Data Scientist: The Sexiest Job
of the 21st Century

Harvard Business Review

Applied Machine Learning Online Course

Eight-week Online ML Bootcamp

The Data Incubator is a Cornell-funded data science training organization. We run an introductory 8-week part-time online program geared towards giving working professionals an immersive hands-on experience with Machine Learning. Developed based on feedback from our hundreds of industry partners and using the same rigorous methodology as our Data Science Fellowship Program, the curriculum gives students the skills they need to harness the power of data.

Take the Assessment

Experienced teachers

Learn from The Data Incubator's experienced data science instructors dedicated to teaching data analytics.

Jumpstart your career

Learn the most sought-after tools and techniques in the industry to help jumpstart your data analyst career.

Designed for working professionals

This course was designed for the busy lives of working professionals with a part-time schedule and recorded lectures.

Build a series of miniprojects

Gain hands-on experience applying the tools employers value to real-world data sets. All powered by a 100-node cluster.

Live lectures

Live lectures allow you to ask your instructor questions and interact with your classmates.

Small class sizes

Small class sizes ensure you have plenty of access to your instructor and can receive personalized feedback on your progress.

Curriculum

Prerequisites:

Intermediate Python
Familiarity with linear algebra
Familiarity with statistical modeling

Target Audience:

This class is for you if:

You have taken our Data Science Foundations course and want to deepen your understanding of machine learning
You have experience with modeling or have a background in data science, and would like to learn the theoretical principles and practical applications of machine learning
You want to gain hands-on experience applying machine learning techniques to real-world data sets

Learning Outcomes:

Upon completion of this course, you will:

Understand and be able to deploy the steps required to complete an end-to-end machine learning project
Have learned about various machine learning models, how they work, pros and cons of each, and how they compare
Know how to best use machine learning to solve business problems and analyze problems using a machine learning perspective
Understand how to effectively use Python's machine learning library

Course Outline:

NumPy and SciPy

NumPy Data types (the nouns): Operations (the verbs): Persisting NumPy objects SciPy

Pandas

Nouns (objects) in Pandas Verbs (operations) in Pandas Loading data (and basic statistics / visualization) Indexing data frames Filtering data Joining data Aggregating data Sorting by indices and columns cut and qcut Unique values Handling missing and NA data Manipulating strings Indices in Pandas Function application and mapping Pandas Timestamps Multi-indices, stacking, and pivot tables Plugging into more advanced analytics Taking advantage of parallel computing

Probability

Random Variables Statistics of Random Variables Quantile Functions Modeling observations with random variables Bernoulli Trials Poisson Distribution Exponential Distribution Normal Distribution Central Limit Distribution Beta Distribution Bayes' Theorem

Introduction to Machine Learning

Statistics vs Machine Learning Data as a Matrix Models as Functions Types of Machine Learning Parameters and Learning

Anomaly Detection, Session 1

CitiBike Ridership Data Detrending Detecting seasonality

Anomaly Detection, Session 2

z-Score Moving-Window Averages Including windowed data in model Online Learning Exercises Bayesian Changepoints

Recommendation Engine, Session 1

Problem definition and data format Exploring a New Data Set Exercise Feature engineering Nearest Neighbors Tag Data Dimensional Reduction Another alternative: NMF A Brief Aside: The Hashing Trick Recommendation for a User Cooperative Learning Exercises

Recommendation Engine, Session 2

Regression of Ratings Baseline Model Overfitting and Cross-validation Modeling Interaction Exercises Surpriselib

Classification

Precision and Recall Other Classification Metrics Probabilistic Models Logistic Regression Multiclass classification problems

Scikit-learn Workflow

Writing custom estimators and transformers Pipelines Feature Unions Data Types Validating Your Implementations

K Nearest Neighbors

Tuning k and other hyperparameters Normalizing features

Transformers and Preprocessing

Feature Scaling Encoding Categorical Variables Imputation Dimension Reduction Natural Language Processing Custom Transformers Answers to Questions

Unsupervised Learning

Metrics for clustering K-Means clustering Gaussian mixture models Dimensionality Reduction Random projections Matrix factorization Principal Components Analysis (PCA) Non-negative Matrix Factorization (NMF) Restricted Boltzmann Machines

Support Vector Machines (SVM)

Maximal Margin Classifier Linear SVM Non-linear SVM Multi-class SVM Approximating kernels Support Vector Regression Outlier detection using SVM

Decision Trees and Random Forests

Decision Trees Ensemble Methods Determining feature importance

Naive Bayes

Predictive modeling using Naive Bayes Classifying mushrooms (an example)

