How many trainable parameters does this layer have (bias included)?
nn.Conv2d(in_channels=3, out_channels=16, kernel_size=3)
- 448
- 432
- 144
- 464
16 filters x (3 x 3 x 3) weights = 432, plus 16 biases = 448.
Play AI & ML ▸
nn.Conv2d(in_channels=3, out_channels=16, kernel_size=3)