WEBVTT

00:00.080 --> 00:00.920
Welcome back.

00:01.240 --> 00:06.920
I want from you to focus with me because in this video, we're gonna introduce something new.

00:06.960 --> 00:12.680
We're going to connect our TensorFlow model with our Android application here.

00:12.680 --> 00:22.640
Inside this app package, create a new others TensorFlow Lite model model location, search for model

00:22.640 --> 00:30.080
location and select create ML directory in main and automatically add build feature and dependencies

00:30.080 --> 00:31.480
to Build.gradle.

00:31.520 --> 00:40.600
We added TensorFlow dependency, but Android will add other libraries like support and metadata and

00:40.600 --> 00:44.240
also will add ML model binding.

00:44.280 --> 00:44.760
True.

00:44.880 --> 00:48.840
Automatically add TensorFlow Lite GPU dependencies to build greater.

00:48.840 --> 00:53.000
This is optional, but I don't prefer to use it with emulators.

00:53.000 --> 00:55.440
So click finish and here we go.

00:55.520 --> 00:58.600
Wait, Gradle project sync to finish.

00:58.600 --> 01:02.120
And you notice that ML package is created.

01:02.120 --> 01:07.680
And inside it we have our model house price model TF Lite.

01:07.720 --> 01:10.800
If you open it you can get this page.

01:10.840 --> 01:11.200
Okay.

01:11.240 --> 01:19.400
So this is how we add TensorFlow to our application and how we add TensorFlow model to the ML package.

01:19.560 --> 01:22.160
Go to Build.gradle file.

01:22.160 --> 01:28.080
And here we have TensorFlow support and metadata being added.

01:28.240 --> 01:35.240
And you see in the build features we have ML model binding equals to true being added.

01:35.240 --> 01:43.760
So in this way we added successfully the TensorFlow model to the Android Studio application.