Coincaster – time series forecasting on cryptocurrency
Project description
Would you pass on the power to predict the future? I certainly won’t, however, in this little project unfortunately we will not obtain the power to predict ‘all’ of the future, we shall focus on the price of cryptocurrencies. The ‘magic’ that we are going to use is the tools and techniques employable by the python programming language and associated packages. That shall be combined with time series data on various cryptocurrencies and sentiment data that we infer from Google search trends and Subreddit metrics.
We name this project “Coincaster”.
I have to give credit to an excellent blog post that sparked some of the investigations undergone in this project: https://towardsdatascience.com/demystifying-cryptocurrency-price-prediction-5fb2b504a110
Table of contents
- Coincaster – Get mongodb running on a Raspberry Pi 3
1. Coincaster – Get mongodb running on a Raspberry Pi 3
To store the data for the coincaster project, we will use a Raspberry Pi 3. In this notebook, all the steps to get the mongo database running on the Raspberry Pi 3 (RPi3) are covered. We will also allow remote access to the database and connect to it using pymongo.
All the steps to do that are described in detail here:
https://github.com/kromerh/coincaster/blob/master/04.Analysis/2020-07-10.Setup_mongdb_on_RPi3.ipynb