In this project I attempt to predict NYC traffic using LightGBM, RNN, GRU and LSTM models.

In this project I created basic chatbot which identifies audio messages, converts them to .wav format with 16khz frequency. In addition, the bot identifies pictures images and saves only those that contain human face. 

Here I experiment with diffusion model using Denoising Diffusion Probabilistic Models (DDPM) and Denoising Diffusion Implicit Models (DDIM) 

In this project we will be creating the Autoencoder model for denoising handwritten images using PyTorch.

This repository contains a data processing workflow for the Energy Forecasting Hackathon. For this forecasting task, the ARIMA (AutoRegressive Integrated Moving Average) model was selected. 

In this project I created a model helping us to estimate credit card approval rate based on the features provided. I used the dataset from Kaggle named Credit card Details Binary Classification Problem.

In this project, I explore the fundamentals of machine learning using the k-nearest neighbors (KNN) algorithm for univariate and multivariate models. In addition, I apply hyperparameter optimization and various cross-validation methods to assess the performance of the model.

In this project, using mainly Matplotlib, I attempt to shed the light on some interesting data regarding Star Wars episodes, characters, fans and provide answers to some important questions, including 'Who shot first, Han or Greedo?'

During this project I applied both exploratory data visualization (graphs used for ourselves to better understand and explore data) as well as explanatory data visualization (creating graphs for others to inform, make a point, or tell a story) techniques. I maximized the data-ink ratio, created visual patterns with Gestalt principles and used pre-attentive attributes. 

In this project, I am going to study the practical side of the multinomial Naive Bayes algorithm by building a spam filter for SMS messages. 

In this project, I work with a dataset of Jeopardy questions to identify patterns that could potentially improve one's chances of winning this famous US TV show. I will be using the chi-squared test for categorical data, which allows me to determine the statistical significance of observing a set of categorical values.

In this project I intend to write functions required for the mobile app which helps lottery addicts better estimate their chances of winning a lottery. I write functions to calculate factorials and combinations, which will be used repeatedly throughout the project. 

In this project, I analyse survey data to find the best two markets to advertise in. An e-learning company offering programming courses wants to identify the two best markets for advertising their subscription product. To minimize costs, data-driven approaches should be explored before considering expensive market surveys.

In this project I apply all the knowledge learned about sampling, variables, scales of measurement, and frequency distributions. I learned about each of these topics in isolation. In this guided project, I attempt to make one step further and learn to combine all these skills to perform practical data analysis.

In this project I attempt to scrape a sports stats website using BeautifulSoup and Requests libraries in order to get data needed for building a prediction model on winning football games in the Premier League.

In this project I attempt to investigate the correlations between SAT scores and demographic factors in New York City public schools including race, gender, AP exams and more. 

This project is aimed at applying data cleaning skills with pandas to practice.

In this repository I clean and prepare the data for the project aimed at identifying the severity of cough due to air pollution in New Delhi using audio analysis and Machine Learning.