Data Science on Google Cloud Platform: Exploratory Data Analytics

58mIntermediate2018-11-14

Authors

Kumaran Ponnambalam

Working with data for 20+ years

Course details

Cloud computing brings unlimited scalability and elasticity to data science applications. Expertise in the major platforms, such as Google Cloud Platform (GCP), is essential to the IT professional. This course—one of a series by cloud engineering specialist and data scientist Kumaran Ponnambalam—shows how to conduct exploratory data analytics with GCP. First, review the concepts of segmentation and profiling. Then get hands on, as you learn to perform both text and visual analysis of data using tools provided by GCP: Cloud Datalab, BigQuery, Cloud Dataflow, and Data Studio. Finally, look at an end-to-end use case that applies what you've learned in the course.

Learning objectives
Setting up Cloud DataLlb for exploratory data analytics
Segmentation and profiling
Reading and writing data from BigQuery
Managing cloud storage buckets
Creating visualizations of BigQuery data with the GCP Charting API
Managing Datalab instances

Skills covered

Google CloudSoftware Development ToolsGoogleData AnalysisCloud PlatformsCloud ComputingData ScienceBusiness Analysis and StrategyBusiness Software and ToolsSoftware DevelopmentOne-Off

Concepts

0. Introduction

01 - Why EDA on Datalab
02 - Data science modules covered

1. Exploration Options in GCP

03 - BigQuery
04 - Datalab
05 - Data Studio
06 - Cloud Dataflow

2. Cloud Datalab Basics

07 - What is Datalab
08 - Setting up the Cloud SDK
09 - Setting up Datalab
10 - Managing Datalab
11 - Using the exercise files
12 - Other capabilities

3. Datalab - BigQuery

13 - Setting up BigQuery
14 - BigQuery commands
15 - Reading data from BigQuery
16 - Working with DataFrames
17 - Writing data to BigQuery

4. Datalab - Cloud Storage

18 - Listing bucket contents
19 - Managing buckets
20 - Reading objects from a bucket
21 - Writing to buckets

5. Datalab - Visualizations

22 - Introduction to the charting API
23 - Line charts with BigQuery data
24 - Pie charts with BigQuery data
25 - Time series analysis with Cloud Storage

6. EDA with GCP - Use Case

26 - Loading data into a DataFrame
27 - Cleansing and transforming data
28 - Statistics and correlations
29 - Segmentation and profiling
30 - Writing results to Cloud Storage

7. Managing Datalab

31 - Datalab instance management
32 - Adding new packages
33 - Managing source code
34 - Datalab best practices

Conclusion

35 - Next steps