Sensing and perception example analyses

This repository contains some example logs and analyses to play around with sensing and perception algorithms and tools.

Getting started

To run the examples, proceed as follows: * Install the dependencies (preferably in a virtual Python environment) by running pip install -r requirements in the repository's root directory. * Play with the examples in ./src/run_examples.py.

You can run the convolutional network on FloydHub by executing:

floyd run --cpu --env tensorflow-1.3 --data /timkam/datasets/cifar/2:my_data "python src/floydhub.py"

If you want to run any of the other networks, adjust the code in ./src/floydhub.py accordingly.

To access the graph data used for the presentation run tensorboard --logdir=graphs/.