https://github.com/manajitpal/deeplearningforselfdrivingcars

This is an implementation in Pytorch of nvidia's model to build a deep learning neural network for self-driving cars.
https://github.com/manajitpal/deeplearningforselfdrivingcars

Last synced: about 2 months ago
JSON representation

This is an implementation in Pytorch of nvidia's model to build a deep learning neural network for self-driving cars.

• Host: GitHub
• URL: https://github.com/manajitpal/deeplearningforselfdrivingcars
• Owner: ManajitPal
• Created: 2018-12-29T13:00:58.000Z (almost 6 years ago)
• Default Branch: master
• Last Pushed: 2019-02-20T07:17:29.000Z (almost 6 years ago)
• Last Synced: 2023-11-07T00:27:45.855Z (about 1 year ago)
• Language: Jupyter Notebook
• Size: 1.52 MB
• Stars: 21
• Watchers: 3
• Forks: 20
• Open Issues: 2
• Metadata Files:
  • Readme: readme.md

Awesome Lists containing this project

README

        

# **Deep Learning For Self-Driving Cars** 

**Goals**

The goals / steps of this project are the following:

* Use the simulator to collect data of good driving behavior.

* Build, a convolution neural network in Pytorch that predicts steering angles from images.

* Train and validate the model with a training and validation set.

* Test that the model successfully drives around the track without leaving the road.

---

### This project is a part of the tutorial published [here](https://medium.com/@manajitpal/deep-learning-for-self-driving-cars-7f198ef4cfa2).

### Overview

#### 1. Project contents:

The project includes the following files:

* self_driving_car.py containing the script to create and train the model.

* Self_Driving_Car.ipynb containing the notebook code to create and train the model.

* model.py as a helper function for loading the required model in drive.py.

* drive.py for driving the car in autonomous mode.

* model.h5 containing a trained convolution neural network.

* Note: This project uses python 3 version.

#### 2. Running the model

Using the Udacity provided simulator and the drive.py file, the car can be driven autonomously around the track by executing 

```sh

python drive.py model.h5

```

#### 3. Readability

The self_driving_car.py/Self_Driving_Car.ipynb file contains the code for training and saving the convolution neural network. The file shows the pipeline used for training and validating the model, and it contains comments to explain how the code works. The tutorial for the project is [here](https://medium.com/@manajitpal/deep-learning-for-self-driving-cars-7f198ef4cfa2).

### Model Architecture

The project is based on the research by Nvidia. The paper is available [here](https://arxiv.org/pdf/1604.07316v1.pdf). The network is a CNN which maps raw pixel inputs along with steering data from a log file and clones the behaviour of the steering in autonomous mode. For more in depth explanation, please refer to the tutorial.