02. Project Instructions
Instructions
The Notebook
You'll be tasked with completing a project notebook Traffic_Light_Classifier.ipynb.
In this notebook, some template code has already been provided for you, but you'll need to implement additional code steps to successfully complete this project. Any code that is required to pass this project is marked with '(IMPLEMENTATION)' in the header. There are also a couple of questions about your thoughts as you work through this project, which are marked with '(QUESTION)' in the header.
Make sure to answer all questions and to check your work against the project rubric to make sure you complete the necessary classification steps!
Helper functions and testing
Also included are some additional Python files: helpers.py and test_functions.py
These provide helper functions (that load in data) and test functions that will let you test your code as you go! You do not need to change these files, but you may add to them if you want. It is especially encouraged to look at the test functions and see how they evaluate your code.
A complete classification model
In the project notebook, you will be tasked with building a classification model step-by-step. Your complete classification code should be able to take in an RGB image and output a predicted label for that image.
Criteria
You should submit your project once you meet the following two criteria:
- Greater than 90% accuracy
- Never classify red lights as green
This project is fairly open-ended and a good portfolio project that you can show off to friends, enemies, and potential employers. For this reason, you are encouraged to get as close to 100% accuracy as possible!
Evaluation
Once you have completed your project, use the Project Rubric to review the project. If you have covered all of the points in the rubric, then you are ready to submit! If you see room for improvement in any category in which you do not meet specifications, keep working!
Your project will be evaluated by a Udacity reviewer according to the same Project Rubric. Your project must "meet specifications" in each category in order for your submission to pass.
Submission
Once you have completed the notebook, make sure you've run every cell, and then submit your project from your workspace. Pressing the submit button zips all the files in the workspace and submits them for grading!