-
Hello, I understand that "atom type information is represented as a density distribution around the atom center" and I believe the function you are using is the one presented in this paper: https://do…
-
Hi,
I am currently using the molgrid.GridMaker in our project to convert data to its grid representation. It works well in the single-GPU setting. But when I tried to use Distributed Data Parallel …
-
-
`molgrid` (installed from PyPI) does not have a `__version__` attribute, which might be useful in some cases.
RMeli updated
3 years ago
-
Hi,
From what I understood, when I have a types file with protein-ligand pairs (3rd and 4th columns) and I use this types file to populate an ExampleProvider, each example will have 28 channels. Th…
-
- Have a pure Python implementation of Becke weights (just for the sake of having it)
- A function which takes all atomic numbers, all atomic coordinates, atomic grid points of one atom and it's an i…
-
Hi, thanks again.
It's really handy that you built-in CSS styling of the labels and tooltips, like `color: red`. It would also be great to support Python string formatting (e.g., to trim the strin…
-
Hi prof. Koes,
I am training a CNN model for predicting the binding pose. Also, I plan to visualize the generated grid density of the input protein-ligand 3d data.
If libmolgrid provides some f…
-
Hi,
This seems like a really neat tool. I have been experimenting along the lines of
```
maker = molgrid.GridMaker(resolution=res, dimension=dimension)
provider = molgrid.ExampleProvider()
p…
-
The following example will give this error: `TypeError: Smallest rgrid.points is negative, got -4.4999`
This error is not very helpful, and probably the `rgrid` should have raised an error not `MolGr…