Saransh Mittal's Projects
In this project I have successfully developed a 2 layer neural network without using any major libraries like Keras or Tensorflow by purely using mathematical calculations to implement gradient descent and back propagation algorithms along with feed forward network
In this kaggle competition, we were tasked with creation of an algorithm that can identify a specific type of cactus in aerial imagery.
The CLI was created to interact with the language models by OpenAI
Autoencoders (AE) are a family of neural networks for which the input is the same as the output. They work by compressing the input into a latent-space representation, and then reconstructing the output from this representation.
This repository contains the benchmark script I wrote to compare the performance of NVIDIA RTX 2070 GPU and Ryzen 5 1600 CPU performance in matrix multiplication operation for Deep Learning models
Prototype of how we can use blockchain technology and create real time update system for people to monitor using BigchainDB and Web sockets
Determining the degree of damage that is done to buildings post an earthquake can help identify safe and unsafe buildings, thus avoiding death and injuries resulting from aftershocks. Leveraging the power of machine learning is one viable option that can potentially prevent massive loss of lives while simultaneously making rescue efforts easy and efficient.
This project was to create a model that can classify and predict the object in the image from the open source dataset CIFAR 10. The dataset is divided into five training batches and one test batch, each with 10000 images.
The project involves generating synthetic data for contact tracing using simple mathematic algebric formulae and then comparing performance between synthetic data and normal data while performing contact tracing on both types.
The repository contains the internship assignment given by Gojek for the position of an iOS engineer to create a application similiar to the Contacts application in Apple devices
Using convolution layers to create a generator discriminator GAN architecture to generate images using Google Street View House Number (SVHN) dataset
In this mini project, I will design an algorithm that can visually diagnose melanoma, the deadliest form of skin cancer. In particular, algorithm will distinguish this malignant skin tumor from two types of benign lesions (nevi and seborrheic keratoses).
An application to organize mock placement interviews and practice to get reviews from peers to get better at iterviews. This repository contains the iOS concept application so that a person can see the details of his past interview mock sessions, get the details and obtain a reminder of any upcoming practice sessions
Backend to sign hash value using secp256k1 curve for blockchain applications using ECDSA algorithm
This project performs emotion analysis on a statement after cleaning the data set to analyse if the emotion attached with the tweet was sad, love etc using LSTM sequence classification model
Server side code in NodeJS for IEEE Evento project
Project using DCGANs to generate human faces using celeba dataset and handwritten digits using MNIST dataset
Facial landmark detection is the task of detecting key landmarks on the face and tracking them (being robust to rigid and non-rigid facial deformations due to head movements and facial expressions)
Python script to recognise human faces using openCV framework and differentiate between human faces
Federated Learning is a distributed machine learning approach which enables model training on a large corpus of decentralised data. The repository tutorial for using PySyft for distributed training of Machine Learning model.
This project contains the Authentication API developed using Flask as the framework and Python as the programming language
Using fully connected layers to generate handwritten digits using GAN architecture on MNIST dataset
Genetic algorithms are commonly used to generate high-quality solutions to optimization and search problems by relying on bio-inspired operators such as mutation, crossover and selection.