DAT_SF_14

Course materials for General Assembly's Data Science course in San Francisco (4/28/15 - 7/9/15).

Instructor:

Yuchen Zhao

Experts-in-Residence:

Alex Chao (Office Hours: 2-6 pm Sundays)
David Feldman (Office Hours: 4-6:30 pm Tuesdays,Thursdays)

Course Schedule (Tentative)

Week	Tuesday	Thursday
1	4/28: Introduction to Data Exploration	4/30: Introduction to Machine Learning
2	5/05: Data Format, Access & Transformation	5/07: K-Nearest Neighbors Classification Final Project Kickoff
3	5/12: Naive Bayes Classification HW1 Due	5/14: Regression and Regularization
4	5/19: Logistic Regression HW2 Due	5/21: K-Means Clustering Project Milestone (PM1): Elevator Pitch
5	5/26: Clustering & Decision Trees	5/28: Tree-based Classifiers
6	6/02: Ensemble Techniques Project Milestone (PM2): Data Ready	6/04: Support Vector Machines
7	6/09: Dimensionality Reduction HW3 Due	6/11: Imbalanced Classes and Evaluation Metrics
8	6/16: Recommendation Systems Project Milestone (PM3): First Draft Due	6/18: Natural Language Processing
9	6/23: Alumni Panel Discussion, Peer Feedback (Assignment) & Final Project Work Session Project Milestone (PM4): Peer Feedback Due	6/25: Map-Reduce & Hadoop
10	6/30: Spark (password required)	7/02: Network Analysis
11	7/07: Project Presentations Day 1 Project Milestone (PM5): Presentation	7/09: Project Presentations Day 2 Project Milestone (PM5): Presentation & Paper

IMPORTANT: Presentation Schedule: link

Homework Schedule

HW	Topics	Dataset	Assigned	Due
1	Data Exploration	titanic	5/05	5/12
2	KNN & Cross Validation	iris	5/07	5/19
PM1	Elevator Pitch	project	-	5/21
PM2	Data Ready	project	-	6/02
3	Decision Trees	bank	6/02	6/09
PM3	First Draft Due	project	-	6/16 (Before Class)
PM4	Peer Feedback Due	project	-	6/23
PM5	Presentation & Paper	project	-	7/07

Resources

Working in the terminal

dat_sf_14's People

Contributors

Watchers

dat_sf_14's Issues

HW 2 Feedback

Good work with implementing your own cross validation! And the plots are helpful in visually determining the best K. Check out the solutions that will be posted soon for an alternate implementation.

HW3 feedback

Nice job on the homework! Your code looks great.

The only change I'd make would be to use a different evaluation metric. Mean squared error is better for evaluating regression accuracy. An accuracy score would be more intuitive in this case.

Check out the first post in this discussion for more info on binary classification metrics.
http://www.quora.com/What-are-benefits-and-weaknesses-of-various-binary-classification-metrics

Have a great weekend!

HW 1 Feedback

Good work! There are a couple of other ways to do some of the same things that you can look into. But otherwise great job! It's good that you added comments to your code to make it more readable.

Recommend Projects