https://github.com/bukosabino/btctrading

Time Series Forecast with Bitcoin value, to detect upward/down trends with Machine Learning Algorithms
https://github.com/bukosabino/btctrading

bitcoin deep-learning deep-neural-networks forecast future machine-learning machine-learning-algorithms prediction price time-series trends xgboost

Last synced: 14 days ago
JSON representation

Time Series Forecast with Bitcoin value, to detect upward/down trends with Machine Learning Algorithms

• Host: GitHub
• URL: https://github.com/bukosabino/btctrading
• Owner: bukosabino
• License: mit
• Created: 2017-04-13T15:52:04.000Z (over 7 years ago)
• Default Branch: master
• Last Pushed: 2018-05-27T09:18:12.000Z (over 6 years ago)
• Last Synced: 2024-10-16T05:27:42.202Z (25 days ago)
• Topics: bitcoin, deep-learning, deep-neural-networks, forecast, future, machine-learning, machine-learning-algorithms, prediction, price, time-series, trends, xgboost
• Language: Jupyter Notebook
• Homepage:
• Size: 11.5 MB
• Stars: 153
• Watchers: 18
• Forks: 47
• Open Issues: 3
• Metadata Files:
  • Readme: README.md
  • License: LICENSE

Awesome Lists containing this project

README

        
# btctrading

This is my personal code to predict the Bitcoin value using Machine Learning / Deep Learning Algorithms.

We will consider our problem as:

1. Regression Problem.

2. Classification Multiclass Problem [UP, KEEP, DOWN].

Get data from https://bitcoincharts.com/charts, you can choose period, symbol and exchange market and save the datas in a csv file.

# Deployment instructions

### Installation (python3):

```sh

> git clone https://github.com/bukosabino/btctrading.git

> cd btctrading

> virtualenv -p python3 virtualenvironment

> source virtualenvironment/bin/activate

> pip install -r requirements.txt

```

### Get data:

API: http://bitcoincharts.com/charts

period = ['1-min', '5-min', '15-min', '30-min', 'Hourly', '2-hour', '6-hour', '12-hour', 'Daily', 'Weekly']

market = ['krakenEUR', 'bitstampUSD'] -> list of markets: https://bitcoincharts.com/charts/volumepie/

```sh

> python get_data.py

```

### Run

```sh

> jupyter lab

```

# TODO:

* Simulator validation (backtesting)

* Add features -> Global Indicators (EUR/USD, S&P500, etc).

* Add different algorithms or ideas (LSTM, Reinforcement Learning, Q-Learning).

* Alert System (email, twitter, telegram).

----

Developed by Bukosabino at Lecrin Technologies - http://lecrintech.com

Please, let us know about any comment or feedback.