Welcome

This sample project shows how to predict the next's day currency rate using machine learning modules.

To run this sample

You need a recent installation of Python and Jupyter. However, due to the dependency on the keras library (recommended version for keras is 2.3.1), you must install a compatible Python version, such as 3.8.x or previous versions. See the library repository Readme.md for details.

If you are new to Python, check out the Beginner's Guide to Python, which contains instructions on how to install Python.

1. Register an application on FusionFabric.cloud Developer Portal, and include the Exchange Rates and Currency Conversion API.
2. Clone the current project and open it in your file explorer.
3. Copy config.csv.sample to config.csv, open it, and enter <%YOUR-CLIENT-ID%>, and <%YOUR-SECRET-KEY%> of the application created at the step 1.

The token_endpoint is provided by the Discovery Service of FusionFabric.cloud Developer Portal.

4. Open a Command Prompt or Terminal and install the machine learning libraries:

pip install datetime IPython json keras matplotlib numpy pandas pandas-datareader pygal re requests seaborn sklearn tensorflow

5. Run the sample with:

python Forex-Data-Science.py

To learn how to create this sample project from scratch, follow the tutorial from Developer Portal Documentation.

This sample client application is released under the MIT License. See LICENSE for details.