PyTorch

Sentence Generation and Classification with Variational Autoencoder and BERT
A conditional variational autoencoder uses a sentence premise and lass label to generate a hypothesis specific to that class type. Then, a BERT model is used to classify the results in order to evaluate the style transfer to generating the sentences.
PDF Slides Repo
Sentence Generation and Classification with Variational Autoencoder and BERT
Image Captioning using an LSTM Network
Created a deep learning network for automatic image captioning of the COCO dataset using a ResNet 50 CNN encoder and a LSTM RNN decoder.
PDF
Image Captioning using an LSTM Network
Convolutional Networks for Semantic Segmentation
Built various deep CNN architectures and trained them to predict the segmentation masks of the images in the India Driving Dataset as a semantic segmentation task.
PDF
Convolutional Networks for Semantic Segmentation