(If you're not from Udacity, or not already familiar with this because I've bored you to death explaining it, start here.)

• Project Report (PDF - GitHub will render it to HTML decently)
• index.ipynb (Jupyter notebook serving as starter point - assume you've glanced through the report)

• Python 2.7+ (should work with Python 3.5+ too as all code has proper future imports, but not tested after latest changes)
• requirements from requirements.txt

(preferably in a virtualenv or docker)

pip install -r requirements.txt

If on a machine with CUDA capable GPU, also do:

pip install -r server-requirements.txt

jupyter notebook

Start with index.ipynb Jupyter notebook.

Use full-runs-btc-24h-ohlcw-sp500-sentimen.ipynb notebook.

Use full-runs-stocks.ipynb notebook.

python btc-5min-full-run.py

(This one can take from a few hours, up to a day, on CUDA GPU machines. Pleas do not try to run on CPU only, it can take days!)