Question2:
I find out that if my model has input that needs to be None, summary function can't work out, can you fix it?
My model's forward function is just like this:
def forward(self, x1, x2, x3=None, x4=None): do something
or like this:
def forward(self, x1, x2=None, x3, x4=None): do something
https://github.com/sksq96/pytorch-summary/blob/345d898d84507b848e92dab4629e03405e19afce/torchsummary/torchsummary.py#L59 Question1: In this line code, you just set the sample batchsize to 2, maybe it's a mistake, could you modify it from 2 to batchsize? Just like this:
x = [torch.rand(batch_size, *in_size).type(dtype) for in_size in input_size]
Question2: I find out that if my model has input that needs to be None, summary function can't work out, can you fix it? My model's forward function is just like this:
def forward(self, x1, x2, x3=None, x4=None): do something
or like this:def forward(self, x1, x2=None, x3, x4=None): do something
Waiting for your reply! Thank you very much.