Try these codelabs to learn hands-on how Firebase can help you use TensorFlow
Lite models more easily and effectively.

## Sentiment analysis

![Screenshot of sentiment analysis app](https://firebase.google.com/docs/ml/images/sentiment_codelab.png)

In this codelab, you use your own training data to fine-tune an existing text
classification model that identifies the sentiment expressed in a passage of
text. Then, you deploy the model using Firebase ML and compare the accuracy
of the old and new models with A/B Testing.

[iOS+](https://firebase.google.com/codelabs/textclassification-iOS)
[Android](https://firebase.google.com/codelabs/textclassification-android)

## Content recommendation

![Screenshot of content recommendation app](https://firebase.google.com/docs/ml/images/recommendation_codelab.png)

Recommendation engines let you personalize experiences to individual users,
presenting them with more relevant and engaging content. Rather than building
out a complex pipeline to power this feature, this codelab shows how you can
implement a content recommendation engine for an app by training and deploying
an on-device ML model.

[iOS+](https://firebase.google.com/codelabs/contentrecommendation-ios)
[Android](https://firebase.google.com/codelabs/contentrecommendation-android)