Unbalanced Classes

Introduction: cancer detection case study Definition and common scenarios for unbalanced data Simple techniques to deal with unbalanced data Undersampling Oversampling Synthetic data augmentation Additional approaches Train/Test split with unbalanced data Probabilities with unbalanced data The Python imbalanced-learn package

Choosing the Correct Machine Learning Algorithm

Few Features Many Features Few Observations Many Observations Underfitting Overfitting Explicability Prediction Speed Parallelization Online Learning Feature Scaling Outlier Detection/Novelty Detection Comparing ML Algorithms

Natural Language Processing

Text as a "bag of words" Word importance: Term frequency–inverse document frequency (TF-IDF) Document similarity metrics Engineering your features Building the Classifier Additional NLP topics and resources

Sentiment Analysis

Bag of Words Model Interpreting the Model Grammar and Other Tools

Deep Learning

Neural Net Basics: The perceptron Chaining Perceptrons into Neural Networks An Example Neural Net: word2vec The model - Continuous Bag of Words (CBOW) Let's try it! Neural Network Libraries

Time Series

Trends in time series data Cross-validation for time series Modeling drift Modeling seasonality Modeling "noise" Using external data sources as features More advanced time series modeling frameworks

Digital Signals

Sampling Noise & Filters Audio files Filters Frequency domain

Outlier Detection

Motivation Concepts Scikit-learn Implementation One-class SVM Isolation Forest Case Study: Anomaly Detection in Time Series Modeling the background Detecting seasonality with Fourier Transforms Detrending z-Score Moving-Window Averages Including windowed data in model Bayesian Changepoints Online Learning References

Introduction to TensorFlow

Flow: The Computation Graph Exercise: Implementing a basic graph Repeating calculations Tensor: The data types and shapes Exercise: Reducing tensors of arbitrary shape Graphs Sessions TensorBoard

Iterative Algorithms

Archimedes' algorithm in Python Archimedes' algorithm in TensorFlow Variables Archimedes' algorithm done idiomatically Exercise: Fibonacci numbers Newton's Method of Root Finding Exercise: Minimizing functions Exercise: Minimizing functions of two variables

Basic Neural Networks

The XOR problem Hidden layers Exercise: Number of hidden neurons Activation functions Initial weights Exercise: Exploring activation functions Exercise: Adding a hidden layer

Sign up

Sign up for the Applied Machine Learning Online Course

The next Applied Machine Learning Online Course will run from July 8th to August 28th, 2019. Classes are generally held on Mondays and Wednesdays from 6:30-9:30 PM ET / 3:30-6:30 PM PT, with some exceptions for holidays. The deadline for registration is July 1st. The course tuition is $3495 with early-bird discounts available.

Please click "Learn More" and/or submit your contact details via the form below to receive updates with information on registration availability and special discounts.

Learn More Schedule a Call

Interested in the course but need help with financing? We've partnered with Climb Credit to help. Climb offers 3-year financing with fixed interest rates and monthly payments. Apply online and get a decision same-day with no impact on your credit. Have questions about how it works? Check out their FAQ.

Success! Please check your inbox for a confirmation email. Occasionally, emails are sent to the spam folder.

Invalid Information! Please enter all your information

Something went wrong! Please resubmit and contact us if the problem persists.

TDI + MRINetwork

The Data Incubator recently teamed up with MRINetwork to increase its access to hiring partnerships worldwide. MRINetwork is comprised of over 1,500 search professionals who specialize in hundreds of industries, many of whom came from the industries in which they now recruit. The addition of MRINetwork, and its network of existing clients, will add thousands of hiring partners on top of TDI's existing 300+ hiring partnerships. As the need for data scientists has increased exponentially over the past few years, MRI provides TDI students with immediate access to new data science positions in geographies worldwide, as well as greater access to companies with a fundamental need for the data science talent required to harness the power of their data.

Testimonials

"The data science skills I sharpened at The Data Incubator helped me analyze diversity in STEM education, model SaaS stock prices, and compare industry growth rates. The instructor's background in quantitative finance made for relevant use cases that benefit not only data scientists, but also data-driven finance people like myself."

Christian Templeton, Google

Formerly Salesforce, Symantec. Accepted to The Data Incubator Fellowship.

"From day one I was getting my hands dirty working with data using industry-relevant tools. Having completed the program I'm now better equipped to manage engineering and product teams, and able to conduct sophisticated ad-hoc analysis. Everyone at TDI blew me away; truly exceptional."

Michelle Miller, Hack Reactor

Formerly Berkeley, Smithsonian

"Their heavy focus on applied learning meant that I was working on real data and solving real problems right from the start. While lectures were a valuable component of their courses, the real learning took place while working on projects. These projects were practical, engaging, and instructive, and generally great opportunities to get hands-on experience with what I had learned in the lectures."