Skip to content

khai88108/Bitcoinprediction

Repository files navigation

Bitcoinprediction

An implementation of bitcoin price prediction using LSTM, RNN, and Neural Prophet.

• 5 years of Bitcoin price dataset collect from Yahoo Finance
• Long Short-Term Memory (LSTM) / Recurrent Neural Network (RNN)
• Neural Prophet

Data

• Collect from Yahoo Finance (https://finance.yahoo.com/)
• Range from the 15th of May 2018 until 14th of May 2023 (5 Years)
• 1827 rows of data
• Close price as a Target Variable • 80%(4 years) for training, 20% for testing (1 year)

Requirement

• Python 3.10
• Jupyter notebook 6.4.5

Libraries and modules

• Pandas
• Numpy
• matplotlib
• Scikit-learn
• TensorFlow
• NeuralProphet
• Sklearn.metrics

Result

LSTM Result


RNN Result


Neural Prophet Result