A ML pipeline for basic XGBoost/LightGBM/simple NN model
pip install scikit-learn shap
optional arguments:
-h, --help show this help message and exit
--data_dir DATA_DIR path to data
--type TYPE classification or regression
--encoder ENCODER categorical feature encoder, one of: label, onehot, target
--algorithm ALGORITHM which model you want to train(XGB or LGB or nn)
--y_col Y_COL column name of predict target
--output_dir OUTPUT_DIR path to output model
--na_rule NA_RULE (if you have domain rule)path to na rule(json)
python train.py --y_col Survived --data_dir ./data/titanic.csv --output_dir ./models/titanic --exclude_col 'PassengerId'
server:
python server.py --model_path ./models/titanic --shap_flag
client:
python python_client.py --data_dir ./data/titanic_client_test.json
python web_server.py