General Assembly Data Science 12

Contact Information

• Gavin Hackeling: gmh283@nyu.edu
• James Beveridge: jamesbev@gmail.com
• Shawn Oakley: shawnoakley@gmail.com

Questions and Discussions:

• Closed and Open Issue List (via pulse)
• Create a new issue

Classes

Lecture 1: Introduction to Data Science

Thursday, 2014/08/07

Class Materials

• Lecture
• lab_01a.md
• lab-1a-submission-example.md
• lab_01b.md

Lecture 2: Simple Linear Regression

Tuesday, 2014/08/12

Class Materials

• Lecture
• simple-linear-regression-lab.pdf
• python-lab.ipynb
• numpy-lab.ipynb

Lecture 3: From Simple Linear Regression to Multiple Linear Regression & an Introduction to scikit-learn

Thursday, 2014/08/14

Class Materials

• Lecture
• lecture_3_numpy_review.ipynb
• lecture_3_intro_to_sklearn.ipynb
• lab_3_multiple_linear_regression.ipynb

Lecture 4: Introduction to pandas & matplotlib, Multiple Linear Regression Review

Tuesday, 2014/08/19

Class Materials

• Lecture
• multiple_linear_regression_review.ipynb
• object_oriented_programming_lab.ipynb
• matplotlib_intro_lecture.ipynb
• pandas_intro_lecture.ipynb
• pandas_lab.ipynb

Project 1

• Project 1: Regression

Lecture 5: From Multiple Linear Regression to Polynomial Regression & Regression Project Workshop

Tuesday, 2014/08/21

Class Materials

• Lecture
• PolynomialRegressionAndRegularization.ipynb
• intro_to_matplotlib.ipynb

Extra Materials

• matplotlib_examples.ipynb

Lecture 6: From Multiple Linear Regression to Logistic Regression & Text Feature Extraction

Tuesday, 2014/08/26

Class Materials

• Lecture
• document_classification.ipynb

Lecture 7: Classification with Logistic Regression and K-Nearest Neighbors

Thursday, 2014/08/28

Class Materials

• Lecture
• k_nearest_neighbors_classification.ipynb

Lecture 8: (optional class) Kaggle Competitions

Tuesday, 2014/09/02

Class Materials

Lecture 9: Non-linear Classification and Regression with Decision Trees and Ensemble Learning

Thursday, 2014/09/04

Class Materials

• Lecture
• decision_trees.ipynb

Lecture 10: Cluster Analysis with K-Means

Tuesday, 2014/09/09

Class Materials

• Lecture
• elbow-method-solutions.ipynb
• semi-supervised-classification.ipynb

Lecture 11: Dimensionality Reduction with Principal Component Analysis

Thursday, 2014/09/11

Class Materials

• Lecture
• eigenfaces.ipynb
• iris-reduction.ipynb

Lecture 12: Machine Learning Review

Thursday, 2014/09/16

Class Materials

• Lecture

Lecture 13: Project Workday

Thursday, 2014/09/18

Class Materials

• Lecture

Lecture 14: The Perceptron

Thursday, 2014/09/23

Class Materials

• Lecture
• document_classification_2.ipynb

No class

Thursday, 2014/09/25

Lecture 15: From the Perceptron to Support Vector Machines

Thursday, 2014/09/29

Class Materials

• Lecture
• natural_image_character_recognition.ipynb

Lecture 16: From the Perceptron to Artificial Neural Networks

Thursday, 2014/10/02

Class Materials

• Lecture
• mlp_xor.ipynb
• mlp_mnist.ipynb

Lecture 17: Web Apps with scikit-learn

Thursday, 2014/10/07

Class Materials

• Lecture
• Lab

Lecture 18: Recommendation Engines

Thursday, 2014/10/02

Class Materials

• Lecture
• Lab

Lecture 19: Guest Lecture 1

Thursday, 2014/10/09

Class Materials

Lecture 20: Guest Lecture 2: Visualization with D3.js

Thursday, 2014/10/14

Class Materials

Lecture 21: Final Project Workday

Thursday, 2014/10/16

Class Materials

Lecture 22: Final Project Presentations

Thursday, 2014/10/21

Class Materials

Lecture 23: Final Project Presentations

Thursday, 2014/10/23

Class Materials

Git Workflow and Command Line Tips:

• Tips

Using a virtual machine

For further help/troubleshooting, feel free to come by the office hours or contact us for further help.

In case you're running into issues setting the environment up on your local environment, you can download a machine image from the following link:

https://www.dropbox.com/s/7nt0rt54m7jtxj5/GADS-InstalledEnv_1%28import%29.ova

Use the following user info to login:

User: GADS

Password: gadspassword

NOTE: VMWare Player appears to occassionaly throw errors when dealing with this image. Therefore, it would probably be easier to use VirtualBox

To install the virtual machine on VirtualBox

• Install VirtualBox
• Select File>Import Appliance
• When prompted, select 'GADS-InstalledEnv_1(import).ova' (or whatever you decided to name the downloaded image)
• Click through the rest of the import process
• In the main menu of VirtualBox, highlight the name of the machine (it should be whatever you named the ova file), and press 'Start'.

The image has Ubuntu 14.04 installed. I found that the default RAM allocation (512 MB) was running a bit slow, so I adjusted the default allocation to 1024 MB. If you're running into performance issues, you may need to adjust the memory allocation settings.

The image has the following libraries installed on it:

• scikit-learn (v 0.15.1)
• numpy (v 1.8.1)
• pandas (v 0.13.1)
• scipy (v 0.13.3)
• pip (v 1.5.6)
• matplotlib (v 1.3.1)
• git
• nltk (v 2.0.4)

Git Repository in VM image

The class's git repo has been cloned as the following directory:

~/Desktop/courseGit/GADS12-NYC

There is a small bash script (~/Desktop/courseGit/GADS12-NYC/update) that allows you to type 'update-GADS' in order to clone the latest version from the repo. The command is only soft-linked to the script, meaning that if 'update' is moved, the command won't work anymore. Even if that happens though, you can just type 'git pull' in the repo's directory to update your local repo.