About: Numerical solver for ODE's for simulated flying insect

1. Run simulations to generate more training data for Neural Networks (parallel sims)
2. Use newly generated training data to train NNet
3. *Build a bigger ball for control parameters.
4. *Penalize high values for controls or speeds. – sift through training and test data to take out “bad” controls.
5. *Try Mixture Density Network
6. *Tru reinforcement learning

• a. Open terminal and navigate to a directory where you want this folder (ex. .../Documents/GitRepos)

• b. clone this repository (download it onto your local computer)

  git clone https://github.com/nifti-coe/ODE_Python.git

• a. Open terminal and navigate to the folder where this repository is stored (ex. .../Documents/GitRepos/ODE_Python)

• b. "pull" the most recent version

  git pull

• a. Open terminal and navigate to the folder where this repository is stored (ex. .../Documents/GitRepos/ODE_Python)

• b. "add" your changes (stage changes). Note that the "." means to add everything in the folder

  git add .

• c. "commit" your changes. This will create a checkpoint on your local computer. include a commmit message with the "-m" flag.

  git commit -m "short commit message"

• d. "push" your changes to the web.

  git push