Hello video data: Train an AutoML video classification model

Use the Google Cloud console to train an AutoML video classification model. After your dataset is created and data is imported, use the Google Cloud console to review the training videos, and begin model training.

This tutorial has several pages:

1. Setting up your project.

2. Creating a video classification dataset.

3. Training an AutoML video classification model.

4. Deploying the model for batch predictions.

5. Cleaning up your project.

Each page assumes that you've already performed the instructions from the previous pages of the tutorial.

Review imported videos

The email you received informing you that importing of the file is complete has a link to the Browse tab for the dataset containing an annotation set. This annotation set contains the data from the imported file. Alternatively, click Go to the Datasets page and select the appropriate dataset.

Begin AutoML model training

To begin training,

1. Select Train new model in the right panel.
   

   

2. Choose training method. Defaults to AutoML. Leave as is.

   

3. Click continue.

4. Edit Model name. Note: If you do not provide a name for your model, one will be assigned based on the dataset name.

   

5. Select Train new model. A "demo_data" training job appears under "Training jobs and models" in the right panel.

   

Training takes several hours for this demo dataset. You will receive an email letting you know what the training time is estimated to be around 15 minutes after training starts.

To cancel ongoing training of a model, first click the Training tab in the content navigation menu, then click  View more >  Cancel training pipeline. The pipeline is canceled in a couple of minutes.

To check the current training status, click the training job "demo_data" in the panel on the right. A model property page appears.

You will also get another notification email once the model training finishes.

"Training time(actual)" is the time used to calculate the training cost and not the time spent training.

Check evaluation results

After training is completed, select your demo_data model to check the model evaluation results.

This takes you to the Evaluate tab where you can view model performance metrics.

What's next

Follow the next page of this tutorial to deploy your trained AutoML model for batch prediction.