different forecast result from different batch_size?

[veelion]

Hi all,

I use the model to forecast use same data but different batch_size, and get different size.

Firstly, batch_size=1, inference process: for i, input in enumerate(inputs): result, _ = tfm.forecast_with_covariate(inputs=[input], ...)

Second, batch_size = 10: for i in rang(0, len(inputs), batch_size): result, _ = tfm.forecast_with_covariate(inputs=inputs[i: i+batch_size], ...)