ML for DevOps
Objectives
Help professionals in DevOps field to use ML techniques and algorithms to automate and optimize DevOps processes as well as predict and prevent potential issues before they occur.
Benefits: Allow DevOps teams to leverage the insights generated by ML models to make informed decisions about how to optimize and improve their processes (quality, speed, agility and cost effectiveness).
Target Audience: DevOps engineers, Software Researchers interesting in DevOps and machine learning fields.
Pre-requisites:
- Python
- Being Comfortable with command Line
- Prior programming experience (at least 1+ year)
The Program
Module 1 : Introduction to ML for DevOps
- Introduction to machine learning
- Why ML for DevOps
Key learnings
- Understand the importance of applying Machine Learning on DevOps
- Understand a lifecycle of Machine Learning project
Module 2 : Machine Learning concepts
- Regression (Algorithms: Linear regression, Ridge, Lasso , ElasticNet, Decision Tree, kNN)
- Evaluation metrics: MSE, RMSE, R^2, Adjusted R^2. Features selection: Recursive Features Elimination
- Time series (Algorithms: AR, MA, ARMA, ARIMA, Prophet facebook. Statistical tests: ADF, KPSS)
- Dimensionality reduction: (PCA)
- Classification (Algorithms: kNN, Logistic Regression, Decision Tree, Random Forest, SVM)
Evaluation metrics: Accuracy, Recall, Precision, F1-score, ROC curve.)
Key learnings
- understand and learn how to use Machine Learning concepts
Module 3 : Machine Learning labs for DevOps
- Regression Lab
- Classification Lab
- Clustering Lab
- Time Series Lab
Key learnings
- Build and compare end-to-end ML models for service reliability prediction
- Build and compare end-to-end ML models for service quality classification and clustering
- Build and compare end-to-end ML models for EC2 CPU utilization forecasting
Effective SRE
Effective MLOps
ML for Life Sciences
ML4 DevOps
Interested in shaping your tailored training ?
Contact us at info@digital-innovation-partner.ch
