TaoSheng

I get another error when i use： def _compute_gradients(tensor, var_list): `def _compute_gradients(tensor, var_list):` ` grads = tf.gradients(tensor, var_list)` ` &nbsp return [grad if grad is not None else tf.zeros_like(var) for...

I finally find out where is key problem： K.gradient return None, because it counter an op that can't gradient I doubt that the Sequential have used to encapsulate vgg16 ,...

Reduce the version of PyQt to 5.12.1 and pip install PyQtWebEngine==5.12.1, arise other problem, pandasgui no response

it has 3 scene： 1. tensor mask op，as example：x = x[x.sum(dim=-1) > 0] 2. torch.where(condition), as example: torch.where(x>0) 3. torch.nonzero(x)

there have another method，use onnx-tool to get shape in numpy