Data Science, Machine Learning and blockchain development enthusiast, currently looking for winter 2021 internships and full time roles.
Previously interned at Artivatic.ai where I worked on building a medical doc extractor which extracts texts from medical
bills and converts them to json/dict format.
Also interned at Vidgyor Media Technologies working with Python/OpenCV and Deep Learning on TV and OTT live streams.
I also worked at the Mobile Computing and Network Research Group(MCNRG), NIT Durgapur, working with live AQI data of
Delhi.
I also co-head the coding club of my college - Codechef KGEC Campus Chapter where we conduct several programs throughout
the year to promote a culture of coding and provide training/guidance to students.
Built a medical doc extractor which extracts texts from medical bills and converts them to json/dict format.
Converted PDFs and images to text files using tesseract .
Used fuzzywuzzy to detect important keys from medical bills and extracted the values in dict format in python.
Technologies used:
Pytesseract, Django, Fuzzywuzzy, Postman
Worked on cutting edge ad-detection and monetization using Python, OpenCV and YoloV3 on live streams.
Built an ad-detection algorithm on live stream using:
1. Machine Learning (opencv) --
Detected the starting and ending points of commercial breaks using correlogram, orb and flann
based matcher.
2. Deep Learning (Yolo_v3) --
Created a signature for each ad using yolo_v3 working on the coco dataset
Detected the starting points of commercial breaks on live streams using the signature created
CGPA - 7.8
A keras model to predict traffic volume from time series data.
Tech used : Python, Jupyter Notebook, Keras, XGBoost, Pandas, Numpy
Facilities : This model predicts the traffic volume of a place from
weather data
A machine learning model which predicts the current contraceptive method choice (no use, long-term methods, or short-term methods)
of a woman based on her demographic and socio-economic characteristics.
Tech used : Python, Jupyter notebook, sklearn