Chunyang
Chunyang
i need it too!tks! i write code to convert onnx but it is not work! `import torch import onnx en2de = torch.hub.load('pytorch/fairseq', 'transformer.wmt19.en-de.single_model') output = en2de.translate('Hello world', beam=5) #print("This is...
> @rGitcy 请问您解决这个问题了吗，我遇到了很您相似的问题。Have you solved this issue? 没有，感觉官方repo有些问题
你好博主，我在跑demo程序的时候，使用的iter_65000.pth模型，结果出来copying a param with shape torch.Size([38, 512]) from checkpoint, the shape in current model is torch.Size([39, 512]).这个报错，demo默认的模型因该是iter_30000这个，是我导入模型的问题，还是需要调节哪边参数，谢谢博主 
> 此为bug，可以的，这样修改是合理的。 非常感谢，我先按这个修改即可对吧？请问官网上会更新修复后的代码吗？
> 可以使用 wget https://bj.bcebos.com/dataset/glue/SST.zip 直接下载吗？ 应该是可以的，请问下载后放到哪个目录？另一方面，经过测试，这个bug和网速有关，可能是官网限速！
i delete the astype in the line 428 of model_utils.py .Now it run successs. 
> glad to see that you have found the trick to fix this issue. > > But, this is actually a bug when doing dynamic-to-static, we will try to fix...
I got the error "RuntimeError: Dynamic shape axis should be no more than the shape dimension for input1_dynamic_axes_4" too.
I also get the same error "KeyError: 'params'".I just want to convert RealESRGAN_x4plus.pth to onnx.
@wj-Mcat hi， do you have fix the bug？thx