She Code Africa: A Community to Belong
The internet is a blessing; if you know where to look.
When I started learning data science, I was surprised by the vast amount of information available on the internet. I clicked every link, read a lot of articles, stalked data scientists on LinkedIn, signed up for more free courses than I can remember. I soon became so overwhelmed that I just left everything for weeks.
Imagine my joy when I received the congratulatory mail informing me of my acceptance into the 2nd cohort of the SCA Mentorship Program. I just knew I had found the community I had been looking for.
The learning path provided is very comprehensive, and I’ve learnt a lot in less than a month, I’ll summarize the lessons in this article.
Python
Python is the most recommended programming language for data science. With the resources provided, I’ve improved my knowledge of Python data types(lists, dictionaries, tuples) and their operators, conditionals and loops(if statements, for & while loops, try&except, break statements). I also learnt how to automate tedious work and avoid monotony by writing functions.
I wrote two Python programs (Guess the number and Password Generator) using what I learnt about functions and control flow.
Basic mathematics
Although you don’t need a mathematics degree to become a data scientist, one has to understand certain mathematical concepts to a reasonable extent in order to understand and gain insights from data. Concepts like probability theory and statistics should be well understood. I took a statistics course on eDX, where I studied the fundamentals and even learnt how to write functions that can estimate the outcome probability of coin flips.
Python libraries
Python libraries are collections of functions and methods that allow us to perform many actions without writing codes. Some of the libraries I learnt about are Pandas(data manipulation), NumPy(manipulation of numerical data), and Matplotlib(data visualization).
As an assignment, I analyzed and drew inferences on the Chinook Music Store dataset.
Data wrangling
It turns out the bulk of a data scientist’s work involves cleaning data, making raw data more appropriate for analysis. Data cleaning involves merging, concatenating, grouping, filtering and dropping missing/null values; these can be done using libraries like Pandas.
I’ve also connected with a handful of amazing ladies who are always ready to help.
I look forward to the remaining two months as I’ll be learning cool topics like SQL, data visualization tools like Tableau and Power BI, machine learning and lots more.
Thank you, SCA, for this opportunity!
