Sudharsan Ananth's Projects
Augmented Reality using OpenCV live using a webcam demo. Draw a cube on top all the April-tags in the given frame. Camera calibration code is included.
Arduino library for reading Neurosky EEG brainwave data. (Tested with the MindFlex and Force Trainer toys.)
Arduino-based robotic arm controlled by brain waves. Uses EEG data extracted from a NeuroSky board to control arm movement. 🤯
carla_vpr_data_collector
Timed programming challenge to code as fast as possible and create simple mini projects.
Deep Reinforcement Learning Embodied Visual Navigation: In this project, the primary focus is on the Point-to-Point7 Navigation (Point-Nav) task when the agent is required to navigate from its initial position to a target location by relying on its egocentric visual data.
This is a 100% offline GPT4ALL Voice Assistant. Completely open source and privacy friendly. Use any language model on GPT4ALL. Background process voice detection. Watch the full YouTube tutorial for setup guide: https://youtu.be/6zAk0KHmiGw
This is a local GPT model for jetson nano
This Repository contains tutorial codes to get started with machine learning without a lot of math.
A self driving car architecture using Jetson. Supporting paralleled driving + training. multiple controller support built-in.
This is the code to train Jetrat a self driving car architecture on a local machine. Have fun!😁
This repo contains code to train MNIST and fashion MNIST using unsupervised learning and visualize clustering using dimensionality reduction like TSNE.
Model Predictive Controller running with Streamlit
This is a traditional method to detect road lanes and get steering inputs.
This consists of all the web projects.
This repo simulates a lidar (mouse cursor) mapping a 2d environment as points using pygame. The map is drawing of any building or roads.
here are questions and solutions to learn python
Reinforcement Learning DQL Agent to win every snake game.
This is where you can find code for self driving car using tensorflow and python.
Hi, This is a short description of me. Thanks for checking my GitHub profile.
This code is an implementation of simple Tabular-Q-Learning algorithm to invert a pendulum. Using only libraries such as numpy and matplotlib we can train a Reinforcement agent without even needing neural network.
In this project, we designed a novel ResNet architecture to maximize classification accuracy on the CIFAR-10 dataset while ensuring that the total number of parameters are less than 5M. We implement a random search approach for hyperparameter optimization. Our best performing architecture, TPSNet achieved a test accuracy of 94.84%.