Zhiyuan Peng
Zhiyuan Peng
> @pzy2000 If the lambda function or dropout layer is causing the NaNs, can we consider removing them and potentially adding alternative functionalities elsewhere in the network? The pre-processing data...
I just tried on latest tf, and it produced the following issue, of which I suppose was caused by inconsistent tf version: Traceback (most recent call last): File "/root/zmx/COMET-master/issue/issue_tf_conv2d.py", line...
the onnx file can be downloaded here: [lstm0-sinewave_origin.zip](https://github.com/ENOT-AutoDL/onnx2torch/files/14616115/lstm0-sinewave_origin.zip)
when selling goods, it is the same issue:) 
it is interesting that when I create a sell order, and check the mail when it is sold. It behaves normally.