Skip to content

Latest commit

 

History

History
3 lines (3 loc) · 925 Bytes

README.md

File metadata and controls

3 lines (3 loc) · 925 Bytes

Made from Scratch: Logistic Regression, K-means Clustering, FFNN Neural Network

Was my first attempt at building Logistic regression, K-means Clustering and a FFNN neural network from scratch whe I was first introduced to Machine Learning. The database used is the Classical Iris Database (link: https://archive.ics.uci.edu/ml/machine-learning-databases/iris/) and consists of 150 measures of sepal length, sepal width, petal length and petal width of 3 types of flowers (Iris-viriginica, Iris-setosa, Iris-versicolor). The objective in the case of Logistic Regression and FFNN Neural Network is predicting the type of flower given its sepal and petal height and width. The purpose of k-means Clustering was to try to find that there are three types of flowers (3 clusters) in the dataset by computing k-means clustering for different number of clusters k and plotting the curve of loss vs k-number (Elbow shaped graph).