In my journey through the realm of Machine Learning, I've delved into numerous crucial areas. These encompass both the theoretical and practical aspects of ML. here is a summary of the topics I learned in the past 40 days.
I've explored various ML models, understanding their unique characteristics and applications.
The core of ML lies in mathematics. I've gained a deep understanding of the mathematical principles that power these models.
The effectiveness of an ML model is determined by how well it's trained. I've learned techniques to train models and optimize their performance.
Understanding data before diving into modeling is crucial. Through EDA, I've learned to summarize, visualize, and interpret data.
Visualization plays a pivotal role in comprehending and representing data. It's a tool that I've honed over time.
Dirty data can derail the best models. I've practiced various techniques to ensure data cleanliness and quality.
Manipulating and transforming features can drastically improve model performance. I've explored numerous strategies to engineer and select the best features for models.
Throughout this journey, I've utilized several libraries relevant to the aforementioned topics. Some notable ones include:
pandasfor data manipulationmatplotlibandseabornfor data visualizationscikit-learnfor modeling and feature engineeringstatsmodelsfor statistical modelingxgboostfor gradient boostingtensorflowandkerasfor deep learningcatboostfor categorical boosting- ...and more!
Embarking on this journey has been both challenging and rewarding, and I'm eager to continue delving deeper into the world of Machine Learning.