299 lines
10 KiB
Python
299 lines
10 KiB
Python
import torch
|
|
from lightning import LightningModule
|
|
from torch_geometric.data import Batch
|
|
import importlib
|
|
from matplotlib import pyplot as plt
|
|
from matplotlib.tri import Triangulation
|
|
from .model.finite_difference import FiniteDifferenceStep
|
|
import os
|
|
|
|
def import_class(class_path: str):
|
|
module_path, class_name = class_path.rsplit(".", 1) # split last dot
|
|
module = importlib.import_module(module_path) # import the module
|
|
cls = getattr(module, class_name) # get the class
|
|
return cls
|
|
|
|
|
|
def _plot_mesh(pos, y, y_pred, y_true ,batch, i, batch_idx):
|
|
|
|
idx = batch == 0
|
|
y = y[idx].detach().cpu()
|
|
y_pred = y_pred[idx].detach().cpu()
|
|
pos = pos[idx].detach().cpu()
|
|
y_true = y_true[idx].detach().cpu()
|
|
# print(torch.max(y_true), torch.min(y_true))
|
|
folder = f"{batch_idx:02d}_images"
|
|
if os.path.exists(folder) is False:
|
|
os.makedirs(folder)
|
|
pos = pos.detach().cpu()
|
|
tria = Triangulation(pos[:, 0], pos[:, 1])
|
|
plt.figure(figsize=(18, 5))
|
|
plt.subplot(1, 3, 1)
|
|
plt.tricontourf(tria, y.squeeze().numpy(), levels=14)
|
|
plt.colorbar()
|
|
plt.title("Step t-1")
|
|
plt.subplot(1, 3, 2)
|
|
plt.tricontourf(tria, y_pred.squeeze().numpy(), levels=14)
|
|
plt.colorbar()
|
|
plt.title("Step t Predicted")
|
|
plt.subplot(1, 3, 3)
|
|
plt.tricontourf(tria, y_true.squeeze().numpy(), levels=14)
|
|
plt.colorbar()
|
|
plt.title("t True")
|
|
plt.suptitle("GNO", fontsize=16)
|
|
name = f"{folder}/graph_iter_{i:04d}.png"
|
|
plt.savefig(name, dpi=72)
|
|
plt.close()
|
|
|
|
def _plot_losses(losses, batch_idx):
|
|
folder = f"{batch_idx:02d}_images"
|
|
plt.figure()
|
|
plt.plot(losses)
|
|
plt.yscale("log")
|
|
plt.xlabel("Iteration")
|
|
plt.ylabel("Loss")
|
|
plt.title("Test Loss over Iterations")
|
|
plt.grid(True)
|
|
file_name = f"{folder}/test_loss.png"
|
|
plt.savefig(file_name, dpi=300)
|
|
plt.close()
|
|
|
|
|
|
class GraphSolver(LightningModule):
|
|
def __init__(
|
|
self,
|
|
model_class_path: str,
|
|
model_init_args: dict = {},
|
|
loss: torch.nn.Module = None,
|
|
start_unrolling_steps: int = 1,
|
|
increase_every: int = 20,
|
|
increase_rate: float = 2,
|
|
max_unrolling_steps: int = 100,
|
|
max_inference_iters: int = 1000,
|
|
inner_steps: int = 16,
|
|
):
|
|
super().__init__()
|
|
self.model = import_class(model_class_path)(**model_init_args)
|
|
# for param in self.model.parameters():
|
|
# print(f"Param: {param.shape}, Grad: {param.grad}")
|
|
# print(f"Param: {param[0]}")
|
|
self.fd_net = FiniteDifferenceStep()
|
|
self.loss = loss if loss is not None else torch.nn.MSELoss()
|
|
self.start_unrolling = start_unrolling_steps
|
|
self.current_unrolling_steps = self.start_unrolling
|
|
self.increase_every = increase_every
|
|
self.increase_rate = increase_rate
|
|
self.max_unrolling_steps = max_unrolling_steps
|
|
self.max_inference_iters = max_inference_iters
|
|
self.threshold = 1e-4
|
|
self.inner_steps = inner_steps
|
|
|
|
def _compute_deg(self, edge_index, edge_attr, num_nodes):
|
|
deg = torch.zeros(num_nodes, device=edge_index.device)
|
|
deg = deg.scatter_add(0, edge_index[1], edge_attr)
|
|
return deg + 1e-7
|
|
|
|
def _compute_loss(self, x, y):
|
|
return self.loss(x, y)
|
|
|
|
def _log_loss(self, loss, batch, stage: str):
|
|
self.log(
|
|
f"{stage}/loss",
|
|
loss,
|
|
on_step=True,
|
|
on_epoch=True,
|
|
prog_bar=True,
|
|
batch_size=int(batch.num_graphs),
|
|
)
|
|
return loss
|
|
|
|
@staticmethod
|
|
def _compute_c_ij(c, edge_index):
|
|
"""
|
|
TODO: add docstring.
