This example shows how to get a Streamlit dashboard for monitoring data and model metrics with Evidently.
Get the evidently code example:
git clone git@github.com:evidentlyai/evidently.git
cd evidently/examples/integrations/streamlit-dashboardNote:
- it's recommended to use Python >=
3.9.12 - the streamlit version
1.19.0doesn't work with Python3.9.7
Create virtual environment named .venv and install python libraries
python3 -m venv .venv
source .venv/bin/activate
pip install -r requirements.txtSet up Jupyter Notebook
python -m ipykernel install --user --name=evidently
jupyter contrib nbextension install --user
jupyter nbextension enable toc2/mainNavigate to streamlit-app/ directory and launch Streamlit application
cd streamlit-app
streamlit run app.pyThis command launches a local Streamlit server and the Monitoring Dashboard app will open in a new tab in your default web browser.
Examples for Bike Sharing project
cd projects/bike-sharing
jupyter notebook- Open notebook
bicycle_demand_monitoring.ipynb - Run all cells to get predictions & generate Evidently reports
Notes:
- All Evidently reports (
.htmlfiles) are stored inreports/directory in each project