Machine Learning / Data Science Lifecycle:

• Define objective
• Data analysis
• Feature engineering
• Model development and evalualtion
• Deployment

Topic: SharpestMinds weekly touchbase

Join Zoom Meeting https://us04web.zoom.us/j/3832725619?pwd=aFFicG5XWWx6TGNYQlF2b0l2RmNRZz09

Meeting ID: 383 272 5619

Passcode: rjSn4i

We will use the conda package manager to setup our development environment. You can download the software on this page.

Once installed, build the environment by running:

cd sm-202011
conda env create

Once all the packages have been installed, activate it by

conda activate sm-202011

The goal of this project is to develop a machine learning model to predict which customers will default on their loan. This project has many of the challenges you might encounter on data science projects in industry: missing records, categorical variables, label you must create yourself. It's a fun problem that is also directly revelant to the banking industry.

• Data source: https://www.kaggle.com/wordsforthewise/lending-club
• Kaggle kernels you might find useful:
  • https://www.kaggle.com/pileatedperch/predicting-charge-off-from-initial-listing-data
  • https://www.kaggle.com/sishihara/simple-lightgbm-kfold

• Data source: https://www.kaggle.com/c/favorita-grocery-sales-forecasting/data