top of page
Vignesh C
Machine learning engineer
Projects (May 2018 - Present)
Welcome to my favourite part
Project | 01
Generating Faces Using General Adversarial Networks
In this project, i defined and trained a DCGAN on a dataset of faces. I first tested it on the MNIST dataset to generate new handwritten digits. Then I trained the GAN on the CelebA dataset for generating new human faces.
Project | 02
Flower Classification Using CNN
This is the final challenge of the Facebook Pytorch Challenge. Out of 10,000 scholars from 149 countries only top 300 scholars were awarded the complete scholarship based on the test accuracy, contributions to the community. I built an image classifier using pre-trained CNN models. I tried various models but DENSENET121 gave me the best accuracy. I am glad that i was among the top 3% of PyTorch scholars to get a full scholarship. The image classifier i built classified 102 species of flowers.
Project | 03
Bike Sharing Predictions With Neural Networks
In this project, i have created neural networks from scratch by using numpy only. In this project, i built a neural network and used it to predict daily bike rental ridership.
Project | 04
Generating TV Scripts using RNN
In this project, i generated my own Seinfeld TV scripts using RNNs. I used part of the Seinfeld dataset of scripts from 9 seasons. The Neural Network i built will generate a new ,"fake" TV script, based on the patterns it recognized in the training data.
Project | 04
Project | 05
Dog Breed Classifier using CNN
In this project, i have created a convolutional neural network from scratch using pytorch and also created a cnn using transfer learning in pytorch. If a dog is detected in the image, it will provide an estimate of the dog's breed. If a human is detected, it will provide an estimate of the dog breed that is most resembling.
Project | 06
Sentimental Analysis using RNN
In this project, i have created a Recurrent neural network from scratch using pytorch and deployed it using Amazon Sage Maker. I deployed my LSTM model and created a simple web page that takes a simple movie review as input and will return if the review is POSITIVE or NEGATIVE.
bottom of page