ngupta23 / data_prep_helper

A helper package for preparing and combining data from a variety of sources

Hi there, I'm Nikhil. Welcome to my Github page!

I am a "problem first" engineer focussed on solving real world challenges with the help of Data Science and Machine Learning.

My personal philosophy is that models are no good if they do not solve a real problem. Hence, when solving any data related problem, I focus on the business need first and on models later.

I also believe that we do not always need the latest and greatest algorithms to solve a problem. Yes, these have their own place for example, image classification, natural language processing - where tradititional methods are lacking. But in my personal experience (over multiple projects), a vast majority of use cases can we solved more efficiently with traditional methods and algorithms as long one is ready to invest some time and diligence in things like domain specific feature engineering.

Some facts about me:

• ✨ I have two Masters in (1) Electrical Engineering, and (2) Data Science (Machine Learning Specialization) and am currently working as a Solution Architect and Data Scientist in the Supply Chain domain.

• ✨ My personal goals include creating and contributing to more open source projects. My contributions so far include

  • Lead Developer of the Time Series Module in pycaret for automated time series modeling and machine learning in Python.
  • Lead Developer of Auto_TS for automated time series modeling in Python.
  • Lead Developer of tswgewrapped for semi automated time series analysis in R
  • Contributor to sktime - a unified framework for time series modeling in Python

• ✨ In my spare time, I volunteer for Vibha which is a non profit organization focussing on child education in India. I am their Data Analytics Lead and perform several tasks including but not limited to

  • Data Cleaning
  • Automation - Donor Data Management
  • Automation - Email Marketing
  • Donor Segmentation
  • A/B testing to increase reach
  • Reporting

• ✨ I try to learn something new about Data Analytics and Machine Learning everyday. Check out what I am reading and evaluating here.

Languages and Tools:

Connect with me:

---

📕 StackOverflow Activity

• Answer by Nikhil Gupta for How to resolve cx_Oracle.DatabaseError: DPI-1043: invalid number error?
• Answer by Nikhil Gupta for How to use tryCatch with withTimeout to timeout Rcpp function without stopping execution of script
• Answer by Nikhil Gupta for PySpark custom UDF ModuleNotFoundError: No module named
• Answer by Nikhil Gupta for How to let new and old subscribers sign up to a Mailchimp segment
• Answer by Nikhil Gupta for How can I receive a notification every time someone sign up through my MailChimp form?

---

📺 Latest Videos

• Aakash's COVID Fundraiser for the People of India
• Build and Tune Your Deep Learning Models Effectively Part 1A
• Forecasting the Next Recession (Final)
• Forecasting the Next Recession (Preliminary Univariate and EDA)

• Image Segmentation Deepdive

---