Zero to Agile Data Science
Master iterative data science model development. Learn a practical, step-by-step approach to building and refining models for impactful results.
Course Content
Zero to Agile Data Science is an intermediate-level Udemy course taught by Shreesha Jagadeesh. It runs 3h 14m, is taught in English, and includes a certificate of completion. Learners rate it 4.3 out of 5 from 80 reviews, with 847 students enrolled. It is a paid course, first published January 2021.
Unlock advanced data science skills with Agile Data Science for Classification. This hands-on course guides intermediate data scientists through practical application of Agile methodologies to solve real-world classification problems.
You'll work through three comprehensive projects:
- Predicting Credit Card Fraud: Develop robust models to identify fraudulent transactions.
- Predicting Customer Churn: Learn to forecast which customers are likely to leave.
- Predicting Financial Distress: Build predictive models to assess financial risk.
Each project is structured into five iterative "Day" modules (Day 1-Day 5), progressing from a foundational Random Forest Classifier to a highly optimized ensemble of five classifiers: XGBoost, LightGBM, Gradient Boosted Decision Trees, Extra Trees, and Random Forest. You'll gain experience with upsampled data techniques to handle imbalanced datasets.
Key skills you'll master include:
- Automated Data Quality: Detect and address problematic columns in raw data (Day 1).
- Imbalanced Dataset Handling: Create custom metrics for accurate evaluation (Day 1).
- Data Resampling: Explore and apply four effective data resampling techniques (Day 2).
- Missing Data Imputation: Learn strategies for handling null values (Day 2).
- Feature Engineering: Develop two powerful feature engineering techniques (Day 3).
- Feature Reduction: Apply four methods to reduce feature dimensionality (Day 3).
- Memory Optimization: Reduce the memory footprint of your models (Day 3).
- Custom Scoring: Implement custom scoring functions within GridSearchCV (Day 4).
- XGBoost Tuning: Modify default scoring metrics for XGBoost models (Day 5).
- Meta-Modeling: Build a meta-model to enhance predictive performance (Day 5).
This course provides complete Jupyter notebooks with readily usable source code and a library of reusable functions – perfect for incorporating these techniques into your own data science projects. Boost your machine learning expertise and become a more effective data scientist!
Want AI in your product? Hire the team behind TutorialSearch.io
LLM features, RAG, agents and automation wired into a real product — built by the devs who shipped this site.