In this course, the participants get access to codes and algorithms in python/tensorflow and they apply these software tools on various types of the data. It is a hands-on course that allows participants to learn by assembling various programming modules to design interesting implementations of machine learning.
We’ll look at the realities of a pure machine learning approach through the lens of applied field cases relevant to formation evaluation, reservoir engineering, and production forecasting.
Intended Audience
Entry Level and Intermediate
Prerequisites (Knowledge/Experience/Education required)
Basic Computer Programming, Numerical Methods, Statistics, Familiarity with concepts like regression, interpolation, and curve fitting.
Equipment/Software Requirements
The instructor will use Windows OS during the course. Participants will execute python and tensorflow modules/codes to understand various Machine Learning concepts. All software used for the course is open source, so participants should bring computers where they can install the open-source software. Participants need at least 4GB of storage and 4GB RAM on their computer.
Course Outline
- Basics of Machine Learning in Python
- Supervised Learning – Classification
- Case Study #1 – Identifying Rock Type
- Supervised Learning – Regression
- Case Study #2 – Saturation Estimation
- Model Evaluation
- Case Study #3 – Image Analysis and Segmentation
- Cross Validation; Hyper-parameter Selection
- Case Study #4 – Shear Traveltime Prediction
- Unsupervised Learning – Transformation
- Feature Engineering and Feature Selection
- Case Study #5 – Waveform Analysis and Clustering
- Neural Networks
Learner Outcomes
- Participants can perform exploratory data analysis on large datasets containing numerical and categorical data.
- Participants can perform exploratory data analysis on time-series data and unsupervised transformations.
- Participants will be proficient with using Decision Tree Classifiers, kNN classifier, Random forest tree classifier, and K-Means Clustering on various datasets.
- Participants can construct training, testing, cross validation, feature elimination, feature ranking, parameter selection, and anomaly detection tasks.
- Participants can implement advanced clustering, regression, and classification techniques, such as DBSCAN, Hierarchical Clustering, neural networks, ElasticNet, and Support Vector Machines.
- Participants can construct deep neural networks for time-series analysis.