Got error while compiling GPT-2 in tvm

Hi all ,
I am putting up the solution here for anyone who runs through the same problem ,
After a lot of debugging and code follow through , i stumbled upon this bert_compilation_problems and value error in bert compilation ,
i thought maybe its a version mismatch error.
So i downgraded pytorch and python and it worked. I was able to convert the torch model to relay model. Pytorch → 1.8.0 with cuda 11.0 support
Python → 3.7