Date: July 29th, 2021
Attendees: Bharath, Peter, Ashwin
Bharath has been mostly working with GSoC Code. Bharath also wrote up a few forum posts:
- Onboarding as a DeepChem Open Source Engineer
- Writing DeepChemic Code
- Figuring out DeepChem's Scaling Story
Ashwin has been working on getting the latest changes into the USPTO loader. Ashwin is also looking to get the SmilesTokenizer to work with the loader.
Arun has been working on a tutorial for material featurization and has been adding more examples for the pytorch models. Arun is also thinking of organizing the tutorials. Peter suggested splitting them into core/application sets of tutorials.
Peter has been working to get the CI passing. Unfortunately there are a lot of breakages still out there.
Date: July 23rd, 2021
Attendees: Bharath, Vignesh, David, Seyone
Bharath gave the same update as he did the previous day.
Vignesh has been working on adding evaluation capabilities to JaxModel and has also been working on a physics inspired neural network model. Bharath mentioned that it might be worth looking at https://github.com/deepchem/deepchem/issues/2629.
David has been working to fix a small bug he found in the atomic convolutions model. David has been facing a memory issue with SwissProt. Bharath mentioned it might be good to look at https://deepchem.readthedocs.io/en/latest/api_reference/data.html#deepchem.data.DiskDataset.create_dataset for tips.
Seyone has been working to fix the
RobertaFeaturizer. The core mypy errors have been mostly solved but there’s still a couple issues to fix there.
Atreya has just put up his PRs for his layers and it looks like everything is passing right now. He has multiple PRs that will need to be merged inn.
Joining the DeepChem Developer Calls
As a quick reminder to anyone reading along, the DeepChem developer calls are open to the public! If you’re interested in attending either or both of the calls, please send an email to X.Y@gmail.com, where X=bharath, Y=ramsundar.