Bitcoin prediction with LSTM
https://colah.github.io/posts/2015-08-Understanding-LSTMs/

Bitcoin prediction with LSTM

Time series forecasting is a?technique for the prediction of events through a sequence of time. It predicts future events by analyzing the trends of the past, on the assumption that future trends will hold similar to historical trends.

Preprocessing method: I just used close value normalized.

Model architecture: a Four LSTM layers with dropout        

Model performance:

No alt text provided for this image

Conclusion: using LSTM we can use deep learning for time depending data, for example in this case where we want to predict future values based on past behavior of some data.

Source code: https://github.com/leocjj/holbertonschool-machine_learning/tree/main/supervised_learning/0x0E-time_series

要查看或添加评论,请登录

社区洞察

其他会员也浏览了