Cao Xuejie

@henryhungle Any update on the finetuned model checkpoints?

> I had the same error with TF2.4.0 and TF2.5.0. Downgrading to TF2.3.0 fixed the problem. I guess this repo needs an upgrade tot support the later TF releases. Downgrading...

Hi! I met the same problem. LeNet5's accuracy drops by almost 90% but VGG16‘s accuracy doesn't change. I've verified that it's not the data or the model itself that‘s wrong....

I met the same problem, have the problem solved?

generate_gpt_codes.py #176

 Sorry, I couldn't make sense of it. Long inputs lead to longer outputs. But if “max_length” is set larger such as 2048, the same error will occur.

I used the data provided by APPS and did not change it. As far as I know, gpt2 has a max input size of 1024 and gpt-neo has a max...