- From the Compute Engine section of the Cloud Console, choose VM instances and create a new virtual machine.
- Select 2 vCPUs and make sure to select "Allow full access to all Cloud APIs"
- SSH to the new instance.
- In the terminal, install git so that you can clone the source repository:
sudo apt-get -y install git
- Then, clone the source repository:
git clone https://github.com/GoogleCloudPlatform/training-data-analyst
- Go to the directory containing demo files and install python packages we need:
cd training-data-analyst/courses/bdml_fundamentals
./install_missing.sh
- Now, ingest data on recent earthquakes from the USGS:
./ingest.sh
- Transform the raw data into an image:
./transform.py
- Go to the Storage | Browser in the Cloud Console and create a new bucket
- Copy the files to the bucket:
gsutil cp earthquakes.* gs://[YOURBUCKET] - Refresh the Storage | Browser to verify that you have new files in Cloud Storage
- Edit the bucket permissions and add a new member named
allUsersand give this member Cloud Storage Object Viewer permissions - Visit https://storageapis.google.com/BUCKET/earthquakes.htm