connorcoley / scscore

MIT License
91 stars 40 forks source link

data issue #11

Open YH-88 opened 3 years ago

YH-88 commented 3 years ago

Hello, You said that your model was trained with 12M reactions from Reaxys, but I can only see ten data from your data, and the rest are all USPTO data. In addition, the USPTO data is already mapped, could you tell me how to find unmapped reactions?Thanks a lot.

connorcoley commented 3 years ago

The Reaxys data is the intellectual property of Elsevier and can't be shared publicly, unfortunately. If you're part of a university or company with a Reaxys license, I'd encourage you to reach out to them and ask about access to their underlying data.

The USPTO is the primary public source of organic reaction data, originally extracted by Lowe

YH-88 commented 3 years ago

Thank you very much.