Skip to content

Commit

Permalink
example de/serialization
Browse files Browse the repository at this point in the history
  • Loading branch information
@tabathadelane
tabathadelane committed Mar 29, 2024
1 parent d2259f0 commit 06608b7
Show file tree
Hide file tree
Showing 3 changed files with 121 additions and 63 deletions.
84 changes: 21 additions & 63 deletions src/content/docs/mlops/bring-your-own/mlops-byo.mdx
View file
Original file line number Diff line number Diff line change
Expand Up @@ -5,78 +5,36 @@ redirects:
- /docs/alerts-applied-intelligence/mlops/bring-your-own/mlops-byo
freshnessValidatedDate: never
---
import modelperformancemonitoringPredictions from 'images/model-performance-monitoring_screenshot-full_predictions.webp'

import modelperformancemonitoringPerformanceDrift from 'images/model-performance-monitoring_screenshot-full_performance-drift.webp'
New Relic's MLOps solution provides a Python library that makes it easy to monitor your machine learning models. New Relic Bring Your Own (BYO) <DoNotTranslate>**an inline DNT**</DoNotTranslate> is an offering of the New Relic product suite.

import modelperformancemonitoringModelFeatures from 'images/model-performance-monitoring_screenshot-full_model-features.webp'
## UL list with no whitespace
To create a Jira destination follow these steps:

1. Go to <DoNotTranslate>**[one.newrelic.com > All capabilities](https://one.newrelic.com/all-capabilities) > Alerts & AI**</DoNotTranslate>, click <DoNotTranslate>**Destinations**</DoNotTranslate>, and then select <DoNotTranslate>**Jira**</DoNotTranslate>.
2. Enter the following information:

import modelperformancemonitoringFeatureDistribution from 'images/model-performance-monitoring_screenshot-full_feature-distribution.webp'
* <DoNotTranslate>**Name:**</DoNotTranslate> Custom name to identify the destination.
* <DoNotTranslate>**URL:**</DoNotTranslate> The URL of the destination.
* <DoNotTranslate>**Username:**</DoNotTranslate> The email address of the user.
* <DoNotTranslate>**API token:**</DoNotTranslate> generated from your Atlassian account.

import modelperformancemonitoringStandardDeviation from 'images/model-performance-monitoring_screenshot-full_standard-deviation.webp'
## UL list with whitespace!!

import modelperformancemonitoringModelPerformanceTab from 'images/model-performance-monitoring_screenshot-full_model-performance-tab.webp'
To create a Jira destination follow these steps:

1. Go to <DoNotTranslate>**[one.newrelic.com > All capabilities](https://one.newrelic.com/all-capabilities) > Alerts & AI**</DoNotTranslate>, click <DoNotTranslate>**Destinations**</DoNotTranslate>, and then select <DoNotTranslate>**Jira**</DoNotTranslate>.
2. Enter the following information:

* <DoNotTranslate>**Name:**</DoNotTranslate> Custom name to identify the destination.

New Relic's MLOps solution provides a Python library that makes it easy to monitor your machine learning models. New Relic Bring Your Own (BYO) <DoNotTranslate>**Machine Learning Model Monitoring**</DoNotTranslate> is an offering of the New Relic product suite that allows users to monitor the performance of their own machine learning models, regardless of the platform or framework they are using. This can help users identify issues with their models and take corrective action before those issues impact the accuracy or performance of their applications.
* <DoNotTranslate>**URL:**</DoNotTranslate> The URL of the destination.

In an increasingly complex digital space, data teams rely heavily on prediction engines to make decisions. New Relic Machine Learning Model Monitoring allows your team to take a step back and look at the whole picture. Our model performance monitoring allows your team to examine your ML models to identify issues efficiently and make decisions by leveraging NR core features like setting alerts on any metrics and get notified once your ML performance has changed
* <DoNotTranslate>**Username:**</DoNotTranslate> The email address of the user.

* <DoNotTranslate>**API token:**</DoNotTranslate> generated from your Atlassian account.

## Easy to Integrate [#integrate]
The [ml-performance-monitoring](https://github.com/newrelic-experimental/ml-performance-monitoring) python package, based on [newrelic-telemetry-sdk-python](https://github.com/newrelic/newrelic-telemetry-sdk-python)​​, allows you to send your model's features and prediction values, as well as custom metrics, by simply adding a few lines to your code.

Use the python package to send the following types of data to New Relic:
## a DNT component as a block element

* <DoNotTranslate>**Inference data:**</DoNotTranslate> Stream your model’s features and prediction values. Inference data is streamed as a [custom event](/docs/data-apis/ingest-apis/introduction-event-api/) named InferenceData.
* <DoNotTranslate>**Data metrics:**</DoNotTranslate> Instead of sending all your raw inference data, select the aggregated statistics option over the features and predictions (for example, min, max, average, or percentile). They will be sent automatically as [metrics](/docs/data-apis/ingest-apis/metric-api/introduction-metric-api/).
* <DoNotTranslate>**Custom metrics:**</DoNotTranslate> Calculate your own metrics, monitor your model performance or model data, and stream them to New Relic using the record_metrics function. They will be sent as [metrics](/docs/data-apis/ingest-apis/metric-api/introduction-metric-api/).

