Sai Teja Vishal Jangala's repositories
Hotspot-analysis-of-Geospatial-data
Built a Large Scale Distributed Data Processing system for Streaming Analytics using Hadoop Ecosystem (Apache Spark and HDFS), in Cloud for real-time spatial analytics.
Classification-Using-Neural-Networks-and-Deep-Learning-SVHN-Dataset
Built a Deep Learning model of Convolution Neural Network algorithm from scratch for the Classification task on the SVHN dataset. Trained the model using SGD (Stochastic Gradient Descent) optimizer with varied learning rates and epochs. Used the model to train on the training dataset and achieved a 96.25% test accuracy.
github-slideshow
A robot powered training repository :robot:
Image-Recognition-as-a-service-on-AWS
Developed an auto-scaling application that delivers Deep Learning Image Classification as a service using the IaaS model of cloud service resources to Horizontally increase the number of instances dependent on customer demand. By using SQS (Standard Queue) to queue messages received by S3 and creating the Load Balancer from scratch in Node.js, the computational costs were lowered by 95%.
ML-Supervised-Learning-Classification-and-Density-Estimation
Performed Classification and Density Estimation using Supervised Learning Algorithms by Implementing them from scratch. Algorithms implemented are - Naive Bayes and Logistic Regression
VAST-2010-Mini-Challenge-2
As part of the application to visualize 16 million records presented in the VAST Challenge by Visual Analytics Community, I implemented the plots Heat Map, Stacked Bar Chart, Kernel Density Estimation Plot, Bubble Chart, and Line Chart.