Watch our on-demand lecture on SVMs featuring Alice Zhao:  Get Recording 

Machine

Learning Foundations

Details

PREREQUISITES

Some experience with Python (ability to write loops and use simple functions). The Python for Data Analysis course is a way to upskill a team to this point.

LENGTH

3-5 Days

LOCATION

On-site or Live Online

STUDENT PROFILE

Data scientists, statisticians, analysts, or others in similar analytical and quantitative roles

Course Description

This course provides a foundation of the two largest areas in machine learning: supervised and unsupervised learning. Instructors will demonstrate how machine learning techniques are applied to business problems, as well as how to implement these techniques using popular Python libraries. Lessons incorporate both lectures and hands-on exercises with a focus on cultivating practical skills.

Metis Senior Data Scientists have real-world business experience and will show your team how to apply machine learning concepts to daily tasks. Your team will then be able to hit the ground running, using their new skills to immediately impact their work.

We offer in-person training, as well as remote training via our Live Online technology. We are able to blend these capabilities so we can teach your entire team, even if they’re not all in one place.

Course Outcomes

Upon completion of the course, attendees should be able to:

Define “Machine Learning” and common terminology

Explain the different types of machine learning and the problems each can solve

Identify if a problem is a regression, classification, or clustering problem

Identify a useful metric for the business problem and optimize a model against it

Estimate the performance of a model on new data

Train and predict on messy datasets, including data that has outliers and/or missing data

Identify important features for the model

Identify the strengths and weaknesses when selecting a model for a problem

Apply and explain clustering (e.g. customer segmentation)

Training Content

DAY 1:

Introduction to supervised learning and regression

Probability and statistics review

Exploratory Data Analysis (EDA)

Introduction to Machine Learning

Linear Regression (Ordinary Least Squares)

Polynomial Regression

Overfitting vs Underfitting

DAY 2:

Regularization

Review of Day 1

Cross-validation and measuring generalizability

Overfitting vs Underfitting with Regularization

Feature engineering

Logistic Regression start (Learning objectives below)

DAY 3:

Introduction to Classification

Review day 2

Introduction to Classification

Ensemble-based methods

DAY 4:

Introduction to Neural Nets and Metis

Classification Metrics

Neural Net Overview

DAY 5:

Unsupervised Learning

Introduction to Unsupervised Learning

Clustering

Feature engineering for clustering

Pairing supervised and unsupervised learning

Details

PREREQUISITES

Some experience with Python (ability to write loops and use simple functions). The Python for Data Analysis course is a way to upskill a team to this point.

LENGTH

3-5 Days

LOCATION

On-site or Live Online

STUDENT PROFILE

Data scientists, statisticians, analysts, or others in similar analytical and quantitative roles