What you need

You must have completed Lab 0 and have the following:

• Logged into GCP Console with your Qwiklabs generated account
• launched Datalab and cloned the training-data-analyst repo

What you learn

In this lab, you will define a 2-layer rnn model to find next value of a time series

Step 1

In Cloud Datalab, click on the Home icon, and then navigate to notebooks/training-data-analyst/courses/machine_learning/deepdive/09_sequence/labs and open sinewaves.ipynb.

Step 2

In Datalab, click on Clear | All Cells. Now read the narrative and execute each cell in turn until the section "Train model locally"

Step 3

Open the model file model.py located in notebooks/training-data-analyst/courses/machine_learning/deepdive/09_sequence/labs/sinemodel and complete the code for the rnn2 model

Step 4

Back in the sinewaves.ipynb notebook, run the remaining cells. Be sure to specify --model correctly as rnn2

Step 5

Record your observation in the result table

©Google, Inc. or its affiliates. All rights reserved. Do not distribute.