[ Should we test ANI performance on a GPU?? ]
While evaluation of DNN methods would be significantly faster on GPUs, and may not be optimized for CPU evaluation, we note that many quantum chemistry methods are also accelerated on GPUs. Thus we retain the single-core CPU timings in Table \ref{310223} and Figure \ref{444292} but note that actual speed of ML methods such as ANI would be faster.
[ Despite training on DFT and coupled-cluster energies, ML methods still not as accurate for conformer energies.. why? Maybe some random errors, maybe don't include multiple minima in parameterization, so there's a bias in training? ]