sample_len and sample_time

[patricxu]

Get confused by the meaning of the variables.

sample_len: The time series batch size. self.sample_time = int(self.data_feature.shape[1] / self.sample_len)

In my view: data_feature.shape[1]: the length(steps) of the time series of a training sample. and it's fixed value for all the sample. so the sample_time should be 0 <= sample_time < data_features.shape[1]

Did I misunderstand?

[fjxmlzn]

Sorry, the variable names are kind of confusing.

data_feature.shape[1]: You are correct. sample_len: The time series batch size, i.e. the number of time steps generated at each RNN rollout (parameter S in Section 4.1 in the paper). sample_time: The maximum number of RNN rollouts needed for generating a sample.

[patricxu]

Thank you!