IntroductionWhat is machine learning?Getting started with Scikit-LearnWhat will we cover today?

---

Figure 1

Image 1 of 1: ‘An infographic showing some of the relationships between AI, ML, and DL’

The image above is by Tukijaaliwa, CC BY-SA 4.0, via Wikimedia Commons, original source

---

Figure 2

Image 1 of 1: ‘Types of Machine Learning’

Figure from the Python Data Science Handbook

---

Figure 3

Image 1 of 1: ‘Types of Machine Learning’

Image from Vasily Zubarev via their blog with modifications in blue to denote lesson content.

---

Supervised methods - RegressionSupervised learningRegression

---

Figure 1

Image 1 of 1: ‘Example of linear and polynomial regressions’

Example of linear and polynomial regressions

---

Figure 2

Image 1 of 1: ‘Comparison of the regressions of our dataset’

Comparison of the regressions of our dataset

---

Figure 3

Image 1 of 1: ‘Comparison of the regressions of our dataset’

Comparison of the regressions of our dataset

---

Figure 4

Image 1 of 1: ‘Comparison of the regressions of our dataset’

Comparison of the regressions of our dataset

---

Figure 5

Image 1 of 1: ‘Comparison of the regressions of our dataset’

Comparison of the regressions of our dataset

---

Supervised methods - ClassificationClassification

---

Figure 1

Image 1 of 1: ‘Artwork by @allison_horst’

Artwork by @allison_horst

---

Figure 2

Image 1 of 1: ‘Artwork by @allison_horst’

The physical attributes measured are flipper length, beak length, beak width, body mass, and sex.

---

Figure 3

Image 1 of 1: ‘Visualising the penguins dataset’

Visualising the penguins dataset

---

Figure 4

Image 1 of 1: ‘Visualising the penguins dataset’

Visualising the penguins dataset

---

Figure 5

Image 1 of 1: ‘Decision tree for classifying penguins’

Decision tree for classifying penguins

---

Figure 6

Image 1 of 1: ‘Decision tree for classifying penguins’

Decision tree for classifying penguins

---

Figure 7

Image 1 of 1: ‘Classification space for our decision tree’

Classification space for our decision tree

---

Figure 8

Image 1 of 1: ‘Performance of decision trees of various depths’

Performance of decision trees of various depths

---

Figure 9

Image 1 of 1: ‘Simplified decision tree’

Simplified decision tree

---

Figure 10

Image 1 of 1: ‘Classification space of the simplified decision tree’

Classification space of the simplified decision tree

---

Figure 11

Image 1 of 1: ‘Classification space generated by the SVM model’

Classification space generated by the SVM model

---

Ensemble methodsEnsemble methods

---

Figure 1

Image 1 of 1: ‘Stacking’

Stacking

---

Figure 2

Image 1 of 1: ‘Stacking’

Stacking

---

Figure 3

Image 1 of 1: ‘Stacking’

Stacking

---

Figure 4

Image 1 of 1: ‘Random Forests’

Random Forests

---

Figure 5

Image 1 of 1: ‘random forest trees’

random forest trees

---

Figure 6

Image 1 of 1: ‘random forest clf space’

random forest clf space

---

Figure 7

Image 1 of 1: ‘Regressor predictions and average from stack’

Regressor predictions and average from stack

---

Unsupervised methods - ClusteringUnsupervised learningClustering

---

Figure 1

Image 1 of 1: ‘Plot of the random clusters’

Plot of the random clusters

---

Figure 2

Image 1 of 1: ‘Plot of the fitted random clusters’

Plot of the fitted random clusters

---

Figure 3

Image 1 of 1: ‘An example of kmeans failing on non-linear cluster boundaries’

An example of kmeans failing on non-linear cluster boundaries

---

Figure 4

Image 1 of 1: ‘Kmeans attempting to classify overlapping clusters’

Increasing n_samples to 4000 and cluster_std to 3.0 looks like this: The straight line boundaries between clusters look a bit strange.

---

Figure 5

Image 1 of 1: ‘Spectral clustering on two concentric circles’

Spectral clustering on two concentric circles

---

Figure 6

Image 1 of 1: ‘Spectral clustering viewed with an extra dimension’

Spectral clustering viewed with an extra dimension

---

Figure 7

Image 1 of 2: ‘Kmeans attempting to cluster the concentric circles’

Image 2 of 2: ‘Spectral clustering on the concentric circles’

---

Unsupervised methods - Dimensionality reductionDimensionality reductionDimensionality reduction with Scikit-Learn

---

Figure 1

Image 1 of 1: ‘MNIST example illustrating all the classes in the dataset’

MNIST example illustrating all the classes in the dataset

---

Figure 2

Image 1 of 1: ‘MNIST example of a single image’

MNIST example of a single image

---

Figure 3

Image 1 of 1: ‘SKLearn image with highlighted pixels’

SKLearn image with highlighted pixels

---

Figure 4

Image 1 of 1: ‘SKLearn image with highlighted pixels’

SKLearn image with highlighted pixels

---

Figure 5

Image 1 of 1: ‘Reduction using PCA’

Reduction using PCA

---

Figure 6

Image 1 of 1: ‘Reduction using PCA’

Reduction using PCA

---

Figure 7

Image 1 of 1: ‘Reduction using PCA’

Reduction using PCA

---

Figure 8

Image 1 of 1: ‘Reduction using PCA’

Reduction using PCA

---

Figure 9

Image 1 of 2: ‘Reduction using PCA’

Image 2 of 2: ‘Reduction using PCA’

---

Figure 10

Image 1 of 1: ‘Reduction to 3 components using pca’

Reduction to 3 components using pca

---

Figure 11

Image 1 of 1: ‘Reduction to 3 components using tsne’

Reduction to 3 components using tsne

---

Neural NetworksNeural networks

---

Figure 1

Image 1 of 1: ‘A diagram of a perceptron’

A diagram of a perceptron

---

Figure 2

Image 1 of 1: ‘A multi-layer perceptron’

A multi-layer perceptron

---

Ethics and the Implications of Machine LearningEthics and machine learningEthics of machine learning in research

---

Find out moreOther algorithms

---