folivo/PINA
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

Fixing self adaptive pinns (#469)

Browse Source 
* fix self adaptive pinn

* clean competitive pinn
This commit is contained in:
Giovanni Canali
2025-02-28 23:22:36 +01:00
committed by Nicola Demo
parent c3aaf5b1a0
commit 375f7f8e2d
2 changed files with 10 additions and 70 deletions
Download Patch File Download Diff File Expand all files Collapse all files

17
pina/solver/physic_informed_solver/competitive_pinn.py
View File

@@ -160,22 +160,6 @@ class CompetitivePINN(PINNInterface, MultiSolverInterface):
self._train_discriminator(samples, equation, discriminator_bets)
return loss_val
def loss_data(self, input_pts, output_pts):
"""
The data loss for the CompetitivePINN solver. It computes the loss
between the network output against the true solution.
:param LabelTensor input_tensor: The input to the neural networks.
:param LabelTensor output_tensor: The true solution to compare the
network solution.
:return: The computed data loss.
:rtype: torch.Tensor
"""
loss_val = super().loss_data(input_pts, output_pts)
# prepare for optimizer step called in training step
loss_val.backward()
return loss_val
def configure_optimizers(self):
"""
Optimizer configuration for the Competitive PINN solver.
@@ -252,7 +236,6 @@ class CompetitivePINN(PINNInterface, MultiSolverInterface):
)
# prepare for optimizer step called in training step
self.manual_backward(loss_val)
return
def _train_model(self, samples, equation, discriminator_bets):
"""

63
pina/solver/physic_informed_solver/self_adaptive_pinn.py
View File

@@ -178,63 +178,20 @@ class SelfAdaptivePINN(PINNInterface, MultiSolverInterface):
:return: The sum of the loss functions.
:rtype: LabelTensor
"""
self.optimizer_model.instance.zero_grad()
# Weights optimization
self.optimizer_weights.instance.zero_grad()
loss = super().training_step(batch)
self.optimizer_model.instance.step()
self.manual_backward(-loss)
self.optimizer_weights.instance.step()
# Model optimization
self.optimizer_model.instance.zero_grad()
loss = super().training_step(batch)
self.manual_backward(loss)
self.optimizer_model.instance.step()
return loss
def loss_phys(self, samples, equation):
"""
Computes the physics loss for the SAPINN solver based on given
samples and equation.
:param LabelTensor samples: The samples to evaluate the physics loss.
:param EquationInterface equation: The governing equation
representing the physics.
:return: The physics loss calculated based on given
samples and equation.
:rtype: torch.Tensor
"""
# Train the weights
weighted_loss = self._loss_phys(samples, equation)
loss_value = -weighted_loss.as_subclass(torch.Tensor)
self.manual_backward(loss_value)
# Detach samples from the existing computational graph and
# create a new one by setting requires_grad to True.
# In alternative set `retain_graph=True`.
samples = samples.detach()
samples.requires_grad_() # = True
# Train the model
weighted_loss = self._loss_phys(samples, equation)
loss_value = weighted_loss.as_subclass(torch.Tensor)
self.manual_backward(loss_value)
return loss_value
def loss_data(self, input_pts, output_pts):
"""
Computes the data loss for the SAPINN solver based on input and
output. It computes the loss between the
network output against the true solution.
:param LabelTensor input_pts: The input to the neural networks.
:param LabelTensor output_pts: The true solution to compare the
network solution.
:return: The computed data loss.
:rtype: torch.Tensor
"""
residual = self.forward(input_pts) - output_pts
loss = self._vectorial_loss(
torch.zeros_like(residual, requires_grad=True), residual
)
loss_value = self._vect_to_scalar(loss).as_subclass(torch.Tensor)
self.manual_backward(loss_value)
return loss_value
def configure_optimizers(self):
"""
Optimizer configuration for the SelfAdaptive PINN solver.
@@ -330,7 +287,7 @@ class SelfAdaptivePINN(PINNInterface, MultiSolverInterface):
)
return super().on_load_checkpoint(checkpoint)
def _loss_phys(self, samples, equation):
def loss_phys(self, samples, equation):
"""
Computation of the physical loss for SelfAdaptive PINN solver.