Shapes #4
Description
I guess it may be a silly question as I'm quite new in both ReasonML and TensorFlow, but I'm gonna ask it anyway. Why i.e. Rank3 has shape with length of four ((int, int, int, int)) instead of three? It's somewhat confusing to me as tensor3d function of tfjs accepts array with three elements. I can clearly see that using IntTensor3D.Create.tensor(...) gives me different result than tf.tensor3d(...). Also, shouldn't Rank0 have a shape of length 0 instead of single int?