This is a Classifier Algorithm that can classify German Traffic-Signs. It uses the good old convolution network inspired by the Nvidia Model used in their self-driving car.
-
Updated
Apr 3, 2018 - Jupyter Notebook
This is a Classifier Algorithm that can classify German Traffic-Signs. It uses the good old convolution network inspired by the Nvidia Model used in their self-driving car.
Behavior cloning of driving a car on a track
This project implements Behavioral Clonning from Udacity Self Driving course
In this project, you will use what you've learned about deep neural networks and convolutional neural networks to clone driving behavior. You will train, validate and test a model using Keras. The model will output a steering angle to an autonomous vehicle.
A self-driving car in a simulated environment. Explore various state-of-the-art methods of autonomous self-driving car in a fun visual format.
Udacity Self Driving Car Engineer Project - End-to-End Driving with Deep Neural Network
Clone driving behavior using a deep convolutional neural network (CNN).
This is the project Behavioral Cloning completed under the Udacity Self Driving Car Engineer Nano-degree Program
Training a robot to classify fruits using deep learning on the NVIDIA DIGITS interface.
Steering Angle prediction
The main goal of the project it to clone the behavoir of a human driver and tests it in a rally game
This is a major project that was completed under the course Deep Learning(CSE674) at University at Buffalo.
Udacity CarND Behavioral Cloning Project, Python, Tensorflow, Keras
Self Driving Car Nvidia Model
Predict steering angle of a self driving car using behaviorial cloning
Contains ML files used to train the self-driving car, its implementation using raspberry pi, lane detection algorithm, and motor control codes.
Add a description, image, and links to the nvidia-model topic page so that developers can more easily learn about it.
To associate your repository with the nvidia-model topic, visit your repo's landing page and select "manage topics."