This repository contains my experince taking COGS108 (WI'22) [Data Science in Practice] as taught by Prof. Jason Fleischer at the University of California, San Diego (UCSD). COGS108 is all about the practice of data science. This hands-on, practical course is intended to get you experience working on data science projects.
This class goes beyond the appreciation for data and data science you (may have) learned in COGS 9 by doing the same things talked about theoretically in that introductory course. Doing is rarely so simple. You will likely attempt to do something, do it wrong, learn from your mistakes, and with a bit of luck and skill, eventually succeed. That’s just part of the scientific process, and data science is no exception.
In focusing on the practice, there are theories that won’t be discussed and mathematical proofs that won’t be done. That is by design. In particular:
- There are entire courses dedicated to each of the topics we’ll cover. To have time to do anything, we can’t teach all the details in a single course.
- Experts in each of these domains are out there and excited to teach you the nitty gritty about each topic.
- We’re promoting data literacy. We believe that everyone who is data literate is at an advantage as they go out into the modern world. Data literacy is not limited to those who are computational gurus or math prodigies. You do not have to be either of those to excel at this course.
In this course, you will try many methods. You’ll even be asked to implement a technique that has not been explicitly taught. Again, this is by design. As a data scientist, you’ll regularly be asked to step outside of your comfort zone and into something new. Our goal is to get you as comfortable as possible in that space now. We want to provide you with a technical and a data science mindset that will allow you to ask the right questions for the problem at hand and set off alarm bells when something in your dataset or analysis is “off.”
---> COGS108 : GitHub
- Formulate a plan for and complete a data science project from start (question) to finish (communication)
- Explain and carry out descriptive, exploratory, inferential, and predictive analyses in Python
- Communicate results concisely and effectively in reports and presentations
- Identify and explain how to approach an unfamiliar data science task
Date | Week | Day | Topic | Section | Assignment | Lecture Quiz |
---|---|---|---|---|---|---|
1/03 | 1 | M | Welcome! | -- | -- | -- |
1/05 | 1 | W | Python Review | -- | -- | -- |
1/07 | 1 | F | Version Control I | -- | -- | -- |
1/10 | 2 | M | Version Control II | -- | -- | Q1 |
1/12 | 2 | W | Data & Intuition | -- | -- | -- |
1/14 | 2 | F | Data Wrangling (pandas ) |
D1 | A1; Group Signup* | -- |
1/17 | 3 | M | No Class - MLK day | -- | -- | Q2 |
1/19 | 3 | W | Ethics | -- | -- | -- |
1/21 | 3 | F | Data Science ?s | D2 | Project Review* | -- |
1/24 | 4 | M | Dataviz I | -- | -- | Q3 |
1/26 | 4 | W | Intro to Analysis | -- | -- | -- |
1/28 | 4 | F | Descriptive Analysis | D3 | Project Proposal* | -- |
1/31 | 5 | M | EDA | -- | -- | Q4 |
2/02 | 5 | W | Inference I | -- | -- | -- |
2/04 | 5 | F | Inference II | D4 | A2 | -- |
2/07 | 6 | M | Inference III | -- | -- | Q5 |
2/09 | 6 | W | Text Analysis I | -- | -- | -- |
2/11 | 6 | F | Text Analysis II | D5 | Checkpoint #1: Data* | -- |
2/14 | 7 | M | Machine Learning I | -- | -- | Q6 |
2/16 | 7 | W | Machine Learning II | -- | -- | -- |
2/18 | 7 | F | Text + ML | D6 | A3 | -- |
2/21 | 8 | M | No Class - President's day | -- | -- | Q7 |
2/23 | 8 | W | Nonparametric | -- | -- | -- |
2/25 | 8 | F | Geospatial I | D7 | Checkpoint #2: EDA* | |
2/28 | 9 | M | Geospatial II | -- | -- | Q8 |
3/02 | 9 | W | Dimensionality Reduction | -- | -- | -- |
3/04 | 9 | F | How to be wrong | D8 | A4 | -- |
3/07 | 10 | M | Guest lecture I | -- | -- | Q9 |
3/09 | 10 | W | Guest lecture II | -- | -- | -- |
3/11 | 10 | F | Data science jobs | D9 | -- | -- |
3/14 | Finals | M | -- | Final project*, video*, team eval survey | -- |
* indicates group submission. All other assignments/quizzes/surveys are completed & submitted individually.