Student consultants are ready to use their technical and analytical skills to support data science and data engineering teams. Students with at least a beginner proficiency in Python will learn applied machine learning using scikit-learn in two phases. The first phase focuses on software engineering for data science and includes skills building with relational and NoSQL databases, data ingestion and wrangling, as well as software engineering topics such as object oriented programming, testing, security, and agile workflows. The second phase focuses on distributional statistical analysis, classification, clustering, and regression modeling using generalized linear models, non-parametric models, probabilistic models, ensemble models, and multi-layer perceptrons. Students will develop model selection techniques using visual diagnostics to interpret model behavior and to find the simplest, most predictive model for a data set using the bias/variance trade-off and analysis and selection of model triples that include algorithm, features, and hyperparameters. 

Learners will work on one main project over the course of the semester, connecting with you as needed with virtual communication tools.

Deliverables will vary depending on the scope of the project. Some Example Project deliverable include, but are not limited to:

• Presentations on data characteristics and model opportunities 
• Trained scikit-learn model and instance data sets 
• Data wrangling and analytics Python code 
• Documentation and visualizations related to model training