Closed lpy1207 closed 2 months ago
Hello @lpy1207,
thanks for using TabLLM!
Basically, the zero-shot results are predictions without training data, so yes this would be possible. Please follow the readme and use the settings for zero-shot in the run script. You can also find some more details about zero-shot experiments here: Large Language Model Capabilities in Perioperative Risk Prediction and Prognostication
I hope that help!
Hello, thank you very much for your work. How can the trained model be used to make predictions on unlabeled data?