logistic

Private Prediction and Training of Logistic Regression Model

The examples are as follows:

• prediction_single.py shows how a single client can run a private prediction
• prediction_joint.py shows how several clients can combine features for a prediction
• training_single.py shows how a model can be privately trained on data from a single owner
• training_joint.py shows how several data owners can jointly train on their combined data

Note that the training data in the above examples are randomly sampled. Here gives a more realistic example showing two data owners jointly training on vertically split dataset, which are separately read from their local csv files.