Predicting with Cloudera Machine Learning

Enterprise data science teams need collaborative access to business data, tools, and computing resources required to develop and deploy machine learning workflows. Cloudera Machine Learning (CML), part of the Cloudera Data Platform (CDP), provides the solution, giving data science teams the required resources. This four-day course covers machine learning workflows and operations using CML. Participants explore, visualize, and analyze data. You will also train, evaluate, and deploy machine learning models. The course walks through an end-to-end data science and machine learning workflow based on realistic scenarios and datasets from a fictitious technology company. The demonstrations and exercises are conducted in Python (with PySpark) using CML.

Who Can Benefit

• The course is designed for data scientists who need to understand how to utilize Cloudera Machine Learning and the Cloudera Data Platform to achieve faster model development and deliver production machine learning at scale. Data engineers, developers, and solution architects who collaborate with data scientists will also find this course valuable.

Skills Gained

• Through lecture and hands-on exercises, you will learn how to
• Utilize Cloudera SDX and other components of the Cloudera Data Platform to locate data for machine learning experiments
• Use an Applied ML Prototype (AMP)
• Manage machine learning experiments
• Connect to various data sources and explore data
• Utilize Apache Spark and Spark ML
• Deploy an ML model as a REST API
• Manage and monitor deployed ML models

Course Content

Introduction to CML

• Overview
• CML Versus CDSW
• ML Workspaces
• Workspace Roles
• Projects and Teams
• Settings
• Runtimes/Legacy Engines

Introduction to AMPs and the Workbench

• Editors and IDE
• Git
• Embedded Web Applications
• AMPs

Data Access and Lineage

• SDX Overview
• Data Catalog
• Authorization
• Lineage

Data Visualization in CML

• Data Visualization Overview
• CDP Data Visualization Concepts
• Using Data Visualization in CML

Experiments

• Experiments in CML

Introduction to the CML Native Workbench

• Entering Code
• Getting Help
• Accessing the Linux Command Line
• Working With Python Packages
• Formatting Session Output

Spark Overview

• How Spark Works
• The Spark Stack
• File Formats in Spark
• Spark Interface Languages
• Introduction to PySpark
• How DataFrame Operations Become Spark Jobs
• How Spark Executes a Job

Running a Spark Application

• Running a Spark Application
• Reading data into a Spark SQL DataFrame
• Examining the Schema of a DataFrame
• Computing the Number of Rows and Columns of a DataFrame
• Examining a Few Rows of a DataFrame
• Stopping a Spark Application

Inspecting a Spark DataFrame

• Inspecting a DataFrame
• Inspecting a DataFrame Column

Transforming DataFrames

• Spark SQL DataFrames
• Working with Columns
• Working with Rows
• Working with Missing Values

Transforming DataFrame Columns

• Spark SQL Data Types
• Working with Numerical Columns
• Working with String Columns
• Working with Date and Timestamp Columns
• Working with Boolean Columns

Complex Types

• Complex Collection Data Types
• Arrays
• Maps
• Structs

User-Defined Functions

• User-Defined Functions
• Example 1: Hour of Day
• Example 2: Great-Circle Distance

Reading and Writing DataFrames

• Working with Delimited Text Files
• Working with Text Files
• Working with Parquet Files
• Working with Hive Tables
• Working with Object Stores
• Working with Pandas DataFrames

Combining and Splitting DataFrames

• Combining and Splitting DataFrames
• Joining DataFrames
• Splitting a DataFrame

Summarizing and GroupingDataFrames

• Summarizing Data with Aggregate Functions
• Grouping Data
• Pivoting Data

Window Functions

• Window Functions
• Example: Cumulative Count and Sum
• Example: Compute Average Days Between Rides for Each Rider

Machine Learning Overview

• Introduction to Machine Learning
• Machine Learning Tools

Apache Spark MLlib

• Introduction to Apache Spark MLlib

Exploring and Visualizing DataFrames

• Possible Workflows for Big Data
• Exploring a Single Variable
• Exploring a Pair of Variables

Monitoring, Tuning, and Configuring Spark Applications

• Monitoring Spark Applications
• Configuring the Spark Environment

Fitting and Evaluating Regression Models

• Assemble the Feature Vector
• Fit the Linear Regression Model

Fitting and Evaluating Classification Models

• Generate Label
• Fit the Logistic Regression Model

Tuning Algorithm Hyperparameters Using Grid Search

• Requirements for Hyperparameter Tuning
• Tune the Hyperparameters Using Holdout Cross-Validation
• Tune the Hyperparameters Using K-Fold Cross-Validation

Fitting and Evaluating Clustering Models

• Print and Plot the Home Coordinates
• Fit a Gaussian Mixture Model
• Explore the Cluster Profiles

Processing Text: Fitting and Evaluating Topic Models

• Fit a Topic Model Using Latent Dirichlet Allocation

Fitting and Evaluating Recommender Models

• Recommender Models
• Generate Recommendations

Working with Machine Learning Pipelines

• Fit the Pipeline Model
• Inspect the Pipeline Model

Applying a Scikit-Learn Model to a Spark DataFrame

• Build a Scikit-Learn Model
• Apply the Model Using a Spark UDF

Deploying a Machine Learning Model as a REST API in CML

• Load the Serialized Model
• Define a Wrapper Function to Generate a Prediction
• Test the Function

Autoscaling, Performance, and GPU Settings

• Autoscaling Workloads
• Working with GPUs

Model Metrics and Monitoring

• Why Monitor Models?
• Common Models Metrics
• Models Monitoring With Evidently
• Continuous Model Monitoring

Appendix: Workspace Provisioning

• Workspace and Environment