DeepChem Minutes
India/Asia/Pacific Call
Date: May 6th, 2021
Attendees: Vignesh, Saurav, Mufei, Ashwin, Gao, Kexin, Tianfan
Summary: Saurav is currently working at weights and biases and a senior in high school who’s looking to get involved with DeepChem. Kexin just finished his Master’s at Harvard data science. Kexin is the lead on the TDC project. Wenhao is a first year PhD student at ChemE working on TDC. Tianfan is a 3rd year PhD student at Georgia Tech also working on TDC.
Bharath has mostly worked on DeepChem issues.
Vignesh is working on adding Joblib support for DeepChem. Once he’s got a basic PR ready he will put it up for review. He also put up a PaGTN PR up.
Mufei reviewed the PaGTN PRs.
Ashwin discussed with Seyone about HuggingFace and found a general strategy for HuggingFace integration with DeepChem and was also looking at Phillipe Schwaller’s work on reaction modeling. Ashwin is also working on cleaning up the datasets.
Americas/Europe/Africa/Middle East
India/Asia/Pacific Call
Date: May 7th, 2021
Attendees: Peter, Stanley, Seyone, Kevin, Ponni, Morgan, Jaewon
Summary: Kevin is an engineer at Weights and Biases and is working on improving integration with DeepChem. Morgan is also an engineer at Weights and Biases and looking to improve some of the integration with DeepChem.
Ponni was a scientist at UW, structural biologist by training and is now interested in using machine learning in drug discovery.
Jae is a bioinformatician and is working at the Broad.
Bharath gave the same update.
Peter had a busy week this week.
Stanley has been doing research on ProteinBERT and what that project might look and some initial sketch of an extended bioinformatics tutorial series.
Ponni asked what os’es people were using and people were on Mac, Linux, Windows.
Seyone this week worked on putting up a rough sketch for implementing ChemBERTa into DeepChem (https://www.notion.so/Expanding-ChemBERTa-and-NLP-techniques-within-DeepChem-3d8eed50c8b64f34883fed85245f29e3) which he plans to post on the forums soon.
Next we moved to the general discussion.
Kevin and Morgan had a short presentation (https://docs.google.com/presentation/d/1HUa8XkQHqhYM0fyjRQz3HFMPHjVJ67e2gMRqyDTnmzw/edit?usp=sharing) for improving weights & biases integration.
Jae asked where he could find the onboarding doc. Bharath pointed to https://docs.google.com/document/d/19RgVv9KKdZl3Dnz4J4nSDjgvWFU8YskIkUI8ZLBCTjo/edit?usp=sharing
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.