Traffic Sign Recognition

This is a Udacity Self-Driving Car NanoDegree project submission that uses Tensorflow to recognize some types of German Traffic Signs.

Installation

Clone or fork this repository.

Usage

Intended user is the Udacity evaluator for this project. It is intended to be used in a Jupyter Notebook. Open Traffic_Sign_Classifier.ipynb to review code.

Files

Project Files

• writeup.md: writeup of project for Udacity evaluator
• Traffic_Sign_Classifier.ipynb: Jupyter Notebook of the Traffic Sign Classifier Project
• report.html: Required HTML version of the Traffic_Sign_Classifier.ipynb

Other files

• env-tsc-gpu.yml: Conda YAML file for installing environment
• signnames.csv: CSV file mapping an integer classId to name of traffic sign
• test.ipyb: Jupyter Notebook for drafting and testing python and markdown code
• visualize_cnn.png: example image used in instructions of Project Step 4
• checkpoints\: folder containing saved model data
• data\: folder containing pickled training, validation, and testing datasets
• traffic_signs\wikipedia\: folder containing German traffic signs from wikipedia for model testing
• wup_assets\: folder containing images used in writeup
• evaluation.png: Udacity evaluation of project

Evaluation