Hi ! I’ve trained a rasa model which contain lots of data . My question is that can we finetune previously trained model or do something similar to transfer learning whenever we add a new intent ? or we need to train it from scratch everytime we add something new ? if yes , how can we finetune it ?
Thanks in Advance.
we’re working on supporting continues training. But right now it is not possible. You have to retrain the model
@Ghostvv is it possible now? 2022 like I have dataset to train . . . is it possible that I fine-tune it simply.
@karen-white when I am finetuning using your article .I am getting this error ,and my model is not showing the new intent that I have added
InvalidConfigException: Cannot finetune because NLU training data contains new labels or does not contain any examples for some known labels. Please make sure that the NLU data that you use for finetuning contains at least one example for every label (i.e. intent, action name, ...) that was included in the NLU data used for training the model which we attempt to finetune now. Moreover, you must not add labels that were not included during training before