search

jind11 / TextFooler

A Model for Natural Language Attack on Text Classification and Inference
MIT License
485 stars 79 forks source link

About Data Size #26

Closed CuiMingyu closed 4 years ago

CuiMingyu commented 4 years ago

Hi, I have a question that if I have 10, 000 samples how can I change all the samples not part of samples.

I used BERT to produce adversaries samples for 10, 000 samples. But I found only part of them were changed. It seemed that the program randomly selected samples from dataset not selected all the samples.

How can I change 10, 000 samples?

Thank you for your time!

jind11 commented 4 years ago

The code now only changes those samples that have correct predictions and skip the wrong predictions, which is why only part of them were changed. If the BERT model has a high accuracy, then the majority of them will be changed.

CuiMingyu commented 4 years ago

I see. I notice the code.

Thank you for your reminder!