search

ebanalyse / NERDA

Framework for fine-tuning pretrained transformers for Named-Entity Recognition (NER) tasks
MIT License
153 stars 35 forks source link

Question about validation_set #11

Closed Shengyu-Liu558 closed 3 years ago

Shengyu-Liu558 commented 3 years ago

I want to ask a simple question. The parameters of the model have been set before model training. What is the purpose of the validation set in model training? Thank you！

smaakage85 commented 3 years ago

The purpose of the validation set is to choose the best model/weights based on the validation performance i.e. performance on the validation data set, that has not been used for training as such. The model is evaluated at the end of each epoch.

Maybe take a look at: https://machinelearningmastery.com/difference-test-validation-datasets/

Did this clearify things?

Best, Lars

Shengyu-Liu558 commented 3 years ago

Thank you，good answer！！