Skip to content

Latest commit

 

History

History
23 lines (21 loc) · 668 Bytes

File metadata and controls

23 lines (21 loc) · 668 Bytes

Neural Network Training

In this repo, you find the code for training a neural network to approximate an MPC that controls swing up and balancing of a cartpole pendulum system.

We recommend using a virtual enviornment to run this code:

python -m venv venv && source venv/bin/activate
pip install -r requirements.txt

Then, you can run the training with:

python train_hpc.py

And the testing with

python test_hpc.py

Once you get a satisfactory result, you can generate a stand-alone C++ implementation of the neural network:

cd embedded_nn_inference
python code_generation.py

There is a pretrained model provided in the models folder.