## View Model Predictions [#view-predictions]
By monitoring the performance of your machine learning models, you can identify any issues that may be impacting their accuracy or performance. You'll see vital information in your dashboard like your total model predictions and distribution over time. Additionally, you can set <InlinePopover type="alerts" /> to track any changes in total predictions or high/low picks.

<img
title="An image of the model performance and predictions in New Relic"
alt="An image of the New Relic MLOps dashboard."
src={modelperformancemonitoringPredictions}
/>

## Detect Model and Data Drift [#detect-drift]
With real time monitoring, you can quickly identify issues with your machine learning models by detecting model and data drift. Model and data drift are changes in your real world environment that can affect the predictive power of your model. By getting real-time data on your model and data drift, you can quickly take corrective action before those issues impact the performance of your applications.

<img
title="An image of the model and data drift in New Relic"
alt="An image of the New Relic MLOps dashboard."
src={modelperformancemonitoringPerformanceDrift}
/>

## Dive Deeper into Model Features [#view-model-features]
Get insight and statistical data on your features and get alerted on any deviation from standard and expected behavior.

<img
title="An image of the model features in New Relic"
alt="An image of the New Relic MLOps dashboard."
src={modelperformancemonitoringModelFeatures}
/>

<img
title="An image of the feature distribution in New Relic"
alt="An image of the New Relic MLOps dashboard."
src={modelperformancemonitoringFeatureDistribution}
/>

<img
title="An image of the model count, missing values, and standard deviation in New Relic"
alt="An image of the New Relic MLOps dashboard."
src={modelperformancemonitoringStandardDeviation}
/>

## How to start monitoring your ML model performance [#get-benefits]

First, use the [getting started](/docs/mlops/bring-your-own/getting-started-byo/) documentation to start streaming your model data. Then, click on <DoNotTranslate>**Model Performance**</DoNotTranslate> in the <DoNotTranslate>**All Capabilities**</DoNotTranslate> page in the New Relic UI (don't forget to pin it), and you will see your model performance view based on the data your model is sending.

<img
title="An image of the model performance tab in the New Relic navigation menu"
alt="An image of the New Relic MLOps dashboard."
src={modelperformancemonitoringModelPerformanceTab}
/>
<DoNotTranslate>This is a whole paragraph that is wrapped in a DNT component. This is a whole paragraph that is wrapped in a DNT component. This is a whole paragraph that is wrapped in a DNT component. This is a whole paragraph that is wrapped in a DNT component. This is a whole paragraph that is wrapped in a DNT component. See you on the other siiide.</DoNotTranslate>
49 changes: 49 additions & 0 deletions test.html
View file
Original file line number Diff line number Diff line change
@@ -0,0 +1,49 @@

