ThinamXx / MLOps

The repository contains a list of projects which I will work on while learning and implementing MLOps.

MACHINE LEARNING OPERATIONS (MLOPS)

Day1 of 66DaysOfMLOps

• Experiment Tracking & MLflow: Experiment tracking is the process of keeping track of all the relevant information from an ML experiment, which includes source code, environment, data, model, hyperparameters, metrics, etc. MLflow is an open source platform for the machine learning lifecycle. It's also a python package that can be installed with pip and it contains four main modules: Tracking, Models, Model Registry & Projects. I have been learning and experimenting MLOps from MLOpsZoomcamp. Today, I have learned MLOps, MLOps Maturity Level, Why we need MLOps, Experiment Tracking Introduction & MLflow and few more topics related to the same. I have presented the implementation of Tracking using MLflow here in the snapshot. I hope you will gain some insights and you will also spend some time learning the topics from the course mentioned below. Excited about the days ahead !!
• 📚 MLOps Zoomcamp
• 📒 MLOps

Day2 of 66DaysOfMLOps

• MLflow: The Model Registry component is a centralized model store, set of APIs, and a UI, to collaboratively manage the full lifecycle of an MLflow Model. It provides Model lineage, Model versioning, Stage transitions & Annotations. A client of an MLflow Tracking Server creates and manages experiments and runs. I have been learning and experimenting MLOps from MLOpsZoomcamp. I have learned about Experiment Tracking and Model Management, Model Registry, MLflow Client, MLOps Workflow Orchestration, and a few more topics related to the same from here. I have presented the implementation of Model Training here in the snapshot. I hope you will gain some insights and spend some time learning the topics from the course mentioned below. Excited about the days ahead.
• 📚 MLOps Zoomcamp
• 📒 MLOps

Day3 of 66DaysOfMLOps

• I have been learning and experimenting MLOps from MLOpsZoomcamp. I have learned about Negative Engineering and Workflow Orchestration, Introduction to Prefect 2.0, Prefect Flow and Basics, Remote Prefect Orion Deployment, Turning Functions into Tasks, Parameters & Type Checking with Prefect, and many more topics related to the same. I have presented the implementation of Prefect Flow & Orion Deployment here in the snapshot. I hope you will gain some insights and spend time learning the topics from the course mentioned below. I am excited about the days ahead.
• 📚 MLOps Zoomcamp
• 📒 MLOps

Day4 of 66DaysOfMLOps

• Deploying Model as a Web Service: Creating a virtual environment with pipenv. Creating a script for prediction. Putting the script into a Flask app. Packaging the app to Docker. I have been learning and experimenting MLOps from MLOpsZoomcamp. I am learning about Model Deployment, Deploying Models with Flask and Docker, MLFlow and Model Registry, Batch Model Deployment, Scheduling, and many more related topics. I have presented the implementation of Model Registry and MLflow here in the snapshot. I hope you will gain some insights and spend time learning the topics from the course mentioned below. I am excited about the days ahead.
• 📚 MLOps Zoomcamp
• 📒 MLOps

Day5 of 66DaysOfMLOps

• I have been learning and experimenting MLOps from MLOpsZoomcamp. I have completed the topics related to Model Deployment. Here, I have learned and implemented Model Deployment using Web services such as Flask & Docker, Model Registry and MLflow, Batch Model Deployment & Scheduling Batch Jobs with Prefect. I have presented the implementation of Prefect flow and MLflow for model deployment here in the snapshot. I hope you will gain some insights and spend time learning the topics from the course mentioned below. I am excited about the days ahead.
• 📚 MLOps Zoomcamp
• 📒 MLOps

Day6 of 66DaysOfMLOps

• I have been learning and experimenting MLOps from MLOpsZoomcamp. Here, I have been learning about Model Monitoring, Evidently Services, Prometheus & Grafana, Model Performance, Data Drift & Concept Drift, Model Bias, Outliers, Explainability, and many more topics related to the same from here. I have presented the implementation of Evidently and Monitoring Services for model monitoring here in the snapshot. I hope you will gain some insights and spend time learning the topics from the course mentioned below. I am excited about the days ahead.
• 📚 MLOps Zoomcamp
• 📒 MLOps

