You need following tools installed:
docker
docker-compose
(included to Docker Desktop for Mac and Docker Desktop for Windows )
- Docker
- Docker compose
- Prefect
- Grafana
- Postgres
- Evidently
Note: all actions expected to be executed in repo folder.
-
Create virtual environment and activate it (eg.
python -m venv venv && source ./venv/bin/activate
orconda create -n venv python=3.11 && conda activate venv
) -
Install required packages
pip install -r requirements.txt
-
I am using this for my environment
`conda create -n py11 python=3.11 && conda activate py11`
- Run
baseline_model_nyc_taxi_data.ipynb
for downloading datasets, training model and creating reference dataset
jupyter notebook
And open http://localhost:8888/tree?token=beb2fbe1b95bf92927c0a58e591e132b74a782932a322fe6
like this which you will find in your terminal and run all the blocks in baseline_model_nyc_taxi_data.ipynb
file.
To start all required services, go to new terminal and execute:
docker-compose up
It will start following services:
db
- PostgreSQL, for storing metrics dataadminer
- database management toolgrafana
- Visual dashboarding tool
Open a new terminal, activate the virtual environemnt by running this conda activate py11
and then run this command to start the prefect server.
prefect server start
And run this in new terminal
prefect config set PREFECT_API_URL=http://127.0.0.1:4200/api
Check out the dashboard at http://127.0.0.1:4200
To calculate evidently metrics with prefect and send them to database, execute:
- open new terminal and run the conda environment :
conda activate py11
Then run this command :
python evidently_metrics_calculation.py
This script will simulate batch monitoring. Every 10 seconds it will collect data for a daily batch, calculate metrics and insert them into database. This metrics will be available in Grafana in preconfigured dashboard.
-
In your browser go to a
localhost:3000
The default username and password areadmin
-
Then navigate to
General/Home
menu and click onHome
. -
In the folder
General
you will seeNew Dashboard
. Click on it to access preconfigured dashboard.
Run debugging_nyc_taxi_data.ipynb
to see how you can perform a debugging with help of Evidently TestSuites
and Reports
To stop all services, execute:
docker-compose down