ML_models/KidneyDiseaseDeepLearningProject

Kidney Disease Classification MLflow DVC

Workflows

1. Update config.yaml
2. Update params.yaml
3. Update the entity
4. Update the configuration manager in src config
5. Update the comp9nents
6. Update the pipeline
7. Update the main.py
8. Update the dvc.yaml
9. app.py

Instalation

STEPS:

Clone the repository

https://github.com/matusoff/ML_models/KidneyDiseaseDeepLearningProject

STEP 01- Create a conda environment after opening the repository

conda create -p kidney python=3.8 -y conda activate kidney

STEP 02- install the requirements

pip install -r requirements.txt

Finally run the following command

python app.py

MLflow

Documentation

'''

cmd

-mlflow ui '''

dagshub

dagshub

MLFLOW_TRACKING_URI=https://dagshub.com/YourLink.mlflow

MLFLOW_TRACKING_USERNAME=YourUserName

MLFLOW_TRACKING_PASSWORD=YourTrackingPassword

python script.py

Run this to export as env variables:

Linux or macOS

export MLFLOW_TRACKING_URI=https://dagshub.com/YourLink.mlflow
export MLFLOW_TRACKING_USERNAME=YourUserName
export MLFLOW_TRACKING_PASSWORD=YourTrackingPassword

Windows

set MLFLOW_TRACKING_URI=https://dagshub.com/YourLink.mlflow
set MLFLOW_TRACKING_USERNAME=YourUserName
set MLFLOW_TRACKING_PASSWORD=YourTrackingPassword

DVC cmd

1. dvc init
2. dvc repro
3. dvc dag

About MLflow & DVC

MLflow

• It is Production Grade
• Trace all of your expriements
• Logging & taging your model

DVC

• It is very lite weight for POC only
• lite weight expriements tracker
• It can perform Orchestration (Creating Pipelines)