Day7 of 66DaysOfMLOps

• MLflow: MLflow is an open source platform for the machine learning lifecycle, with a focus on reproducibility, training, and deployment. MLflow enables everyday practitioners to learn in one platform to manage the ML life cycle, from iteration on model deployment up to deployment in a reliable and scalable environment that is compatible with modern software system requirements. On my journey in Machine Learning Operations (MLOps), I have started reading the book Machine Learning Engineering with MLflow. Here, I have learned about the basics of MLflow, MLflow Modules such as MLflow Tracking, MLflow Projects, MLflow Models, End-to-End pipeline in MLflow, Dockerfile, and many more topics related to the same. I have presented the implementation of the MLflow Pipeline here in the snapshot. I hope you will gain some insights and spend time learning the topics from the course mentioned below. I am excited about the days ahead.
• 📚 Machine Learning Engineering with MLflow
• 📒 MLOps

Day8 of 66DaysOfMLOps

• Parquet: AWS recommends using the Parquet format because the Parquet format is up to 2x faster to unload and consumes up to 6x less storage in Amazon s3, compared to text formats. On my journey in Machine Learning Operations (MLOps), I am reading the book Designing Machine Learning Systems by Chip Huyen. Here, I have read about Machine Learning Systems & Design, Business and ML Objectives, Data Engineering Fundamentals, Data Sources, Data Formats, ML Use Cases, ML in Research Versus in Production, and many more topics related to the same. I have shared the notes about Data Sources, and Data Formats here in the snapshot. I hope you will gain some insights and spend time learning the topics from the book mentioned below. I am excited about the days ahead.
• 📚 Designing Machine Learning Systems
• 📒 MLOps

Day9 of 66DaysOfMLOps

• ETL: Extract, Transform, and Load: Extract is extracting the data from all data sources. In the extraction phase, we need to validate the data and reject the data that doesn't meet our requirements. Transform is the meaty part of the process, where most of the data processing is done. We can apply operations such as transposing, deduplicating, sorting, aggregating, deriving new features, and more data validation. Load is deciding how and how often to load the transformed data into the target destination, which can be a file, a database, or a data warehouse. On my journey in Machine Learning Operations (MLOps), I am reading the book Designing Machine Learning Systems by Chip Huyen. Here, I have read about Data Models, Relational Models, Document & Graph Models, Structured & Unstructured Data, Data Storage Engines, and Processing, Transactional & Analytical Processing, ETL, Modes of Dataflow, and many more topics related to the same. I have shared the notes about Data Models, and ETL here in the snapshot. I hope you will gain some insights and spend time learning the topics from the book mentioned below. I am excited about the days ahead.
• 📚 Designing Machine Learning Systems
• 📒 MLOps

Day10 of 66DaysOfMLOps

• Why Kubernetes? Kubernetes is an open-source system for automating the deployment, scaling, and management of containerized applications. It allows our pipelines to be scalable, without sacrificing portability, enabling us to avoid becoming locked into a specific cloud provider. The goal of Kubeflow is to standardize the entire process of MDLC and make it substantially easier and more efficient. On my journey in Machine Learning Operations (MLOps), I am reading about Kubeflow. I have read about Kubeflow, Model Development Life Cycle, Containerization, Kubernetes, Kubeflow's Design and Core Components, Data & Feature Preparation, Hyperparameter, Model Validation, Inference, and Pipelines, and many more topics related to the same. I have shared the notes about Kubeflow and Kubernetes here in the snapshot. I hope you will gain some insights and spend time learning the topics from the book mentioned below. I am excited about the days ahead.
• 📚 Kubeflow for Machine Learning from Lab to Production
• 📒 MLOps

Day11 of 66DaysOfMLOps

• Kubernetes: Kubernetes is a container orchestration platform used to manage containerised applications which is used to automate container management process. Machine Learning models can be easily easily scaled and scheduled when containerised, and the management of workload performance can be automated. TensorFlow Serving*: TensorFlow Serving is a flexible, high-performance serving system for machine learning models, designed for production environments. TensorFlow Serving makes it easy to deploy new algorithms and experiments, while keeping the same server architecture and APIs. I am learning about Kubernetes, TensorFlow Serving, Docker, Flask, and many other topics related to the same from Machine Learning Zoomcamp. I have shared the implementation of TF-Serving here in the snapshot. I hope you will gain some insights and spend time learning the topics mentioned below. I am excited about the days ahead.
• 📚 Machine Learning Zoomcamp
• 📒 MLOps