Sheetal Bongale's Projects
All things Beer! Beer Educator and Recommender Web App | Deployed on GCP > https://alegorithm-fxljyqhslq-uc.a.run.app/ | UT Data Analysis and Visualization Nov 2019 - May 2020.
Ale-gorithm 2 is a full stack beer recommender app using Collaborative Filtering Machine Learning Models. ALE-gorithm-2 will recommend your next beer in two ways using a custom recommender model built with Surprise-scikit KNN Model | Deployed on GCP > https://alegorithm2-fxljyqhslq-uc.a.run.app
Project analyzing how COVID-19 lockdown affected the City of Austin's scooter rides and scooter complains.
Demonstrate serving a Flask Web App for City of Austin's Open Data about Shared Mobility Complaints and Shared Mobility & Dockless rides. Deployed using Heroku: https://atx-sharedmobility.herokuapp.com/
Demonstrate ETL processes in Big Data using PySpark and using AWS's Relational Databases to analyze and perform statistical analysis of Amazon's Vine program customer reviews.
Analyze, preprocess, design, optimize and train a binary classification model using neural networks with TensorFlow.
A simple command-line Blackjack game implemented using Python and Object Oriented Programming concepts
Building data pipelines to improve the multistage workflow of a ML/DL project to production. SkLearn | Keras | MLFlow | Airflow
Lexus feedback survey web-application using Python, Flask, Postgres and deployed on Heroku that sends the survey data to a Postgres database and emails admin.
Unsupervised learning project using Amazon SageMaker demonstrating data dimension reduction with PCA and clustering the cryptocurrencies using K-Means Clustering
Exploring COVID-19 Open Research Dataset (CORD-19) to apply natural language processing(NLP) to generate new insights in support of the fight against this infectious disease.
Exploring historical Dollar-Yen exchange rate futures data and apply time series analysis and forecasting modeling including linear regression, ARMA, ARIMA and GARCH to determine whether there is any predictable behavior.
Interactive data visualization using D3.js to show some state-level data about three risk factors (obesity, smoking and lack of healthcare) plotted against three perhaps underlying factors (income, poverty rate and age) based on 2014 U.S. Census data. Find it here: https://sheetalbongale.github.io/D3-Data-Journalism/
Cheat Sheets
Train and deploy a brick/ball/cylinder classifier using deep learning computer vision models.
A simple neural network using Keras and Python to solve a binary classification problem to detect an onset of diabetes.
Build, train, optimize, and deploy a machine learning model to a virtual edge device that accurately detects environmental anomalies.
Kickstart My Chart - Organize and analyze a database of 4,000 past projects in order to uncover any hidden trends.
Developed a face recognition system that can detect faces in images, identify the faces, and even modify faces with "digital makeup" like you've experienced in popular mobile apps.
Develop a model to predict the weight of a fish from their measurements by exploring various models including Linear Regression, Gradient Boosting Regressor model.
A collection of useful .gitignore templates
A series of Jupyter notebooks that walk you through the fundamentals of Machine Learning and Deep Learning in python using Scikit-Learn and TensorFlow.
Build and test multiple sci-kit learn machine learning models to predict species from flower measurements.
An interactive webpage to display a table of UFO sightings filtering data through an HTML form using JavaScript and D3.js | UT Data Analysis and Visualization Nov 2019 - May 2020
Python notebooks with ML and deep learning examples with Azure Machine Learning | Microsoft
Used Pandas and Python plotting libraries - Matplotlib & Bokeh to analyze and visualize data of a clinical trial to test four potential treatment drugs and their possible effects on tumor volume, metastasis and survival rate | UT Data Analysis and Visualization Nov 2019 - May 2020
Data Analysis using Pandas to solve 2 real world problems - 1. Analyze purchasing data for a gaming company into meaningful insights 2. Analyze the district-wide school's standardized test score results | UT Data Analysis and Visualization Nov 2019 - May 2020
Practice your pandas skills!
Built a full-stack interactive dashboard to explore the Belly Button Biodiversity DataSet using Plotly, Javascript, D3.js, Flask and SQLAlchemy. Deployed app on Heroku: https://belly-button-dashboard-sheetal.herokuapp.com | UT Data Analysis and Visualization Nov 2019 - May 2020
Python script to visualize the weather of 500+ cities across the world of varying distance from the equator and analyze the relation between latitude and different climate parameters | UT Data Analysis and Visualization Nov 2019 - May 2020
React
Typescript
Django
Laravel
Facebook
Microsoft
Google
Alibaba
D3
Tencent