General Assembly Data Science Class

6/14/2016 to 8/18/2016

Instructor: Hamed Hasheminia

Tuesdays	Thursdays
6/14: Data Science - Introduction Part I	6/16 Data Science - Introduction Part II
6/21: Linear Regression Lines Part I	6/23: Linear Regression Lines Part II
6/28: Model Selection	6/30: Missing Data and Imputation
7/5: K-Nearest Neighbors	7/7: Logistic Regression Part I
7/12: Logistic Regression Part II	7/14: In Class Project
7/19: Tree-Based Models Part I	7/21: Tree-Based Models Part II
7/26: Natural Language Processing	7/28: Time Series Models
8/2: Principal Component Analysis	8/4: Data Visualization
8/9: Naive Bayes	8/11: Course Review
8/16: Final Project Presentations I	8/18: Final Project Presentations II

##Lecture 1 Summary (Data Science - Introduction Part I)

• Data Science - meaning
• Continuous, Discrete and Qualitative Data
• Supervised vs Unsupervised Learning
• Classification vs Regression
• Time series vs cross-sectional data
• Numpy
• Pandas

Resources

• Lecture 1 - Introduction - Slides
• Intro Numpy - Code
• Intro Numpy - Code - Solutions
• Intro Pandas - Code
• InClass Practice Code - Pandas
• InClass Practice Code - Solutions

Set up GitHub - Self-study guide

• Lecture 0 - GitHub - Slides
• excellent videos to set-up github. Students who have not used GitHub before must watch these videos.
• A hands-on introduction to Git and GitHub, and how to make them work together! More Git resources for beginners here

Pre-work for second lecture

• Review all lecture notes including Lecture Slides, Numpy notebook, and Pandas notebook
• Finish self-study Github guidlines listed above
• Finish Inclass Practice Code
• Review final project requirements. You can find timelines for final project at slide 11 of Lecture 1 PowerPoint Slides

Additional Resources

• Official Pandas Tutorials. Wes & Company's selection of tutorials and lectures
• Julia Evans Pandas Cookbook. Great resource with examples from weather, bikes and 311 calls
• Learn Pandas Tutorials. A great series of Pandas tutorials from Dave Rojas
• Research Computing Python Data PYNBs. A super awesome set of python notebooks from a meetup-based course exclusively devoted to pandas

Lecture 2 Summary (Data Science Intorduction - Part II)

• Measures of central tendency (Mean, Median, Mode, Quartiles, Percentiles)
• Measures of Variability (IQR, Standard Deviation, Variance)
• Skewness Coefficient
• Boxplots, Histograms, Scatterplots
• Central Limit Theorem
• Class/Dummy Variables
• Walkthrough describing and visualizing data in Pandas

Resources

• Lecture 2 - Slides
• Basic Statistics - Part 2 - Lab Codes
• Basic Statistics - Part 2 - Practice Code
• Basic Statistics - Part 2 - Practice Code - Solutions

HW 1 is Assigned

• Please read and follow instructions from readme
• This homework is due on June 23rd, 2016 at 6:30PM

Additional Resources

• Here you can find valuable resources for matplotlib
• A good Video on Centeral Limit Theorem

Lecture 3 Summary (Linear Regression Lines - Part I)

• Linear Regression lines
• Single Variable and Multi-Variable Regression Lines
• Capture non-linearity using Linear Regression lines.
• Interpretting regression coefficients
• Dealing with dummy variables in regression lines
• intro on sklearn and searborn library

Resources

• Lecture 3 - Slides
• Linear Regression - Part I - Lab Codes
• Linear Regression - Part I - Practice Code
• Linear Regression - Part I - Practice Solutions

Additional Resources

• My videos on regression lines. Video 1, Video 2
• This is an excellent book. In Lecture 3 and Lecture 4, we are going to cover Chapter 3 of this textbook.
• Seaborn
• Weighted Least Square Method (WLS)
• Good resource for heteroskedasticity
• Here Contours are elegantly introduced.)

Lecture 4 Summary (Linear Regression Lines - Part II)

• Hypothesis test - test of significance on regression coefficients
• p-values
• Capture non-linearity using Linear Regression lines.
• R-squared
• Interaction Effects

Resources

• Lecture 4 - Slides
• Linear Regression - Part II - Lab Codes
• Linear Regression - Part II - Practice Code
• Linear Regression - Part II - Practice Solution

Additional Resources

• My videos on regression lines. Video 1, Video 2
• This is an excellent book. In Lecture 3 and Lecture 4, we covered Chapter 3 of this textbook.
• statmodels.formula.api

HW 2 is Assigned

• Please read and follow instructions from readme
• Here you can find iPython notebook of your 2nd assignment.
• This homework is due on June 30th, 2016 at 6:30PM

Lecture 5 Summary (Model Selection)

• Bias-Variance Trade off
• Validation (Test vs Train set)
• Cross-Validation
• Ridge and Lasso Regression
• (Optional) Backward Selection, Forward Selection, All Subset Selection. (If you want to use these methods you need to use R)

Resources

• Lecture 5 - Slides
• Model Selection - Lab Codes
• Model Selection - Practice Code
• Model Selection - Practice Solutions
• HW 1 - Key

Additional Resources

• Preprocessing Library
• Cross-Validation Library
• This is an excellent book. You can find theory of Cross-Validation in Chapter 5. You can also learn about Lasso and Ridge regression in Chapter 6 of the mentioend textbook.
• Here you can find my video on Cross-Validation
• Here you can find my video on Ridge and Lasso Regression
• Here you can find my video on Best subset selection.

