This is a basic self-steering car model implemented in TensorFlow. It is used as a demo project to get started with the Clusterone deep learning computation platform.
Follow the Getting Started guide for Clusterone and read the author's blog post about building this demo here and here.
To run this project, you need:
- Python 2.7 or 3.5 (or higher).
- Git and Git Large File Storage (LFS)
- The TensorFlow Python library. Get it with
pip install tensorflow - The Clusterone Python library. Install it with
pip install clusterone - A Clusterone account. Sign up here for free if you don't have an account yet.
Running a job on Clusterone is simple with the just command line tool that comes included with the Clusterone Python package.
To run the model on Clusterone, you first need a Clusterone account. Log in with just login.
cd into the directory where you cloned this repository to and create a new project with just init project self-driving-demo. The data is already uploaded to Clusterone, so you don't need to worry about it.
Push the project code to Clusterone with git push clusterone master.
When the upload is complete, create a job to run the model on Clusterone:
just create job distributed --name first-job --project self-driving-demo \
--datasets tensorbot/self-driving-demo-data --module main_tfNow the final step is to start the job:
just start job -p self-driving/first-jobYou can monitor the execution of your job on Clusterone using just get events.
Instead of running the model from the command line, you can also use Clusterone's graphical web interface Matrix.
For a more detailed guide on how to run this project on Clusterone, check out the Getting Started guide. To learn more about Clusterone, visit our website.
MIT © Clusterone Inc.
Comma dataset and data_reader.py by comma.ai, licensed as specified.