You will be taught how to:

• Leverage your Oracle skills in the ML data science realm
• Use Python with a data science ML analysis focus
• Use the industry standard ML Python libraries
• Properly sample and split your data to properly train, test and validate
• Use Python scripts to build, test and use your ML projects
• Evaluate various ML algorithms using a variety of techniques
• Use industry standard data science ML project methodologies
• Build and use two ML projects that you can use in your work
• Optimize model goodness based on numbers and charts
• Minimize data volume and columns/features
• Explore your data looking for interesting and useful relationships
• Build and use both supervised and unsupervised ML models
• Extract AWR and ASH needed for your ML projects using SQL
• How to deploy your ML projects in a production IT environment

Class Pricing:

• $715 USD for Free members
• $572 USD for Essential and Advantage members (ask for your coupon code)
• FREE for Prime members (just ask to be registered)

Class Outline:

Session 0. Pre-class ML Sandbox Environment Setup HERE.
Session 1. Your Future With ML! Oracle Activity Anomaly Detection Basics
Session 2. Oracle Activity Anomaly Detection Advanced Topics
Session 3. Supervised Learning, Models And Accuracy
Session 4. Supervised Learning, Advanced Accuracy Determination
Session 5. Supervised Learning, Incorporating ASH & Model Improvement Strategies
Session 6. Supervised Learning, Feature Reduction, Model Tuning & Prediction

Some Of The Topics Covered:

Jupyter notebooks, DBAs as Data Scientists, working with Python dataframes and arrays, combining datasets, pivoting and/vs one-hot-encoding (OHE), selecting data, supervised vs unsupervised models, supervised models (dummy/naive, decision trees, artificial neural networks, etc.), accuracy determination, checking multiple models quickly, using sysmetric and ASH data, one hot encoding (manual and automatic and saving/using model), grouping samples, filtering out sample rows, recursive feature evaluation, feature correlation analysis, feature importance analysis, center and scaling, cross validation, confusion matrix, rare event problems and solutions, upsampling, SMOTE, label pre-processing, label threshold analysis and determination, dataset split and stratification, validation data, hyper parameter optimization with tuning grid, kmeans algorithm and strategies, time shifting, how to find "labels", deployment challenges and solutions, two very different anomaly detection strategies, the entire process, final model validation, cluster evaluation, anomaly detection process and deployment, cluster data point distance measurement in multi-dimensional space, dimensional reduction techniques, principle component analysis (PCA), independent component analysis (ICA) and more!

Details: 

Class Hours: There are 15 hours of live, interactive webinar training divided into six 2.5-hour sessions. To maximize learning and to work with your busy schedule, there will be a 1-3 day gap between sessions.

Class Structure: OraPub believes you learn more when you freely and actively engage during the training. This is why each session is full of live demonstrations and lots of Q&A time. Craig uses Jupyter Notebooks throughout the class, presentation slides and simply typing notes into a text file! All this is available after each session. If you miss a session, you are free to attend that session when the class is offered again.

Watching Recorded Sessions: Each session is recorded and used to create a unique web page just for your class. The recordings can be streamed using any device, but not downloaded. You can re-watch the videos as many times as you wish. The videos and your unique class web page will be available for three years!

After Class Activation/Homework: After each live session, an "activation" (think: homework) will be available for you to download. The activation is project based, uses the exact same Jupyter Note Craig used during the session, focusing on what Craig demonstrated and taught during class, plus some additional analysis. Any slides presented during the sessions will also be available for you to download.

Class Requirements:

You will need:

• It is recommended, but not required, that you attend OraPub's LVC, Core Truths for Oracle Professionals - Performance Engineering 1
• The ability to connect to the online training over the internet, hosted by Zoom. You do not need a microphone. Craig is the only one that speaks but all students are able to type their questions, answers, and comments... it can be very lively and a lot of fun!
• Time to do the homework. You can expect the homework to take between 60 to 120 minutes. This LVC is especially intense and very hands-on.
• The ability to setup your Machine Learning sandbox training environment. This should be completed BEFORE our first session. Your activation/homework will be done in your sandbox environment. I blogged how setup your environment here: https://blog.orapub.com/20200107/how-to-setup-your-machine-learning-sandbox-environment.html
• Oracle software is not required during the live training or for your homework.
• You do NOT need the ability to access an Oracle database. All class data is provided and will be available for download after the first session. 

Certificate of Mastery:

An additional Certificate of Mastery project is available for those who demonstrate they have successfully completed both class projects. Once you complete the certificate project, it is reviewed by the instructor (Craig Shallahamer). Once you receive a passing grade, a signed/authorized/numbered certificate (PDF) is available at no additional cost, i.e. it's FREE!