Murali Manohar Kondragunta's Projects
Models and examples built with TensorFlow
Unsupervised Statistical Machine Translation
A library for Multilingual Unsupervised or Supervised word Embeddings
Image captioning applied on VQG Dataset with beam-search
Using transformers to predict next word and predict <mask> word
Research in Word Sense Disambiguation(WorkFlow Codes)
Will be logging my daily learning progress here.
Takes a number and generates the images of those numbers. Trained on MNIST using GAN.
Becoming 1% better at everything I want
Contains implementation of models like BiLSTM CRF, Hierarchical BiLSTM for POS Tagging.
Takes screenshots of the employees' desktop at regular intervals.Records the exact time spent by the employee and has a task sheet.Files can be uploaded to the company's local server.Employees must select the client and worktype before the timer starts.HR Manager/Project Manager can view all these details using a website.IP validation,remember me are also included.
Attendance app
Repository for ongoing core project "Determining Question Relevance in the context of VQA".
Selenium based Bot to check if there are any available slots.
A desktop application(Done using NetBeans) which is based on CPU scheduling algorithms providing gantt chart,results table
Visualizing my swiggy information
NER using T5 - Template based
Built for tracking traffic during CoVID-19 pandemic lockdown.
Used CoNLL Dataset. Leveraging Transfer Learning for POS tagging.Forked from fastai
A model for finding mentions of adverse drug reactions in Twitter posts
Phrase-Based & Neural Unsupervised Machine Translation
Demo script explaining VAEs with understandable math, code and references
Uses an architecture of VGG16 with LSTM to predict.
Bot which plots time by manual writing using servo movements powered by arduino
I take PyTorch's code on Neural Style Transfer and rip it apart to see what's going on. I visualize every filter to know what's going on.
Tool to visualize heat maps of CNN models. Ideal goal is to make it work for any model & any framework.
Models and Codes for the paper Question Relevance in VQA: Identifying Non-Visual And False-Premise Questions
This website is created to showcase my university's ATV(All Terrain Vehicle) work.ATV-Team has been named as Frigators.
Crosslingual word embeddings described in our EMNLP paper
PyTorch original implementation of Cross-lingual Language Model Pretraining.