What happens if the observation of time series is an array

chen0040 / keras-anomaly-detection

Anomaly detection implemented in Keras

MIT License

370 stars 157 forks source link

What happens if the observation of time series is an array #4

Open cipher009 opened 5 years ago

cipher009 commented 5 years ago

Demo example for credit card has train_data shape: (284807,29) If I am not mistaken, 29 is the timestep here (and not the number of the values in the observation)?

I have my data in shape (samples, timesteps, values) can I still use keras anomaly detection?