Abhishek Sharma's repositories
DASS_Project
This project aims to turn 2D photographs provided by the user into 3D versions that can be lip-synced with an audio file of the user's choice. The final product will be a lifelike video of the subject. The project also considers future additions such as integration with different backgrounds, text inclusion, and social media sharing.
StackOverFlow_Duplicate_Question_Detection
DupPredictor is a framework to detect duplicate questions on Stack Overflow using machine learning techniques . The algorithm consists of LDA (latent Dirichlet Allocation) for topic modelling to classify the text into topics.
AAD-Project
A course project which is used to compare few of graphs algorithms and on the basis of that there is debt simplifier which gives the best way to minimise the number of transactions.
COC-lite
A 2 D game in Python3 (terminal-based), heavily inspired by Clash of clans where the user will control the king, move it up, down, forward and backward, while destroying buildings and fighting defences on its way. Concepts of object oriented programming is present within the code and the game is a basic version of Clash of clans .
Computer_System_Organisation
Assignments Solved in CSO course
DigitsSumNeuralNetwork
A Convolutional Neural Network (CNN) model for the task of predicting the sum of digits in an image based on MNIST training.
FoodPortal_Mern
A Food Ordering Portal following where users have the option to search and order various food items. At the same time, various vendors have the ability to list food items.
hw1
A place to put the first assignment.
Two-Phase-Commit-Protocol
This is a simple implementation of the two-phase commit protocol. It is a distributed transaction management protocol that ensures that a transaction can be committed or rolled back across multiple distributed resources.
Word-Sense-Disambiguation
The repository contains the implementation of four baseline models for Word Sense Disambiguation, including Lesk, Naive Bayes, K-Nearest Neighbors, and BiLSTM. Additionally, a BERT-based model has been included for comparison. The models have been trained on the SemCor dataset and evaluated on the Senseval and SemEval datasets