Streamlit Monitoring Tool

The aim of this project is to monitor machine learning models on new input data via a broad range of metrics.

Setup and Run

pip3 install -r requirements.txt
cd src/
streamlit run main_front.py

Configuration File

You can modify the configuration file found in /params/conf/conf.json.

• Add a dict_info_files information containing your data and an equivalent mapping to the one shown below.

{
    "target": "VARIABLE TO PREDICT",
    "numerical_features": [NAME OF NUMERICAL COLUMNS],
    "categorical_features": [NAME OF CATEGORICAL COLUMNS]
}

• Choose the correct dict_info_file in the selected_dataset parameter of the configuration file.

Video Demo of the App

wine_monitoring_demo.mp4