search

mlech26l / ncps

PyTorch and TensorFlow implementation of NCP, LTC, and CfC wired neural models
https://www.nature.com/articles/s42256-020-00237-3
Apache License 2.0
1.94k stars 311 forks source link

Input dimension #56

Open MarcProbst12 opened 11 months ago

MarcProbst12 commented 11 months ago

Hi, thanks for your work. I have a question regarding how to set up my input dimension for finance application.

My input currently is ( #examples, #features) (1.000.000, 177) . It's grouped according to the share and sorted according to the Date. Can you maybe be so kind and explain me how to set it up so that i can use it for an LTC?

Thanks a lot in advance!

mlech26l commented 11 months ago

Hi, the data format should be (#examples, #timesteps, #features).

For example, if we have 100 samples, each being a time-series with 20 timesteps and having 8 features, the training set should be of the shape (100, 20, 8).