Scribd
Upload
35 views

M3 - Big Data Analytics With Cloud AI Platform Notebooks Slides

This document discusses AI Platform Notebooks and how they can be used with BigQuery. It begins by explaining that notebooks are interactive documents that contain code, outputs, and explanations, and that they support an iterative development process. It then discusses how notebooks can be launched on AI Platform with preconfigured machine learning frameworks. Finally, it explains that BigQuery queries can be run from notebooks using "magic" functions, and the results can be loaded into Pandas DataFrames for further analysis. Notebooks thus provide an integrated environment for querying data from BigQuery and processing it for machine learning.

Uploaded by

Deepak Kapoor
Copyright
© © All Rights Reserved
Available Formats
Download as PDF, TXT or read online on Scribd
35 views

M3 - Big Data Analytics With Cloud AI Platform Notebooks Slides

This document discusses AI Platform Notebooks and how they can be used with BigQuery. It begins by explaining that notebooks are interactive documents that contain code, outputs, and explanations, and that they support an iterative development process. It then discusses how notebooks can be launched on AI Platform with preconfigured machine learning frameworks. Finally, it explains that BigQuery queries can be run from notebooks using "magic" functions, and the results can be loaded into Pandas DataFrames for further analysis. Notebooks thus provide an integrated environment for querying data from BigQuery and processing it for machine learning.

Uploaded by

Deepak Kapoor
Copyright
© © All Rights Reserved
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 17

Big Data Analytics with

Cloud AI Platform
Notebooks
Agenda
What’s a Notebook

BigQuery Magic and Ties to


Pandas
Increasingly, data analysis and machine learning are
carried out in self-descriptive, shareable, executable
notebooks A typical
notebook
contains code,
Share

charts, and
Code explanations

Output

Markdown
Notebooks are developed in an iterative,
collaborative process
PHASE 5 PHASE 1 2

5
Share and Write code in
collaborate Python
1

Development
Process in 3

PHASE 4 Cloud Datalab PHASE 2

Write Run cell


commentary (Shift+Enter)
in markdown
4

PHASE 3

Examine Output
Spin up a JuypterLab instance, pre-configured with
the latest machine learning and data science
frameworks in one click.
AI Platform Notebooks uses the latest open-source
version of the industry-standard JupyterLab
You can easily change hardware including adding
and removing GPUs
Use any GCE instance type. You can pick the
hardware that makes sense, and scale up or down as
needed
You can even add and remove GPUs
Notebook instances are standard GCE instances
that live in your projects
How does it work?

Hosted on
DLVM - Compute
JupyterLab
Engine
Laptop
Read/Write
Pull/Push
Data
Code

Notebook files Data


Cloud repository Sources
Agenda
What’s a Notebook

BigQuery Magic and Ties to


Pandas
You can execute BigQuery commands from AI
Platform Notebooks Jupyter “magic” function

● Useful for checking query


validity
● Viewing query output
● But… can’t use query
output for anything
Can use the BigQuery API in Notebooks to return
query results as a Pandas DataFrame

Pandas DataFrame
Pandas + BigQuery in Notebook rocks!
Lab
BigQuery in Jupyter Labs
on AI Platform
Objectives

● Instantiate a Jupyter notebook on AI Platform


● Execute a BigQuery query from within a Jupyter notebook and process the
output using Pandas
Module Summary
● AI Platform Notebooks are ideal for prototyping
machine learning pipelines and models
● Notebooks integrate nicely with BigQuery and other
GCP services

You might also like