Abhishek Sehgal

This project involved using python to construct an ML model through six different approaches including Random Classification/ Decision Tree modeling, linear, KNN, etc. It consisted of 10 main inputs with 3 iterations of mean, standard error and largest/worst figures totaling 30 inputs to classify a binary output of 1/0 using an ML classification model. The highest accuracy on test data consisting of around 113 inputs gave an accuracy of 97%. Below is the screenshot of the final output with heatmap for each model using the seaborn library in Python showing accuracy results in the title along with F scores.

This is a small script I wrote in ruby to allow customer checkout discount to be tiered based upon total cost. In Shopify, the main problem is that the checkout item cannot be discounted as a whole and thus only each line item can be discounted. So to solve this problem I generated a % discount on each line item adding up to the final discount needed in the checkout. For example if a customer shopped for $110 with three items X, Y, Z and the tiered discount range specified in the code for $110 is $20 then that $20 is spread to X,Y,Z as weighed so that total end discount is $20 with $90 being final payment. This is an important workaround for Shopify stores looking to customize discounting features according to promotions. Can only be implemented through Shopify plus.