<div data-type="frontmatter" data-value="eyJ0aXRsZSI6IkJyaW5nIHlvdXIgb3duIGRhdGEiLCJtZXRhRGVzY3JpcHRpb24iOiJCcmluZyB5b3VyIG93biBNTCBtb2RlbCB0ZWxlbWV0cnkgaW50byBOZXcgUmVsaWMsIGFuZCBzdGFydCBnZXR0aW5nIHZhbHVlIGZyb20geW91ciBNTCBtb2RlbCBkYXRhIGluIGp1c3QgYSBmZXcgbWludXRlcyIsInJlZGlyZWN0cyI6WyIvZG9jcy9hbGVydHMtYXBwbGllZC1pbnRlbGxpZ2VuY2UvbWxvcHMvYnJpbmcteW91ci1vd24vbWxvcHMtYnlvIl0sImZyZXNobmVzc1ZhbGlkYXRlZERhdGUiOiJuZXZlciJ9">
<div data-key="title">Bring your own data</div>
</div>
<p>New Relic's MLOps solution provides a Python library that makes it easy to monitor your machine learning models. New Relic Bring Your Own (BYO) <span data-type="component" data-component="DoNotTranslate" class="notranslate"><strong>an inline DNT</strong></span> is an offering of the New Relic product suite.</p>
<h2>UL list with no whitespace</h2>
<p>To create a Jira destination follow these steps:</p>
<ol>
<li>
<p>Go to <span data-type="component" data-component="DoNotTranslate" class="notranslate"><strong><a href="https://one.newrelic.com/all-capabilities">one.newrelic.com > All capabilities</a> > Alerts &#x26; AI</strong></span>, click <span data-type="component" data-component="DoNotTranslate" class="notranslate"><strong>Destinations</strong></span>, and then select <span data-type="component" data-component="DoNotTranslate" class="notranslate"><strong>Jira</strong></span>.</p>
</li>
<li>
<p>Enter the following information:</p>
<ul>
<li><span data-type="component" data-component="DoNotTranslate" class="notranslate"><strong>Name:</strong></span> Custom name to identify the destination.</li>
<li><span data-type="component" data-component="DoNotTranslate" class="notranslate"><strong>URL:</strong></span> The URL of the destination.</li>
<li><span data-type="component" data-component="DoNotTranslate" class="notranslate"><strong>Username:</strong></span> The email address of the user.</li>
<li><span data-type="component" data-component="DoNotTranslate" class="notranslate"><strong>API token:</strong></span> generated from your Atlassian account.</li>
</ul>
</li>
</ol>
<h2>UL list with whitespace!!</h2>
<p>To create a Jira destination follow these steps:</p>
<ol>
<li>
<p>Go to <span data-type="component" data-component="DoNotTranslate" class="notranslate"><strong><a href="https://one.newrelic.com/all-capabilities">one.newrelic.com > All capabilities</a> > Alerts &#x26; AI</strong></span>, click <span data-type="component" data-component="DoNotTranslate" class="notranslate"><strong>Destinations</strong></span>, and then select <span data-type="component" data-component="DoNotTranslate" class="notranslate"><strong>Jira</strong></span>.</p>
</li>
<li>
<p>Enter the following information:</p>
<ul>
<li>
<p><span data-type="component" data-component="DoNotTranslate" class="notranslate"><strong>Name:</strong></span> Custom name to identify the destination.</p>
</li>
<li>
<p><span data-type="component" data-component="DoNotTranslate" class="notranslate"><strong>URL:</strong></span> The URL of the destination.</p>
</li>
<li>
<p><span data-type="component" data-component="DoNotTranslate" class="notranslate"><strong>Username:</strong></span> The email address of the user.</p>
</li>
<li>
<p><span data-type="component" data-component="DoNotTranslate" class="notranslate"><strong>API token:</strong></span> generated from your Atlassian account.</p>
</li>
</ul>
</li>
</ol>
<h2>a DNT component as a block element</h2>
<div data-type="component" data-component="DoNotTranslate" class="notranslate">
<p>This is a whole paragraph that is wrapped in a DNT component. This is a whole paragraph that is wrapped in a DNT component. This is a whole paragraph that is wrapped in a DNT component. This is a whole paragraph that is wrapped in a DNT component. This is a whole paragraph that is wrapped in a DNT component. See you on the other siiide.</p>
</div>
51 changes: 51 additions & 0 deletions test.mdx
View file
Original file line number Diff line number Diff line change
@@ -0,0 +1,51 @@
---
title: Bring your own data
metaDescription: 'Bring your own ML model telemetry into New Relic, and start getting value from your ML model data in just a few minutes'
freshnessValidatedDate: never
translationType: !<tag:yaml.org,2002:js/undefined> ''
---

New Relic's MLOps solution provides a Python library that makes it easy to monitor your machine learning models. New Relic Bring Your Own (BYO) <DoNotTranslate>**an inline DNT**</DoNotTranslate> is an offering of the New Relic product suite.

## UL list with no whitespace

To create a Jira destination follow these steps:

1. Go to <DoNotTranslate>**[one.newrelic.com > All capabilities](https://one.newrelic.com/all-capabilities) > Alerts & AI**</DoNotTranslate>, click <DoNotTranslate>**Destinations**</DoNotTranslate>, and then select <DoNotTranslate>**Jira**</DoNotTranslate>.

2. Enter the following information:

* <DoNotTranslate>**Name:**</DoNotTranslate>

Custom name to identify the destination.

* <DoNotTranslate>**URL:**</DoNotTranslate>

The URL of the destination.

* <DoNotTranslate>**Username:**</DoNotTranslate>

The email address of the user.

* <DoNotTranslate>**API token:**</DoNotTranslate>

generated from your Atlassian account.

## UL list with whitespace!!

To create a Jira destination follow these steps:

1. Go to <DoNotTranslate>**[one.newrelic.com > All capabilities](https://one.newrelic.com/all-capabilities) > Alerts & AI**</DoNotTranslate>, click <DoNotTranslate>**Destinations**</DoNotTranslate>, and then select <DoNotTranslate>**Jira**</DoNotTranslate>.

2. Enter the following information:

* <DoNotTranslate>**Name:**</DoNotTranslate> Custom name to identify the destination.
* <DoNotTranslate>**URL:**</DoNotTranslate> The URL of the destination.
* <DoNotTranslate>**Username:**</DoNotTranslate> The email address of the user.
* <DoNotTranslate>**API token:**</DoNotTranslate> generated from your Atlassian account.

## a DNT component as a block element

<DoNotTranslate>
This is a whole paragraph that is wrapped in a DNT component. This is a whole paragraph that is wrapped in a DNT component. This is a whole paragraph that is wrapped in a DNT component. This is a whole paragraph that is wrapped in a DNT component. This is a whole paragraph that is wrapped in a DNT component. See you on the other siiide.
</DoNotTranslate>

0 comments on commit 06608b7

Please sign in to comment.