ML Demos

A collection of implementions for an assortment of machine learning and deep learning algorithms with basic overviews/explanations written using Jupyter notebooks. A list of implemented algorithms can be found below.

The purposes of these implementations is to emphasize familiarity with the built-in functions available in the scikit-learn and PyTorch libraries, so they only make use of the scikit-learn built-in.

List of Implmented Algorithms:

ML_demos/basic:

• Regression.ipynb: linear and logistic regression
• naiveBayes.ipynb: Naive Bayes
• SVM.ipynb: support vector machine
• kNN.ipynb: k-nearest neighbors
• Trees.ipynb: decision tree and random forest
• Clustering.ipynb: k-Means and DBScan
• XGBoost.ipynb: xgboost
• fromScratch.ipynb: various ML evaluation metrics, sampling, kNN, and feed-forward neural network from scratch (coding interview practice exercise)

ML_demos/deep_learning:

• DNN.ipynb: simple multilayer perceptron implementation, implement a basic neural network
• CNN.ipynb: implements a basic CNN
• CNN_architectures.ipynb: implements some known CNN achitectures including LeNet, AlexNet, VGG, and ResNet
• RNN.ipynb: implements an RNN with LSTM for time series data

ML_demos/reinforcement_learning:

To implement:

• General deep learning
  • Transformer
  • GAN
  • Autoencoder
• RL algorithms