Week 1 at Data Science Bootcamp

My first week at Metis has been demanding, challenging, exciting, and exhausting. I chose Metis because of the rigor of the curriculum, so when the instructors assigned us an MTA turnstile project on the first day of the program, I was excited to get started on my journey into Data Science.

This focus of the MTA turnstile project was getting familiar with several of Python’s data manipulation and analysis packages (Numpy, Pandas, Matplotlib, and Seaborn). The goal of the project was to help WomenTechWomenYes (a hypothetical non-profit organization) promoting women in Tech come up with a strategy for spreading awareness of their organization at NYC subway stations. We were given the flexibility to create assumptions about what characteristics to look for in a station and who we wanted to target. Also, we were permitted to incorporate any data (i.e. demographic data) we thought would be useful in addition to the MTA’s official turnstile traffic data.

My team's goal was to leverage publicly available MTA subway traffic data and identify the NYC subway stations. Initially, our team decided to explore the data without any knowledge of the steps to follow for the Data Science workflow.

The Metis instructors intentionally gave us a full afternoon to struggle and to fail many times. There were many failed attempts to figure out how to manipulate the data to convert it into an form that could be easily analyzed. The following day, were not only then provided a lesson on the Data Science workflow, but also a deep dive into Exploratory Data Analysis (EDA) and how to do it properly. Exploring the data without any structure or organization made conducting Exploratory Data Analysis (EDA) more challenging than it should have been because we weren't familiar with what were the common things (I.e. duplicates, outliers, inconsistent numbers, etc..) to look for when cleaning the data and what was the Python syntax to conduct those tasks. I struggled with finding the most efficient Python syntax but it was a good kind of struggle. I didn't feel overwhelm. Once the instructors finally reviewed EDA with the class, it was nice to compare my method to the instructors and see that I was pretty close to executing my EDA the same way as the workflow. Additionally, I learned more efficient ways of writing Python syntax that accomplished the same goals.

My team also discussed what our Minimal Viable Product (MVP) and guiding principles were for the project. This provided more structure and identified what were the priorities for the project. After we agreed upon what the MVP was and how we were going to get it, we were able to analyze with a purpose and organization.

At the end of the week, my team was able to successful go through each phase of the Data Science Work Flow for the MTA turnstile project. We obtained a business understanding and narrowed down our focus by identifying a Minimal Viable Product (MVP) to guide our decisions. Leveraged Pandas and Numpy libraries to clean and explore the data. Created data visualizations using Matplotlib and Seaborn libraries to identify key insight (i.e. the top 5 NYC subway stations, best day of the week, and best time of day). Additionally, we effectively communicated our results and recommendations to the class.

Here's a link to my GitHub with my MTA Turnstile presentation.

After a reflection of the tasks and work that I have been able to accomplish during the first week, I am confident that I have made the right decision to leverage Metis to FINALLY make a career switch to the Data Science. My personal goals for this program were to improve my technical skills (Python, SQL, Hadoop, etc..), build a network with other really intelligent data nerds who are passionate about Data Science (like myself), find a job after the program, and ,most importantly, HAVE FUN.

I know that my journey with Metis will be an uphill battle that is long, scary, and sometimes dark. But with patience, hard work, and a strong desire to become an intelligent Data Scientist, I know that I can make it top of the hill and scream "Victory!". I am excited to continue this climb into Data Science along side this amazing cohort and instructors.

RAD Guy