aiqm / torchani

Accurate Neural Network Potential on PyTorch
https://aiqm.github.io/torchani/
MIT License
459 stars 126 forks source link

improve documentation and fix buggy behavior of shuffle() and cache() #570

Closed IgnacioJPickering closed 3 years ago

IgnacioJPickering commented 3 years ago

Current behavior of shuffle() and cache() is not very performant since calling dataset.cache().shuffle(), dataset.cache().cache(), dataset.shuffle().shuffle() or other combinations creates copies instead of letting the arguments go through if the input is already cached, which is costly for no reason.

I also fixed a couple of typos in the documentation.