Open hpicsk opened 2 years ago
In Colab notebook implementation of Class Siren foward def forward(self, coords): coords = coords.clone().detach().requires_grad_(True) # allows to take derivative w.r.t. input output = self.net(coords) return output, coords
def forward(self, coords): coords = coords.clone().detach().requires_grad_(True) # allows to take derivative w.r.t. input output = self.net(coords) return output, coords
why are coords set requiregrad(True)? Shouldn't the weights of the network only updated, not coordinates?
In Colab notebook implementation of Class Siren foward
def forward(self, coords): coords = coords.clone().detach().requires_grad_(True) # allows to take derivative w.r.t. input output = self.net(coords) return output, coordswhy are coords set requiregrad(True)? Shouldn't the weights of the network only updated, not coordinates?