Priyanshi Gupta's Projects
Built a Multinomial Naïve Bayes classifier and compared the results with the built-in classifier in Sklearn. Developed a classifier with accuracy 82%.
A website to play the American Checkers game with an AI built using the Mini-Max Search Tree algorithm and Alpha-Beta pruning
Built my first basic app in Android Studio
CNN classifier on Oxford Pets Dataset deployed using AWS SageMaker
BERT with the huggingface PyTorch library on CoLA dataset for sentence classification and evaluate predictions using Matthew's correlation coefficient
Coded Gradient Descent for N features and came up with predictions by trying and testing with various combinations of learning rates and number of iterations.Developed a model with accuracy 93.6%
List of competitive programming practice problems and my code solution for them topic vise. Under Construction !
Designed Connect-3 Game App in Android Studio with Animation features
Successfully participated and qualified till the pre-final round in E-Yantra Nationals 19-20 and worked on the theme Biped Patrol which aimed at developing a Self-Balancing Bot embedded with a magnometer and a gyrometer.
5th Position submission for E-Yantra Nationals 2018-2019 in the track Hungry Bird: Autonomous Drone Control Using Whycon Markers.
A model-based cleaner using Laser sentence embeddings to exploit embeddings to filter misaligned segment pairs. Product scaled by asynchronously building the Task Queues, dispatching the tasks in a Round Robin method and adding multiple workers on the RabbitMQ server for consumption.
This repository illustrates the task of applying Machine Translation ( Seq2Seq Attention Network ) for Product Categorization of an E-Commerce Website data (Flipkart), classification of the description of products into the primary category of their category tree, and documenting the path to an optimal model pipeline
Code and trained model for Hybrid ranking network for text-to-SQL on WikiSQL
Implemented Machine Learning Algorithms in Hyperbolic Geometry (MDS, K-Means, Support vector machines, etc.)
Predict the correct class for each object in Cifar-10 dataset using Tensorflow and implementing Convolutional Neural Network
List and Code Solutions of 450 Important Data Structures & Algorithms Questions focused on Placement and Interview Preparation
An automated web scraper built using Selenium WebDriver to parse "food" based data from Instagram and perform a case study on it
Text Classification using BERT on SNIPS Dataset
Built and implemented the Decision Tree successfully with split on the basis of information gain from scratch
Built an Object-Oriented Program which has both user and an admin interface that allows to buy, add or customize the product
NVIDIA's NeMo and Pytorch Lighting API is used to load an audio_sample and convert it to text with QuartzNet ASR model. To convert text back to audio, spectrogram with Tacotron2 is generated first and then convert into actual audio signal using WaveGlow vocoder.
A tool to translate Multilingual Legal Document PDFs to English using Tesseract, EasyOCR and EasyNMT's Opus-MT Machine Translation model locally
25 Python Pandas tricks to work faster and write better pandas code
Everything about me is here!
Using Python Pandas & Python Matplotlib to analyze and answer business questions about 12 months worth of sales data.