> Also, I can't justify why, but my gut feeling is that the database should be an order of magnitude faster than Python, provided the implementation were to be improved.
Would be curious how that could end up being the case. Perhaps if NumPy wasn't used at all? That would mean no vectorization or optimized storage.
Would be interesting to see how it scaled with length and dimensionality
> Also, I can't justify why, but my gut feeling is that the database should be an order of magnitude faster than Python, provided the implementation were to be improved.
Would be curious how that could end up being the case. Perhaps if NumPy wasn't used at all? That would mean no vectorization or optimized storage.
Would be interesting to see how it scaled with length and dimensionality