Tutorial: Build Data validation and Model Monitoring pipelines with DVC and Evidently

Create virtual environment

Create virtual environment named .venv and install python libraries

python3 -m venv .venv
echo "export PYTHONPATH=$PWD" >> .venv/bin/activate
source .venv/bin/activate
pip install --upgrade pip setuptools wheel
pip install -r requirements.txt

Run pipelines

train

dvc repro pipelines/train/dvc.yaml

predict

dvc repro pipelines/predict/dvc.yaml

monitor

dvc repro pipelines/monitor/dvc.yaml

View reports

Enter directory reports/, open required period folder and launch HTML report in a browser.