Stephenzza

 I found after quantization the matmul operator max value is too big, which 0~255 can not represent all value.

This operation is torch.bmm() in decoder.py line79 `inst_features = torch.bmm(iam_prob, features.view(B, C, -1).permute(0, 2, 1))` assume input tensor shape (B,3,512,512), inst_features =(B,400,4096)*(B,4096,256), after this matrix multiplication i got the exploded...

Same problem, and I can also ensure the key is correct.