> Well, the obvious improvement to this would be to use numeric identifiers rather than implicit numeric identifiers where you derive the number from the positi

> Well, the obvious improvement to this would be to use numeric identifiers rather than implicit numeric identifiers where you derive the number from the position of the only set bit in a very long bitstring. I think you need to spend a bit of time learning how deep learning algorithms work. One hot encoding is used because the data is represented as matrices and the algorithms often use one-hot encoded as shorthand for the position in columns/rows for those matrices.