MLFLOW

This repository explains how to train, monitor, make versions, register and server Machine Learning Models using MLFlow.

FastAPI

It also shows how to deploy Machine Learning Models as a Microservice using FastAPI and MLFLOW.

####### Create Python Virtual Environment and install all dependencies as follows ######

create venv using command: python -m venv venvname

activate venv: venvname\Scripts\Activate

install dependencies: pip install -r requirements.txt

run using: uvicorn main:app --reload

Open up the terminal, change to bash terminal and run the following command to run the bash script but first run your mlflow and fastapi server:

curl_iris.sh

Name		Name	Last commit message	Last commit date
Latest commit History 1 Commit
.ipynb_checkpoints		.ipynb_checkpoints
__pycache__		__pycache__
artifacts		artifacts
mlruns		mlruns
LRClassifier.pkl		LRClassifier.pkl
Logistic Regression_FASTAPI.ipynb		Logistic Regression_FASTAPI.ipynb
MLFlow Model Serving Live Demo.ipynb		MLFlow Model Serving Live Demo.ipynb
MLflow Live Demo.ipynb		MLflow Live Demo.ipynb
README.md		README.md
banking.csv		banking.csv
confusion_matrix.png		confusion_matrix.png
curl_iris.sh		curl_iris.sh
fast_api_code.py		fast_api_code.py
mlflow.db		mlflow.db
mlflow.png		mlflow.png
requirements.txt		requirements.txt
roc_auc_curve.png		roc_auc_curve.png