Biswarup G.

Biswarup G.

Sr Machine Learning Engineer

Montreal , Canada

Experience: 3 Years

Biswarup

Montreal , Canada

Sr Machine Learning Engineer

72000 USD / Year

  • Notice Period: Days

3 Years

Now you can Instantly Chat with Biswarup!

About Me

LANGUAGES AND TECHNOLOGIESProficient: Python,R, SQL, Java, Spark, Pandas, Keras, Scikit-learn, Pytorch, Hadoop,Hive, Matplotlib, Numpy, Dask, Git, Docker, ggplot2, Shiny, FlaskExposure: Javascript, c++, scala, clojure, tensorflow, gluon, mxnet, cassa...

Show More

Portfolio Projects

Description

Analyzing data and created Random Forest and Xgboost machine learning models for internal tool to help in predicting project's turnaround time and budget using Python and Scikit-learn.

Analyzed data with Pandas and Matplotlib to understand why human forecast did not work well and created deep learning model to predict total hours of project; models improved MSE by over 150%.

Show More Show Less

Description

Created model based on LSTM architecture to find rare diseases in patients via EHR data utilizing Python and Pytorch, improving ROC to 0.73 from 0.61 used by previous models.

Built RESTful API in Flask to serve model and used Docker for microservices to productionize system.

Show More Show Less

Description

Transitioned SAS Oracle based analytics system to automated analytics pipeline with Python, Pyspark for ETL, Spark-ml for Recommendation Engine and automated workflow using Airflow; reduced analysis time by 92%.

Show More Show Less

Description

An interactive tool to visualize vancouver crime data across last decadeused vancouver crime data to produce visualization tool which used animation to showcase crime data across vancouver cityUsing R, Shiny ,plotly and leaflet js library to build visualization and animation I was able to showcase crimes and different statistics over vancouver city map.Showed where crime happens often , how crime patterns changes over time in different districts of city

Show More Show Less

Description

Developed image analysis tools to enable automated diagnosis of melanoma from dermoscopic images and recommend medications from medical textsDeveloped algorithm for automated skin lesion segmentation in form of binary masks using u-net architectureAlso worked towards meaningful automatic classification of lesion images into melanoma, seborrheic keratosis and nevus using VGG-16Build models using tensorflow ,python achieved 95.6% accuracy in testing data setCombined with the analysis a recommendation engine was built which extracted first line treatment from pubmed texts for the skin lesion

Show More Show Less

Description

Chinese to English statistical machine translation system with emphasis on feedback exchange between decoder and rerankerUsed future cost based beam search decoder and PRO reranking algorithm with ordinal regression to build our decoder-reranker systemUsed python ,numpy and pandas to develop this algorithm from scratch purpose was to build statistical machine translation tool to translate chinese text to english text for Msc course project for NLP , we achieved more than 96% accuracy in train in corpus and 93% accuracy in testing corpus .

Show More Show Less