-
Notifications
You must be signed in to change notification settings - Fork 8
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
Add custom forcefields #35
Conversation
Codecov Report
Additional details and impacted files@@ Coverage Diff @@
## main #35 +/- ##
==========================================
- Coverage 92.24% 89.37% -2.87%
==========================================
Files 18 19 +1
Lines 1328 1384 +56
==========================================
+ Hits 1225 1237 +12
- Misses 103 147 +44
|
One thing we might want to consider is whether or not we want to include pytorch in the default conda environments. Is pytorch pretty big and slow to install? For some dependencies that will rarely be used, we could add import checks and if they aren't installed throw errors with a message of the correct conda install command. What do you think? |
Yeah, I agree. Installing Pytorch with CPU usually doesn't slow installation but the GPU installation may cause some complications with Cuda. Since Pytorch is not being used anywhere else in this package, I agree that we should take it out of the env file. |
This PR adds a simple example of ML based force predictors to the forcefields library. (issue #15 )
I also separated the python file for xml based forcefields from custom forcefields.
The
custom_forcefirlds.py
would be a good place for now to add any kind of force object we can think of. We can think of a better name for it if we want.I need to add a unit test for the ML based torch custom force. That requires some offline model training.