supervised_learning

Repository of notebooks used to complete assignment 1. The code covers analysis that was desired for assignment 1 but its not very cohesive, but all code required for a particular model is found within the notebook with the models name.

Usage

Assuming python 3.9 and conda are installed create the env using. conda env create -f ML.yml

Open jupyter jupyter notebook execute code as needed

Template used for Latek

https://www.overleaf.com/project/5ebe55ec08e09b0001676b6a

Make a directory

mkdir data download both of these files into data

https://www.kaggle.com/uciml/adult-census-income

https://www.kaggle.com/uciml/indian-liver-patient-records

Note

The parent directory path inside the notebooks will need to be changed to reflect absolute path to the data directory

Code used to aid in discovery

https://scikit-learn.org/stable/auto_examples/model_selection/plot_learning_curve.html

https://medium.com/datadriveninvestor/understanding-adaboost-and-scikit-learns-algorithm-c8d8af5ace10

https://www.kaggle.com/alexandrago/income-prediction-xgbclassifier-auc-0-926

https://towardsdatascience.com/accuracy-precision-recall-or-f1-331fb37c5cb9

https://scikit-learn.org/stable/modules/model_evaluation.html

https://scikit-learn.org/stable/auto_examples/model_selection/plot_multi_metric_evaluation.html#sphx-glr-auto-examples-model-selection-plot-multi-metric-evaluation-py

https://scikit-learn.org/stable/modules/learning_curve.html