|
- Q1: Tutorial's datasets and code: where is the download link?
- The tutorial relies on live sensor data, as well as simulator generated data.
No need for dataset download. The tutorial's code is skeleton-based, to be filled during the tutorial,
and will be provided during the tutorial.
- Q2: Is a GPU or high performance cloud support required?
- Although a GTX-1070 based computer will be available during the tutorial for a 2-TeraFLOP demonstration,
a CPU-only [i5+] configuration is sufficient for this tutorial, as it performed 7-14 fps
for non-image based tasks, and 1fps for image-based tasks.
- Q3: Can I use TensorFlow instead of Pytorch in this tutorial?
- Yes - you can use TensorFlow, or any other framework you wish, as long as you are proficient in it.
The instructions will be provided in Pytorch only - so you'll need to translate to your own framework.
Deployment on the rover will be carried out using Pytorch only.
- More questions? Just let us know with the form below!
|
