Code used to obtain results for my Medium articles - medium@conorosullyDS
| Article | Code | Description |
|---|---|---|
| The Power of Feature Engineering | feature_engineering.ipynb | Compare the performance of logistic regression to a NN Classifier on a non-linear dataset |
| Identifying Restaurant Hotspots with a Gaussian Mixture Model | gmm_restaurant.ipynb | Using a GMM to identify intuitive restaurant clusters in Toronto, Canada |
| Visualising the Classification Power of Data using PCA | pca_visualisation.ipynb | Using PCA to explore how well your data can separate classes |
| Deep Neural Network Language Identification | language_classification.ipynb | Classifying the language of a piece of text using a DNN and character n-grams |
| Too Many Terms Ruins the Regression | polynomial_regression_overfitting.ipynb | Overfitting with polynomial regression and how to avoid it |
| US Election Choropleth with Python | US_election_map.ipynb | Creating time-series choropleths of US election results |
| Finding and Visualising Interactions | interactions.R | Analysing interactions using feature importance, Friedman’s H-statistic and ICE Plots |
| Introduction to Sentiment Analysis | sentiment_analysis.ipynb | Creating your first sentiment analysis model with Python |
| Finding and Visualising Non-Linear Relationships | nonlinear_relationships.R | Analysing non-linear relationships with partial dependence plots (PDPs), mutual information and feature importance |