Projects in Awesome Lists by shilpakancharla
A curated list of projects in awesome lists by shilpakancharla .
https://github.com/shilpakancharla/event-based-velocity-prediction-snn
Neuromorphic computing uses very-large-scale integration (VLSI) systems with the goal of replicating neurobiological structures and signal conductance mechanisms. Neuromorphic processors can run spiking neural networks (SNNs) that mimic how biological neurons function, particularly by emulating the emission of electrical spikes. A key benefit of using SNNs and neuromorphic technology is the ability to optimize the size, weight, and power consumed in a system. SNNs can be trained and employed in various robotic and computer vision applications; we attempt to use event-based to create a novel approach in order to the predict velocity of objects moving in frame. Data generated in this work is recorded and simulated as event camera data using ESIM. Vicon motion tracking data provides the ground truth position and time values, from which the velocity is calculated. The SNNs developed in this work regress the velocity vector, consisting of the x, y, and z-components, while using the event data, or the list of events associated with each velocity measurement, as the input features. With the use of the novel dataset created, three SNN models were trained and then the model that minimized the loss function the most was further validated by omitting a subset of data used in the original training. The average loss, in terms of RMSE, on the test set after using the trained model on the omitted subset of data was 0.000386. Through this work, it is shown that it is possible to train an SNN on event data in order to predict the velocity of an object in view. (Spring 2022 MS Computer Science Thesis - North Carolina State University)
event-camera neuromorphic-computing regression-models spiking-neural-networks
Last synced: 19 Nov 2024
https://github.com/shilpakancharla/snn-networking
Neuromorphic computing is a new technology which uses very-large-scale integration (VLSI) systems containing electrical analogs that mimic neurobiological structures. Such technology can be simulated using the Python framework snnTorch, a derivative of the deep learning framework PyTorch that has the capacity to run spiking neural networks (SNNs) which in turn can be run on neuromorphic chips. This project serves to explore the use of the snnTorch framework on a networking dataset from the Barcelona Neural Networking Center. Not only are the performance results and construction of a spiking neural network seen from this experiment, but also important characteristics of types of data that can be collected that will better fit the spatio-temporal data characteristics that are ideal to be run in an SNN. Moreover, cloud computing tools from Amazon Web Service (AWS) such as S3 and SageMaker were used to aid in the completion of running deep learning experiments. This work serves as an artificial intelligence proof of concept for the snnTorch framework and cloud computing in the deep learning environment.
Last synced: 19 Nov 2024
https://github.com/shilpakancharla/galaxy-image-classification
SDSS telescopes have captured over 40 TB worth of galaxy images and classification of these images is the first step towards obtaining a deeper understanding of physical processes within them, star formation, and the nature of the universe. Since we could not find an easily accessible dataset for galaxy classification, we compiled a dataset for Galaxy classification and provided benchmarks with some of the common learning algorithms that would help in automating the galaxy classification which until recently had to be performed by hand by expert astronomers. We classify the images of galaxies into four classes: spiral, elliptical, irregular, and invalid.
astronomical-images astronomy astropy big-data convolutional-neural-networks deep-learning galaxy-classification image-classification machine-learning multilayer-perceptron-network principal-component-analysis random-forest support-vector-machines
Last synced: 11 Apr 2025
https://github.com/shilpakancharla/quantumlearning
Designed two-layer neural network using TensorFlow to find solutions to Schrödinger's equation. Used mean square distance to assess performance of the model.
deep-learning gradient-descent machine-learning machine-learning-algorithms neural-network quantum-mechanics
Last synced: 19 Jan 2025
https://github.com/shilpakancharla/gams-problems
Repository of general algebraic modeling problems using linear programming for different scenarios (STOR 415).
Last synced: 13 Mar 2025
https://github.com/shilpakancharla/deep-rl-lunar-lander
Using a deep Q-learning network and searching for optimal hyperparameters in order to solve the lunar lander problem provided by OpenAI Gym.
deep-reinforcement-learning hyperparameter-optimization lunar-lander openai-gym openai-gym-environments
Last synced: 13 Mar 2025
https://github.com/shilpakancharla/supernova
Computational model for supernovae neutrino physics.
Last synced: 13 Mar 2025
https://github.com/shilpakancharla/graphplan
Graphplan is an automated planning algorithm that takes as input a planning problem expressed in STRIPS and produces, if one is possible, a sequence of operations for reaching a goal state. I have implemented it here, alongside a few problem examples.
algorithms-and-data-structures artificial-intelligence artificial-intelligence-algorithms artificial-intelligence-application automated-planning graphplan-algorithm java-application
Last synced: 13 Mar 2025
https://github.com/shilpakancharla/ml-for-bank-data
Feature engineering project to build a model that predicts if clients will subscribe to term deposit based on marketing call campaign data from a Portuguese bank. Used bagging, random forests, boosting, logistic LASSO regression with and support vector machines to study trends and optimize bank actions for maximizing bank account subscriptions.
Last synced: 13 Mar 2025
https://github.com/shilpakancharla/metoo-classification
Social media provides a unique opportunity to shed light on phenomena that have previously been ignored or swept under the rug. The #MeToo movement began in 2006 when Tarana Burke first coined the phrase, but a viral Twitter post in 2017 served to lift the hashtag into the mainstream accompanied by the public reckoning of several high-profile men. This hashtag allows people to seek solidarity on a much larger scale in calling out the perpetrators of their abuse. We are going to categorize tweets with #MeToo into the years they were posted. In particular, we look at the years 2017, 2018, and 2019.
classification convolutional-neural-networks datasets fasttext hashtag kaggle lime lime-explainer lstm-neural-networks metoo-movement metoo-tweets social-media
Last synced: 13 Mar 2025
https://github.com/shilpakancharla/rover
2017-2018 project using Arduino to program functionality of rover and AutoCAD for hardware design.
Last synced: 13 Mar 2025
https://github.com/shilpakancharla/numerical-analysis
Library of numerical analysis methods.
Last synced: 13 Mar 2025
https://github.com/shilpakancharla/optimization
Optimization methods (STOR 415)
Last synced: 13 Mar 2025
https://github.com/shilpakancharla/hate-speech
As citizens, how can we keep track of hate speech online that's affecting our fellow peers and neighbors? While I believe there are myriad solutions to helping each other out, I wanted to try a solution using machine learning models. Machine learning classifiers, alongside a vast amount of data gathered through API calls, can offer valid solutions to organizations and companies attempting to monitoring content on their platforms. Ultimately, a logistic regression classifier was created that achieved 98% accuracy when classifying tweets as positive (non-offensive) and negative (offensive).
convolutional-neural-networks decision-trees flask jupyter-notebook logistic-regression-algorithm machine-learning-algorithms pipeline python twitter-api
Last synced: 13 Mar 2025
https://github.com/shilpakancharla/machine-learning-tutorials
Machine Learning (STOR 565)
Last synced: 13 Mar 2025
https://github.com/shilpakancharla/terrain-identification
Given time series data from an accelerometer and gyroscope attached to the prosthetic lower leg limb of 10 subjects, we create two different models (1D CNN and bidirectional LSTM) to investigate if we can classify the type of terrain a user is traversing.
bidirectional-lstm cnn lower-limb-amputations series-data terrain-identification
Last synced: 11 Apr 2025