Is there any pytorch module which automatically gives me result of the number of parameters in a Convolutional Neural Network layer (similar keras topology.py)?

1 Like

I think it is easy to calculate the number of elements in PyTorch. Suppose you have a model called `net`

. You can use the following snippet to calculate the number of parameter in your model:

```
count = 0
for p in net.parameters():
count += p.data.nelement()
```

4 Likes

That snippet can be reduced to:

sum(p.data.nelement()forpinnet.parameters())

1 Like