[Question Description] I have fine-tuned a classifier on my own dataset. At present, i need to test the classifier on my test dataset. However, it seems that I can only use each piece of data in the test dataset as a prompt input the fine-tuned classifier, which will takes a long time.
[My Question] Can I directly upload the test dataset file and get the final output results of classification? If possible, how can i do it?
I’m not sure if OpenAI API has a provision to run validation after the fine-tune has been completed, but you can write your own script and make calls to the model to compare against a validation dataset. Even better, you can use W&B to do that.