For implementing Machine learning algorithm I needed various tools csv data extraction, tensor processing, saving model etc. So I implemented a tensor class that represents multidimensional data. I used operator overloading to implement most mathematical operations like addition, subraction, multiplication and division. I also implementated operations like mean,log, exp etc. Next I needed to parse csv data. I created a parser to parse csv data then created a datastructure called frame to store the data. I has various indexing features like getting specific columns and rows. For saving the models. I save the trained weights in json. I also created random train test split to split the dataset into train and test. Utils.
Basic Adaline Model for classification. Adaline.
Basic Logistic Regression for classification. Logistic Regression
Created confusion matrix and various evaluation metrics like accuracy, precision, recall and f1 score. Evaluation