Get acquainted with the two libraries at the heart of data analysis in python: NumPy and pandas: • Explain the differences between ndarrays, Series, and DataFrames • Construct and carry out basic operations with these data structures • Understand the need for efficient libraries for maths and data analysis.

Dig into the data exploration powers of pandas: • Index into and select from DataFrames • Clean and replace missing values • Combine and merge datasets

Test your data exploration skills by analysing the top 50 songs of the 2019 spotify dataset with pandas: • Group and aggregate rows • Calculate new variables • Answer statistical questions about the dataset

Organise time-series data with pandas: • Manipulate time indices • Calculate rolling statistics • Resample time series • Interpolate missing values

Make your first steps into the worlds of Natural Language Processing and Computer Vision: • Operate on python strings and build regular expressions • Tokenize and find entities in text with spacy • Manipulate images with pillow

Leverage data visualisation to explore and communicate: • Learn fundamentals of data and visual literacy • Create graphs with matplotlib • Format plots to make them more effective • Create advanced charts

Cluster a geospatial dataset using k-Means: • Explore data using clustering • Implement k-means with sklearn • Visualize geospatial data • Build a ridgeline graph

Reduce the dimensionality of a dataset using PCA: • Use dimensionality reduction to compress images • Implement PCA on an emoji dataset with sklearn • Analyse the cumulative explained variance of a PCA model • Use dimensionality reduction for data visualization

Use gaussian distributions to detect anomalies in an image dataset: • Implement Gaussian distribution anomaly detection with sklearn • Debug the optimisation of a learning algorithm • Discuss the imperfection of learning algorithms • Learn about other outlier detection methods

Understand the different types of learning algorithms and machine learning tasks: • Define features & labels • Compare supervised and unsupervised learning • Differentiate classification and regression • Explain the basics of optimisation procedures

Use linear regression to predict the happiness of countries: • Describe a regression problem • Fit a linear regression model to a dataset • Visualise the predictions of a trained linear regression model

Use logistic regression to predict the quality of wines: • Describe classification problem • Fit a logistic regression model to a dataset • Visualise the decision boundary of a trained logistic regression model

Evaluate machine learning models to predict their expected future performance: • Split train, validation, and test sets with sklearn • Run end to end machine learning experiments • Compare model quality • Tune a hyperparameter

Boost your features with scaling and polynomials to improve model accuracy: • Scale and descale a feature matrix using standardization • Calculate polynomial features • Use both methods to optimise the fitting of a regression model

Regularize a model to improve its generalisation properties: • Overfit a polynomial regression model • Use ridge regularization to prevent overfitting • Analyse the effect of the regularization weight on model parameter sizes • Train a regularized logistic regression model with polynomial features

Use SVMs to detect fraudulent banknotes: • Train linear & RBF kernel SVMs • Visualize and compare large margin classifier decision boundaries • Analyse the effect of regularization hyperparameters

Continue detecting fake banknotes with Naive Bayes & Decision Trees: • Train & evaluate a Naive Bayes classifier • Train & evaluate a Decision Tree • Carry out error analysis to better understand model hyperparameters

Strength in numbers, combine models to improve accuracy: • Use bagging to train random forest models • Use boosting to train AdaBoost and XGBoost models • Analyse accuracy vs performance tradeoffs in ensemble models