MLops and ML
We started as a small interior design firm in downtown Michigan, aiming to help Machine Learning (ML) involves creating algorithms that enable computers to learn from and make predictions based on data. It includes processes like data preprocessing, model training, and evaluation. MLops (Machine Learning Operations) extends DevOps practices to ML workflows, focusing on streamlining the deployment, monitoring, and maintenance of ML models in production. MLops ensures scalability, reproducibility, and continuous integration/continuous deployment (CI/CD) of ML models, bridging the gap between data science and IT operations. By integrating MLops, organizations can effectively manage the entire ML lifecycle, leading to more reliable and efficient AI-driven solutions.
Our mission
We're on a mission to change the way the housing market works. Rather than offering one service or another, we want to combine as many and make our clients' lives easy and carefree. Our goal is to match our clients with the perfect properties that fit their tastes, needs, and budgets.
Our vision
We want to live in a world where people can buy homes that match their needs rather than having to find a compromise and settle on the second-best option. That's why we take a lot of time and care in getting to know our clients from the moment they reach out to us and ask for our help.
Course Content
Basic Module: Introduction to Machine Learning (6 hours)
Hour 1-2: Understanding machine learning: definition, types, and applications (Why ML or AI; Difference Between AI and ML; Will AI Take Jobs?).
Hour 3-4: The machine learning pipeline: data collection, pre-processing, feature engineering.
Hour 5-6: Mathematical foundations: linear algebra basics and introductory calculus.
Start: Supervised Learning (6 hours)
Hour 7-8: Introduction to supervised learning: features, labels, and training data.
Hour 9-10: Linear regression: theory, cost functions, gradient descent.
Hour 11: Logistic regression: theory, binary and multiclass classification.
Hour 12: Decision trees and ensemble methods: theory, random forests.
Preserve: Unsupervised Learning and Evaluation (7 hours)
Hour 13-14: Introduction to unsupervised learning: clustering, dimensionality reduction.
Hour 15-16: Clustering algorithms: k-means, hierarchical clustering.
Hour 17: Dimensionality reduction techniques: principal component analysis (PCA), t-distributed stochastic neighbor embedding (t-SNE).
Hour 18-19: Model evaluation metrics: accuracy, precision, recall.
Advanced Topics and Practical Projects (12 hours)
Hour 20-21: Transfer learning: concepts, pre-trained models, fine-tuning.
Hour 22: Ethical considerations in machine learning: bias, fairness, transparency.
Hour 23-28: Hands-on projects: participants work on supervised and unsupervised learning tasks using real-world datasets.
Hour 29-31: Capstone project presentations and discussion of new ideas to be implemented in the future: participants showcase their machine learning projects and share insights.
Hour 32-35: Miscellaneous and Doubt session
Play on real time- Productions (8 hours)
ML Model Workflows
Introduction to MLops
Building Data Pipelines
Time Series Data Pipeline: AWS - S3
Development Frameworks and Tools installation
ML Model Development Framework: Jupyter
Replicating Production Environment: Docker
Development of Prod Code-Base: VS Code
Is machine performing the way we have designed for (3 hours)
Deployment and Monitoring
Deployment to Production: Server (GitLab)
Monitoring Models in Production: Sentry
Monitoring Model Output: Grafana
Tracking Model Performance: CSI and PSI
Learning Algorithms Effectively
Better Ways to Learn Algorithms
Study All Algorithms and Search Real-Life Problems
List Domains and Implement Algorithms
Projects Ideas (3 hour)
Clothing Application
Extraction of Rotten Food Grains from Mixture of Good and Bad
Automated Thermal Screening at Entry Gates
Age and Gender Prediction
Non-Invasive Estimation of Heart Rate and Blood Oxygen Level
Document Verification