akshaybabloo / car-nd-term-1

Udacity "Self-Driving Car Engineer" Nanodegree files - January 2017

Self-Driving Car Engineer Nanodegree

Udacity "Self-Driving Car Engineer" Nanodegree files

• Documentations
• Term 1
  • 1 Finding Lanes
  • 2 MiniFlow
  • 3 TensorFlow
  • 4 Deep Neural Network
  • 5 Convolutional Neural Network
  • 6 Traffic Sign Classifier
  • 7 Keras
  • 8 Transfer learning
• License and Disclaimer

Documentations

To build Python documentations

Make sure you have installed Node.js 6

Once you have installed Node.js, do the following

1. Open the Car-ND folder in terminal or Command Prompt.
2. Type in the following once

   npm install
   

3. To build the Python docs, type in

   gulp
   

If there are no errors, the index.html will be in build/html/index.html, double clicking that will open in a web browser.

or

Open docs folder in Command Prompt or Terminal and type in

make html

Term 1

Term 1 is mainly focused on finding the lanes and objects correctly using various algorithms and computer vision.

1 Finding Lanes

In this section we were introduced to the basic knowledge of Computer Vison using OpenCV and Python. The files are available here.

More on this in my blog Part 1.1 Car-ND: Detect road lanes using Computer Vision and Python 3

Outcome

2 MiniFlow

A Stochastic Gradient Descent based Backpropogation algorithm to minimise the error.

Files can be found here

3 TensorFlow

An introduction to TensorFlow.

Files can be found here

4 Deep Neural Network

Files for DNN can be found here.

5 Convolutional Neural Network

Files for CNN can be found here

6 Traffic Sign Classifier

Files for CNN can be found here

7 Keras

Files for CNN can be found here

8 Transfer learning

Files for CNN can be found here

License and Disclaimer

All product and company names are trademarks™ or registered® trademarks of their respective holders. Use of them does not imply any affiliation with or endorsement by them. Except as otherwise noted, the content of this page is licensed under the Creative Commons Attribution 3.0 License, and code samples are licensed under the MIT License.