Ryan Zotti: How to Build Your Own Self Driving Toy Car | PyData DC 2016

Ryan Zotti: How to Build Your Own Self Driving Toy Car PyData DC 2016 I’ve spent the past 6 months building a self-driving toy car using a Raspberry Pi, OpenCV, and TensorFlow. If you’ve ever thought about building your own self-driving toy car, this presentation will help you avoid common pitfalls and shed light on important tradeoffs that you’ll have to weigh along the way. I’ll cover things like how to parse images, how to effectively tune machine learning neural. www.pydata.org PyData is an educational program of NumFOCUS, a 501(c)3 non-profit organization in the United States. PyData provides a forum for the international community of users and developers of data analysis tools to share ideas and learn from each other. The global PyData network promotes discussion of best practices, new approaches, and emerging technologies for data management, processing, analytics, and visualization. PyData communities approach data science using many languages, including (but not limited to) Python, Julia, and R. PyData conferences aim to be accessible and community-driven, with novice to advanced level presentations. PyData tutorials and talks bring attendees the latest project features along with cutting-edge use cases. 0:00 How to build your own Self-Driving Car 0:51 About Me 3:16 Why I did this project 6:24 Materials 10:26 Internals 13:49 Training Track 19:26 What I’ll demo today 20:19 Car camera demo 21:46 Tech Stack 22:44 Raspberry Pi 24:09 Raspberry Pi vs Arduino 28:58 Hello World for the Pi: LED light 29:41 Gradually added more complexity 30:16 Pro Tip: Don’t inhale lead 31:16 OpenCV 32:32 Haar Cascades 37:50 Stop sign detector 42:35 TensorFlow 44:43 Machine Learning 101 46:49 How a Neural Network “Sees” 50:39 Convoluted Neural Networks 52:06 Deep Learning 52:53 Standard Neural Network 56:20 Tips for tuning Neural Networks 59:11 Tensor Board 1:00:44 GPU Machine Learning 1:02:50 GPUs now commonplace 1:04:28 NVIDIA stock price 1:05:23 Machine Learning takeaways 1:09:01 Car and hardware tips 1:10:58 To Do 1:12:18 That’s it! 1:13:00 Question & Answer S/o to https://github.com/BofeiCHEN for the video timestamps! Want to help add timestamps to our YouTube videos to help with discoverability? Find out more here: https://github.com/numfocus/YouTubeVideoTimestamps