Data Science Stack Exchange is a question and answer site for Data science professionals, Machine Learning specialists, and those interested in learning more about the field. It only takes a minute to sign up.
Sign up to join this community
I am working on a time-series kind of problem. I have (timesteps, features) and the length of output will be (timesteps,). All outputs as float numbers. But the problem is I have training samples with different timesteps.
Suppose one example has 1500 timesteps then it will output 1500 dimensional vector. Another example has 1000 timesteps and it will have 1000 dimensional output.
I am looking for a way to train such a model using keras.
Its the number of features that has to remain consistent not the number of timestamps.
Outputs will be batches of one row predictions, so in your case its 1500 and than 1000. Model does not care, features should remain same. And beware of dataleakage with time series.
