AssertionError: idx 0 != idx_new 20 during testing.

[pmgautam]

I'm getting the error AssertionError: idx 0 != idx_new 20 during testing.. Can someone help me with this. I've tried looking in issues but couldn't get much help. Thanks in advance!

[fuzheng1209]

I have the same error,do you solve the problem?

[neverstoplearn]

I meet tehe save error,have anyone solve it?

[shivpoojansaini]

I'm getting the error AssertionError: idx 0 != idx_new 20 during testing.. Can someone help me with this. I've tried looking in issues but couldn't get much help. Thanks in advance!

have you got this while training the model? Or only inferencing on your own data.

[shivpoojansaini]

I meet tehe save error,have anyone solve it?

One possible solution https://github.com/FangShancheng/ABINet/issues/34#issuecomment-1465623885

[neverstoplearn]

I meet tehe save error,have anyone solve it?

One possible solution #34 (comment)

will this affect the performance of the model？

[shivpoojansaini]

I meet tehe save error,have anyone solve it?

One possible solution #34 (comment)

will this affect the performance of the model？

No, we just skipped some data there is no modification from the model perspective.

[shivpoojansaini]

Fo more details understand the exception cause of def get(self, idx): under dataset.py

[fuzheng1209]

I use Baidu cloud datasets provided by the author.But I also meet the same problem.

[shivpoojansaini]

Modify the function under dataset.py if not self.is_training: if idx!=idx_new: pass

under function def __getitem__(self, idx): Since def get(self, idx): going in exception for some id of data