Hi, I'm learning machine learning sentiment analysis and your code is a great reference, but I'm running into some issues that I can't resolve

[peikolee]

From the code it appears that train_loss and val_loss contain only a single sentiment category label, not the full loss value. This may be due to problems with data processing or saving.
train_acc and val_acc also contain only a single sentiment category label instead of the full accuracy value. This may also be due to problems with data processing or saving. How can these problems be solved, and how do I retrain the model to improve it so that I can draw specific loss curves and accuracy curves during model training? Thank you very much for your answer.

Translated with www.DeepL.com/Translator (free version)