Books I read
"In the multiverse of Data Science and Machine Learning, books have been my guiding stars. They have illuminated my journey, shaping my understanding and honing my expertise. As each page turned, I relentlessly find patterns and extract the best practice, finding not just knowledge, but wisdom and inspiration. The books I've read have not just built my professional prowess, but also profoundly shaped who I am as a data scientist.
Below, I share with you the books that have left an indelible imprint on my memory, hoping they might also spark your curiosity, light up your learning path, and help you unravel the mysteries of data science and machine learning as they have for me."
Effective Pandas
-Matt Harrison
What I learned: The concept of chaining Pandas' operation has immensely improved my code readability and memory efficiency. What's more? my code becomes cleaner, more concise, and more purposeful.
Effective XGBoost
-Matt Harrison
What I learned: A very powerful ensembling method for supervised machine learning. Here, I learned the ins and outs of tuning XGBoost hyperparameters, techniques to prevent overfitting, and so much more!
Matt goes far and beyond by exploring SHAP for feature importance and also serving ML models with MLFlow.
Data Cleaning and Exploration with Machine Learning
-Michael Walker
What I learned: This book introduces me to the idea of leveraging Pipeline to implement seamless Scikit-learn code, and most importantly, prevent data leakage. It also offers many practical tips and advice in dealing with data preprocessing.
Storytelling with Data
-Cole Nussbaumer Knaflic
What I learned: Every page offers invaluable insights on what to do and what not to do with your plot. Although less hands-on practice and more theoretical-based, I quickly learned that visualization technique goes beyond tools and packages. It also takes time to get one's eyes tuned to visually-aesthetic plots and truly master the tool one is comfortable with.
An Introduction to Statistical Learning
-Gareth James. Daniela Witten. Trevor Hastie. Robert Tibshirani
What I learned: One of the first Machine Learning books that I read. It has a great balance of theoretical and practical aspects that is beginner-friendly. This book exposes me to various ML models with the appropriate intuition on how the model works. While the examples in this book are written in R when I first read it, the authors recently published a new version with Python.
Getting Started with Streamlit for Data Science
-Tyler Richards
What I learned: What's next after building and evaluating an ML model? Building a full-stack website is not the main focus of a Data Scientist, spending more time refining the model is. I learned from Tyler how to get feedback for my model and showcase my work using Streamlit Share and Github.