</strong> to enable or disable the option.</p><div data-gb-custom-block data-tag="hint" data-style="info" class="hint hint-info"><p>This option is enabled by default.</p></div></td></tr><tr><td>Time Series Progression: Bucket Size</td><td><p>Define the time span in minutes over which measured values within an interval are combined into a single data point.</p><div data-gb-custom-block data-tag="hint" data-style="info" class="hint hint-info"><p>The default time span is <strong>5 minutes</strong>.</p></div></td></tr><tr><td>Time Series Progression: Threshold for Anomaly Detection</td><td><p>Define the percentage deviation from the baseline within an interval at which a deviation is classified as an anomaly.<br><br>If the measured value exceeds or falls below this threshold, the ML model triggers an <a href="broken://pages/36a0764274766a9d13f26e9d949fc830e576e0a7#step-1-alert-type">action</a> after a user-defined grace period.</p><div data-gb-custom-block data-tag="hint" data-style="info" class="hint hint-info"><p>The default threshold is <strong>85%</strong>.</p></div></td></tr><tr><td>Time Series Progression: Grace Period</td><td><p>Define the time in minutes that the ML model waits before triggering an action to avoid false positives.</p><div data-gb-custom-block data-tag="hint" data-style="info" class="hint hint-info"><p>The default grace period is <strong>5 minutes</strong>.</p></div></td></tr><tr><td>Time Series Variation: Bucket Size</td><td><p>Define the time span in minutes over which measured values within an interval are combined into a single data point to calculate the fluctuation.</p><div data-gb-custom-block data-tag="hint" data-style="info" class="hint hint-info"><p>The default time span is <strong>5 minutes</strong>.</p></div></td></tr><tr><td>Time Series Variation: Threshold for Anomaly Detection</td><td><p>Define the percentage deviation from the baseline within an interval at which a deviation is classified as an anomaly.<br><br>If the measured value exceeds or falls below this threshold, the ML model triggers an action.</p><div data-gb-custom-block data-tag="hint" data-style="info" class="hint hint-info"><p>The default threshold is <strong>85%</strong>.</p></div></td></tr><tr><td>Time Series Variation: Grace Period</td><td><p>Define the time in minutes that the ML model waits before triggering an action to avoid false positives.</p><div data-gb-custom-block data-tag="hint" data-style="info" class="hint hint-info"><p>The default grace period is <strong>5 minutes</strong>.</p></div></td></tr></tbody></table>

<i class="fa-computer-mouse">:computer-mouse:</i> After configuring all settings, click **Apply** to save the changes.

<i class="fa-computer-mouse">:computer-mouse:</i> Click **Close** to leave the advanced settings without saving.

***
