-
Notifications
You must be signed in to change notification settings - Fork 1.6k
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
DeepChem Model Wishlist #2680
Comments
If you have other models you'd like to see in DeepChem, please chime in below as well and I'll update the main wishlist at the top of the thread |
Maybe seen and implementation of MolGAN (https://github.com/nicola-decao/MolGAN https://arxiv.org/abs/1805.11973) would be nice 😃 |
@nicola-decao We already have MolGAN in DeepChem! https://deepchem.readthedocs.io/en/latest/api_reference/models.html#basicmolganmodel. I hadn't realized you were the lead author :) |
how about reinforcement learning, such as Reinvent. https://github.com/MolecularAI/Reinvent |
@rbharat Oh WOW I did not know that! 😱 I'll proudly share that with people then! |
MegaMolBART (released last month) would be interesting to compare against the other generative models: https://ngc.nvidia.com/catalog/models/nvidia:clara:megamolbart |
We have some funding now to support people who want to implement models. Please apply following directions at https://forum.deepchem.io/t/deepchem-model-wishlist-internships/569 if you are interested! |
Hi. I would love to take a crack at implementing SE(3) Transformer. Since it has a high quality implementation available already, it should give me a gentle intro to deepchem's repo! |
Flow Network based Generative Models for Non-Iterative Diverse Candidate Generation (https://arxiv.org/abs/2106.04399, https://github.com/bengioe/gflownet) paper proposes an approach for tasks like molecule synthesis. Maybe this can be added to the list. |
Neural Flows (https://proceedings.neurips.cc/paper/2021/file/b21f9f98829dea9a48fd8aaddc1f159d-Paper.pdf) proposes an |
I'm currently working on adding a tutorial for scikit-learn's Gaussian Processes. However, I've been having trouble with getting calibrated uncertainty out of it. I plan to try again with a different implementation of Gaussian Processes like GPy. Is someone still working on implementing ChemProp (D-MPNN)? |
@TheRealSalmon I believe that @atreyamaj was working with a student on ChemProp but I'm not sure if that project is still proceeding. Perhaps it might be worth teaming up and discussing? I'd really like to see D-MPNN make it into ChemProp so I can help on the effort as well :) |
Hi! I'm really sorry for the delay, got caught up with my thesis project for this semester. Now that the project is almost over, I will be resuming working on DMPNN starting now. |
@atreyamaj and @rbharath I would like to work on DPMNN project as well, can you please give me just a heads up how to proceed? |
@Zymrael I believe @shivance and @VIGNESHinZONE are both working on different ODE/PDE/NODE models and might be interested in discussing with you |
@Zymrael Can we please hop on a call sometime to discuss on the same or maybe you could talk during deepchem's office hours ? |
@rbharath this also seems interesting! |
@gusty1g Is this Model Wishlist updated? |
Mostly. I am working on grover. If you are interested to work on some, post that you are working on it here. |
Hey! Can I work on the Kalman Filter? Is it still available? |
Hey @rbharath and fellow DeepChem contributors! Can I work on HM Models,if someone else isnt' working on it? |
Hey @rbharath, can I work on implementing Neural ODEs under GSOC 2024. |
There are a number of important molecular and scientific models that I would love to see implementations for in DeepChem. Implementations of any of these models would be a great contribution to DeepChem. As an incentive for contributors, contributing a model would earn authorship on one of our upcoming manuscripts! If you have bandwidth to contribute an implementation, please chime in below
Equivariant Models
Classical Machine Learning
The text was updated successfully, but these errors were encountered: