Stefan Doerr

```python import tangent import numpy as np def test(x, y): return np.dot(x, y) xxx = tangent.grad(test) xxx(np.random.rand(1, 3), np.random.rand(1, 3)) ```

Oh right, sorry. Well you can still make it work with numpy but fail with tangent even though it becomes a different error now ```python import tangent import numpy as...

It seems to me like you cannot overwrite a variable because it tries to assign to the value of the variable in the second iteration.

I found a fix, not sure if it applies to other cases as well https://github.com/conda/conda-build/issues/4498#issuecomment-1247819570

@raimis do you know something about this?

Related to this issue https://github.com/Acellera/htmd/issues/888

Hi @n-salvi did you manage to build the TIP4P system? If so we could close this issue.

I gave it a try with CHARMM and managed to build and run the system but it doesn't really work yet due to psfgen [not supporting LONEPAIR statements](https://www.ks.uiuc.edu/Research/vmd/mailing_list/vmd-l/30666.html) yet. Once...

there seems to be lots of active development currently on psfgen: https://www.ks.uiuc.edu/Research/vmd/vmd-new/devel.html I think with version 2.0 of psfgen in VMD 1.9.4 we should be able to do it.

hi @Lizeerd , can you give my fork a try? https://github.com/stefdoerr/DeeplyTough I updated their code to use moleculekit but I don't have time to test it. If it works fine...