1. Course description.

Python has become a powerful language of data science and is now commonly used as the leading programming language for predictive analytics and artificial intelligence. During this hands-on “Machine Learning with Python” training course, you will learn to utilise the most cutting edge Python libraries for clustering, customer segmentation, predictive analytics and machine learning on the real-world data.

The course explores practical applications of the most frequently used machine learning approaches such a Multiple Linear, Polynomial (non-linear) and Logistic Regressions, k-Means and Hierarchical Clustering, k-Nearest Neighbours, Naive Bayes, Decision Trees and ensemble algorithms e.g. Random Forests, Adaptive Boosting or Extra Gradient Boosting approaches using Python’s major scientific libraries such as NumPy, pandas, SciPy as well as more specialised, statistical and machine learning oriented packages e.g. scikit-learn, statsmodels, pycaret and h2o.

Apart from this, you will learn to evaluate the predictive models based on the obtained metrics such as sensitivity, specificity, F-score, Kappa etc., and optimise the accuracy and efficiency of these models using various methods of cross-validation, grid-search and performance boosting.

Please note this training course doesn’t include Neural Networks and Deep Learning approaches.

2. Course schedule.

This instructor-led course duration is planned over 6 teaching weeks.

In between the six weekly online live tutorials (2.5 hours long each) you will improve your skills by watching pre-recorded video tutorials at our Mind Project Learning Platform and working through set tasks (e.g. quizzes) as well as homework coding exercises which will require 4-6 hours of your time commitment per week (24-36 hours). We estimate that the total time commitment is 40-50 hours over 6 teaching weeks.

Start date: Wednesday, 17th of March 2021 @10:00 am London (UK) time

Schedule of sessions: Every Wednesday at 10:00 am London (UK) time for 6 weeks

Deadline for registrations: Monday, 15th of March 2021 @ 17:00 London (UK) time

Week 1: Introduction to Machine Learning with Python

• Concepts, terminology and context: unsupervised vs. supervised vs. semi-supervised approaches,
• Overview of methods and applications,
• Preparing data for Machine Learning tasks: revision of probability distributions, data normalisation and standardisation techniques, feature engineering, dealing with missing values,
• Dimensionality reduction with Singular Value Decomposition, Principal Component Analysis and Factor Analysis.

Week 2: Unsupervised learning with clustering approaches

• K-means and k-medians clustering,
• Hierarchical clustering,
• Evaluating clustering solutions, describing clusters and estimating cluster profiles,
• Overview of other important clustering methods: mean-shift, DBSCAN, Gaussian mixtures.

Week 3: Predicting continuous data with linear and non-linear models

• Multiple linear regression and selecting suitable predictors with stepwise regression,
• Ridge and lasso regularisation,
• Regression metrics for model evaluation, comparing models,
• Polynomial regression, splines and generalised additive models (GAMs).

Week 4: Binary and multinomial classification - part 1: methods, evaluation metrics, model selection

• Introduction to classification with logistic regression - understanding probabilities and log-odds,
• Model selection and classification metrics: sensitivity, specificity, F score, Kappa, log-loss, R-squared etc.,
• Linear and quadratic discriminant analysis,
• Cross-validation and bootstrapping.

Week 5: Binary and multinomial classification - part 2: overview of other important approaches

• Stochastic Gradient Descent classifier,
• Distance-based classification: k-Nearest Neighbours algorithm,
• Probabilistic Naive Bayes classifier and kernel-based Support Vector Machines,
• Semi-automated and automated tuning of classification models.

Week 6: From decision trees to ensembles

• Classification and Regression Trees (CART),
• Estimating variable importance, bagging and boosting,
• Tree-based Random Forests ensemble,
• Extra Gradient Boosting (XGBoost) algorithm.

3. Course pre-requisites and further information.

• We recommend that all attendees have the most recent version of Anaconda Individual Edition of Python 3.8 (or at least Python 3.5) installed on their PCs (any operating system). Anaconda’s Python is a free and fully-supported distribution and you can download it directly from https://www.anaconda.com/products/individual#Downloads. Please contact us should you have any questions or issues with the installation process. A short list of additional Python libraries to pre-install before the course will be sent to the enrolled attendees in the Welcome Pack alongside other Joining Instructions.
• We recommend that the attendees have practical experience in data processing or quantitative research – gathered from either professional work or university education/research. A good knowledge of statistics would be beneficial. We suggest that the course is preceded with our “Python for Data Analysis” open-to-public tutor-led online training course.
• Your PC needs to be connected to a stable WiFi/Internet network (either home or office-based) and have Zoom video-conferencing application installed.
• You will need at least one commonly used web browser installed on your PC (e.g. Chrome, Safari, Firefox, Edge etc.) to access our Mind Project Learning Platform.

Should you have any questions please contact Mind Project Ltd at aW5mbyB8IG1pbmRwcm9qZWN0ICEgY28gISB1aw== or by phone on 0203 322 3786. Please visit the course website at https://www.mindproject.io/product/machine-learning-with-python-tutor-led-online-course-mar21/.