Raimondas Galvelis

We should put the from-source installation instructions to a separate file (e.g. `INSTALL.md`) and link it from the main `READM.md`. Most users can install with `conda`, so they don't need...

I'll make an example.

@jchodera I have created a draft of a tutorial (#62). You can see it better here: https://github.com/raimis/openmm-torch/blob/example/tutorials/openmm-torch-nnpops.ipynb Or just [](https://colab.research.google.com/github/raimis/openmm-torch/blob/example/tutorials/openmm-torch-nnpops.ipynb)

The tutorials is done (#62). @jchodera @peastman could you check it?

@jchodera could tweet about the new tutorial: https://github.com/openmm/openmm-torch/blob/master/tutorials/openmm-torch-nnpops.ipynb Also, this should be considered beta, but we need people trying it and find the remaining issues.

I keep this issue open as a reminder to update the tutorial, when following issue are fixed: - [x] An official NNPOps package is still not available: https://github.com/openmm/NNPOps/issues/26, https://github.com/openmm/openmm-torch/pull/77 -...

Currently the `ENERGY` CV isn't supported. OpenMM computes the forces (and energies) concurrently, so the potential energy is not available for the individual force objects. @peastman is it possible to...

A significant part of the computation overhead is coming form libTorch, which is more or less constant regardless the GPU.

We just made a tutorial how to speed up TorchANI with NNPOps: https://github.com/openmm/openmm-torch/blob/master/tutorials/openmm-torch-nnpops.ipynb

Could you have the list of your conda environment?