What order at the level of timesteps in a Keras RNN should we use?

An RNN, in keras, accepts an input with the following shape

array([[['Xa0', 'Ya0'],
           ['Xa1', 'Ya1'],
           ['Xa2', 'Ya2']],

         [['Xb0', 'Yb0'],
          ['Xb1', 'Yb1'],
          ['Xb2', 'Yb2']],

        [['Xc0', 'Yc0'],
         ['Xc1', 'Yc1'],
         ['Xc2', 'Yc2']]], dtype='<U3')

The numbers are associated with the time steps (3 in our case), and X,Y are features. We have 3 batches (a,b,c).

Which order should I use:

  1. 0 → 2 as in from past to present ?
  2. or 2 → 0 as in from present to past?
array([[['Xa2', 'Ya2'],
           ['Xa1', 'Ya1'],
           ['Xa0', 'Ya0']],

         [['Xb2', 'Yb2'],
          ['Xb1', 'Yb1'],
          ['Xb0', 'Yb0']],

        [['Xc2', 'Yc2'],
         ['Xc1', 'Yc1'],
         ['Xc0', 'Yc0']]], dtype='<U3')

Nothing in the documentation tells me which is the order I should use

In RNN the information goes through a loop. When it makes a decision, it considers the current input and also what it has learned from the inputs it has received previously because the sequence of data contains crucial information about what is coming next. So the order should always be 0 → 2 (i.e from past to present). Thank you