Date: June 17th, 2021
Attendees: Bharath, Ashwin, Mufei, Peter, Alana
Summary: Bharath has been mainly working with GSoC students this week and doesn’t have much to report outside GSoC updates.
Ashwin has been working on updating his Molecule Net loader to use the Specified Splitter. He’s been working on processing the USPTO dataset cleanly and has uploaded the dataset to S3. He still needs to write tests and write docs.
Peter is working on converting the multitask regressor and classifier to PyTorch. He’s made some good progress.
Mufei has been busy but has come by to catch up.
Alana is working on the protein Tokenizer and the fasta loader. The shapes aren’t what we want yet so there’s still some more debugging to do.
Date: June 18th, 2021
Attendees: Bharath, Stanley, Atreya, Calum, Vignesh, Seyone, David
Summary: Calum is a PhD student at the University of Bath working on designing porous structures and is interested to learn more.
Bharath gave the same update as at the last meeting.
Stanley has gotten caught up on the GSoC updates and is playing around with an extended bioinformatics notebook.
Atreya has been working on his MATFeaturizer PR. He’s currently working on responding to review comments on the PR and needs to add another text.
Vignesh has been working on his Jax fit/predict methods for his JaxModel. Vignesh is currently working on the dependency management challenges.
Seyone has been working on getting up a WIP PR for the ChemBERTa masked language model. We need to do a bit of re-engineering to match the DeepChem API. This weekend, Seyone will try to start using DiskDataset.
David has been working on adding SwissProt to MoleculeNet. He’s working on getting the PR fixed and running.
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.