ParlAI
ParlAI copied to clipboard
Pre-train r2c2 from scratch
I would like to pre-train r2c2-base-400M/3B from scratch with reddit+robert+cc100en datasets on ParlAI. Is there any way / script to do that ?
You will need to look elsewhere for the data, as we do not provide it within ParlAI. The SeeKeR paper outlines the training hyperparamters
Ok, thank you !