End-to-End-Kidney-Disease-Classification-Using-MLflow-DVC

MLflow for experiment tracking and DVC for ML Pipeline tracking, Data versioning and experiment tracking too. I'm using mlflow for experiment tracking.

Tasks

• Project template creation
• Project setup and requirements installation
• Logging, Utils and exception module
• Project Workflows
• All components notebook experiment
• All components modular code implementation
• training pipeline
• MLflow (MLOps tool for experiments tracking and model registration)
• DVC (MLOps tool for pipeline tracking and implementation)
• Prediction pipeline and user app creation
• Docker
• Final CI/CD Deployment on AWS

Workflows

• Update config.yaml
• Update secrets.yaml [Optional]
• Update params.yaml
• Update the entity
• Update the configuration manager in src config
• Update the components
• Update the pipeline
• Update the main.py
• Upddate the dvc.yaml
• Update app.py

How to install

Clone the repository

git clone https://github.com/MartinKalema/Kidney-Disease-Classification-MLflow-DVC.git

Create a conda environment after opening the repository and activate it

conda create -n kidney python=3.8 -y

conda activate kidney

Install the requirements

pip install -r requirements.txt

This Project is connected to Dagshub so all my experiments are sent to dagshub and can be viewed on dagshub itself or on the mlflow platform integrated there.

View experiments locally.

Do not set the tracking uri using,

mlflow.set_tracking_uri()

All experiments will be stored inside an auto generated folder called mlruns. Use the command below to view them in the mlflow web interface

mlflow ui

For remote views & collaboration.

Connect your github account to DagsHub @ https://dagshub.com

Whenever you close your editor and open it again, the mlflow uri configs are erased and you must run the commands below in your bash terminal again, otherwise the experiments will be saved on your local machine in a folder called mlruns

export MLFLOW_TRACKING_URI=https://dagshub.com/kalema3502/Kidney-Disease-Classification-MLflow-DVC.mlflow

export MLFLOW_TRACKING_USERNAME=kalema3502

export MLFLOW_TRACKING_PASSWORD=fb3845efcc3b2e46a4157b1d2c977a21e02dd16e

DVC setup

Initialise dvc

dvc init

Add the project pipelines to the dvc.yaml file, then run the command below.

dvc repro

To the pipeline structure, use the command below

dvc dag