Lecture 6 Summary (Missing Data and Imputation)

• Types of missing data (MCAR, MAR, NMAR)
• Single imputation and their limitations
• Imuptation using regression lines and error
• Hot deck imputation
• multiple imputation

Resources

• Lecture 6 - Slides
• Missing Data and Imputation - Lab Codes
• Missing Data and Imputation - Practice Code and HW 3
• Missing Data and Imputation - Solution Code

Additional Resources

• Great Video by Dr. Elizabeth A. Stuart from John Hopkins University

Announcements

• HW 3 is assigned (Due at 6:30PM - July 7th)
• Please read this before starting your assignment.
• HW3 starter code can be found here

Lecture 7 Summary (K-Nearest Neighbors)

• Classification Problems
• Misclassifciation Error
• KNN algorithm for Classification
• Cross-Validation for KNN Algorithm
• Limitations of KNN Algorithm
• KNN algorithm for Regression

Resources

• Lecture 7 - Slides
• K-Nearest Neighbors - Lab Codes
• K-Nearest Neighbors - Practice Code
• K-Nearest Neighbors - Practice Solution

Announcements

• HW 2 Solutions are posted.

Lecture 8 Summary (Logistic Regression Part I)

• Logistic Regression - Intro
• Odds vs Probability
• Using Logistic Regression to Make predictions
• How one interprets coefficients of a Logistic Regression model
• Strength and weaknesses of Logistic Regression Model

Resources

• Lecture 8 - Slides
• Logistic Regression Part I - Lab Codes
• Logistic Regression Part I - Practice Code
• Logistic Regression Part I - Practice Solutions

Additional Resources

• Logistic Regression video

HW 3 Solutions Posted

• HW 3 Soltions

Lecture 9 Summary (Logistic Regression Part II)

• Unbalanced observations and Logistic Regression
• FP/FN/TP/TN/FPR/TPR
• The effect of changing Threshhold
• ROC Curves
• Area Under Curve
• How to compare classifciation algorithms

Resources

• Lecture 9 - Slides
• Logistic Regression Part II - Lab Codes
• Logistic Regression Part II - Practice Code
• Logistic Regression Part II - Practice - Solutions

Lecture 10 (In-Class Projects)

• Breast Cancer Project
• Breast Cancer - Group Notebook
• Energy Efficiency Project
• Energy Efficiency - Group Notebook
• Income Prediction Project
• Wine Quality Project
• Wine Quality - Group Notebook

Lecture 11 Summary (Tree-Based Models - part I)

• Decision Tree for Regression
• Greedy Approach
• Decision Tree for Classification
• Gini Index and Entropy index
• Limitation of Simple Decision Trees

Resources

• Lecture 11 - Slides
• Decision Trees Part I - Lab Codes
• Decision Trees Part I - Practice Code

Additional Resources

• Tree-Based Models - Video 1
• Tree-Based Models - Video 2
• If you are among the ones who hate dealing with dummy variables, enjoy working with this dummify function

Lecture 12 Summary Tree-Based Models - part II)

• Bagging
• Random Forest
• Boosting
• Tuning parameters for boosting and Random Forest

Resources

• Lecture 12 - Slides
• Decision Trees Part II - Lab Codes
• Decision Trees Part II - Practice Code

Additional Resources

• Bagging, Boosting, and Random Forest - Video
• BootStrap - Video

Announcement

• HW 4 is assigned and is due on July 28th 2016 at 6:30PM.
• Please read ReadMe file before working on your project.

Lecture 13 Summary (Natural Language Processing)

• Definition of Natural Language Processing
• NLP applications
• Basic NLP practice
• Stop words, bag-of-words, TF-DIF

Resources

• Lecture 13 - Slides
• Natural Language Processing - Lab Codes
• Natural Language Processing - Practice Code
• Natural Language Processing - Practice Solutions

Additional Resources

• If you want to learn a lot more NLP, check out the excellent video lectures and slides from this Coursera course (which is no longer being offered).
• Natural Language Processing with Python is the most popular book for going in-depth with the Natural Language Toolkit (NLTK).
• A Smattering of NLP in Python provides a nice overview of NLTK, as does this notebook from DAT5.
• spaCy is a newer Python library for text processing that is focused on performance (unlike NLTK).
• If you want to get serious about NLP, Stanford CoreNLP is a suite of tools (written in Java) that is highly regarded.
• When working with a large text corpus in scikit-learn, HashingVectorizer is a useful alternative to CountVectorizer.
• Automatically Categorizing Yelp Businesses discusses how Yelp uses NLP and scikit-learn to solve the problem of uncategorized businesses.
• Modern Methods for Sentiment Analysis shows how "word vectors" can be used for more accurate sentiment analysis.
• Identifying Humorous Cartoon Captions is a readable paper about identifying funny captions submitted to the New Yorker Caption Contest.

Pre-Work

• Bayes Examples
• Bayes Example - Solutions
• Principal Component Analysis Video

Lecture 14 Summary (Principal Component Analysis)

• Principal Component Analysis
• Computation of PCAs
• Geometry of PCAs
• Proportion of Variance Explained

Resources

• Lecture 14 - Slides
• Principal Component Analysis - Lab Codes
• Principal Component Analysis - Practice Code
• Principal Component Analysis - Practice Solutions

Additional Resources

• This tutorial on Principal Components Analysis (PCA) includes good refreshers on covariance and linear algebra
• To go deeper on Singular Value Decomposition, read Kirk Baker's excellent tutorial.
• Chapter 10 of Statistical Learning with applications in R

Lecture 15 Summary (Time Series Models)

• AutoRegressive Models
• Moving Averages
• ARMA
• ARIMA

Resources

• Lecture 15 - Slides
• Time Series Models - Lab Codes
• Time Series Models - Practice Code

Additional Resources

• This is a good resource for AR models
• Relatively easy to read book on time series.