search

SVAIGBA / WMSeg

MIT License
172 stars 42 forks source link

the problem of evaluating model #7

Open huzheng1111 opened 3 years ago

huzheng1111 commented 3 years ago

Excuse me,For the SIGHAN 2005 dataset, do you use the test data to evaluate the model, and then use the test data to test the final performance of the model?

yuanheTian commented 3 years ago

Thank you for your question.

We use a part of the training set as the dev set to find the best hyperparameter settings (e.g., batch size, learning rate, etc.) and then use that hyperparameter setting to train our model on the entire training set and finally evaluate it on the test set.