Project Structure:
├── data <- Data files.
├── notebooks <- Jupyter notebooks.
├── output <- Model predictions, logs, etc.
└── src <- Source code(.py) for use in this project.
├── data <- Scripts to load and preprocess data.
├── features <- Scripts to turn raw data into features for modeling.
└── models <- Scripts to train models and make predictions.
Default data directory is data/ and default config directory is configs/.
Run logistic regression pipeline locally:
python local.py --model LRRun logistic regression pipeline in a spark cluster:
spark-submit --packages com.microsoft.azure:synapseml_2.12:1.0.1 parallel.py --model LR