|
|
"""
|
|
return (0.5 * (c[edge_index[0]] + c[edge_index[1]])).squeeze()
|
|
|
|
def _compute_model_steps(
|
|
self, x, edge_index, edge_attr, boundary_mask, boundary_values
|
|
):
|
|
|
|
out = x + self.model(x, edge_index, edge_attr)
|
|
# out[boundary_mask] = boundary_values.unsqueeze(-1)
|
|
plt.figure()
|
|
return out
|
|
|
|
def _check_convergence(self, out, x):
|
|
residual_norm = torch.norm(out - x)
|
|
if residual_norm < self.threshold * torch.norm(x):
|
|
return True
|
|
return False
|
|
|
|
def _preprocess_batch(self, batch: Batch):
|
|
x, y, c, edge_index, edge_attr = (
|
|
batch.x,
|
|
batch.y,
|
|
batch.c,
|
|
batch.edge_index,
|
|
batch.edge_attr,
|
|
)
|
|
# edge_attr = 1 / edge_attr
|
|
c_ij = self._compute_c_ij(c, edge_index)
|
|
edge_attr = edge_attr * (c_ij) # / 100)
|
|
return x, y, edge_index, edge_attr
|
|
|
|
def training_step(self, batch: Batch):
|
|
x, y, edge_index, edge_attr = self._preprocess_batch(batch)
|
|
# deg = self._compute_deg(edge_index, edge_attr, x.size(0))
|
|
losses = []
|
|
# print(x.shape, y.shape)
|
|
# # print(torch.max(edge_index), torch.min(edge_index))
|
|
# plt.figure()
|
|
# plt.subplot(2,3,1)
|
|
# plt.scatter(batch.pos[:,0].cpu(), batch.pos[:,1].cpu(), c=x.squeeze().cpu())
|
|
# plt.subplot(2,3,2)
|
|
# plt.scatter(batch.pos[:,0].cpu(), batch.pos[:,1].cpu(), c=y[:,0,:].squeeze().cpu())
|
|
# plt.subplot(2,3,3)
|
|
# plt.scatter(batch.pos[:,0].cpu(), batch.pos[:,1].cpu(), c=y[:,1,:].squeeze().cpu())
|
|
# plt.subplot(2,3,4)
|
|
# plt.scatter(batch.pos[:,0].cpu(), batch.pos[:,1].cpu(), c=y[:,2,:].squeeze().cpu())
|
|
# plt.subplot(2,3,5)
|
|
# plt.scatter(batch.pos[:,0].cpu(), batch.pos[:,1].cpu(), c=y[:,3,:].squeeze().cpu())
|
|
# plt.subplot(2,3,6)
|
|
# plt.scatter(batch.pos[:,0].cpu(), batch.pos[:,1].cpu(), c=y[:,4,:].squeeze().cpu())
|
|
# plt.suptitle("Training Batch Visualization", fontsize=16)
|
|
# plt.savefig("training_batch_visualization.png", dpi=300)
|
|
# plt.close()
|
|
# y = z
|
|
pos = batch.pos
|
|
boundary_mask = batch.boundary_mask
|
|
boundary_values = batch.boundary_values
|
|
# plt.scatter(pos[boundary_mask,0].cpu(), pos[boundary_mask,1].cpu(), c=boundary_values.cpu(), s=1)
|
|
# plt.savefig("boundary_nodes.png", dpi=300)
|
|
# y = z
|
|
print(y.shape)
|
|
for i in range(self.current_unrolling_steps * self.inner_steps):
|
|
out = self._compute_model_steps(
|
|
# torch.cat([x,pos], dim=-1),
|
|
x,
|
|
edge_index,
|
|
edge_attr,
|
|
# deg,
|
|
batch.boundary_mask,
|
|
batch.boundary_values,
|
|
)
|
|
x = out
|
|
# print(out.shape, y[:, i, :].shape)
|
|
losses.append(self.loss(out.flatten(), y[:, i, :].flatten()))
|
|
|
|
print(losses)
|
|
|
|
loss = torch.stack(losses).mean()
|
|
# for param in self.model.parameters():
|
|
# print(f"Param: {param.shape}, Grad: {param.grad}")
|
|
# print(f"Param: {param[0]}")
|
|
self._log_loss(loss, batch, "train")
|
|
return loss
|
|
|
|
# def on_train_epoch_start(self):
|
|
# print(f"Current unrolling steps: {self.current_unrolling_steps}, dataset unrolling steps: {self.trainer.datamodule.train_dataset.unrolling_steps}")
|
|
# return super().on_train_epoch_start()
|
|
|
|
def on_train_epoch_end(self):
|
|
if (
|
|
(self.current_epoch + 1) % self.increase_every == 0
|
|
and self.current_epoch > 0
|
|
):
|
|
dm = self.trainer.datamodule
|
|
self.current_unrolling_steps = min(
|
|
int(self.current_unrolling_steps * self.increase_rate),
|
|
self.max_unrolling_steps
|
|
)
|
|
dm.unrolling_steps = self.current_unrolling_steps
|
|
return super().on_train_epoch_end()
|
|
|
|
def validation_step(self, batch: Batch, _):
|
|
# x, y, edge_index, edge_attr = self._preprocess_batch(batch)
|
|
|
|
# deg = self._compute_deg(edge_index, edge_attr, x.size(0))
|
|
# for i in range(self.max_inference_iters * self.inner_steps):
|
|
# out = self._compute_model_steps(
|
|
# x,
|
|
# edge_index,
|
|
# edge_attr,
|
|
# deg,
|
|
# batch.boundary_mask,
|
|
# batch.boundary_values,
|
|
# )
|
|
# converged = self._check_convergence(out, x)
|
|
# x = out
|
|
# if converged:
|
|
# break
|
|
# print(y.shape, out.shape)
|
|
# loss = self.loss(out, y[:,-1,:])
|
|
# self._log_loss(loss, batch, "val")
|
|
# self.log("val/iterations", i + 1, on_step=False, on_epoch=True, prog_bar=True, batch_size=int(batch.num_graphs),)
|
|
# return loss
|
|
|
|
x, y, edge_index, edge_attr = self._preprocess_batch(batch)
|
|
# deg = self._compute_deg(edge_index, edge_attr, x.size(0))
|
|
losses = []
|
|
pos = batch.pos
|
|
for i in range(self.current_unrolling_steps * self.inner_steps):
|
|
out = self._compute_model_steps(
|
|
# torch.cat([x,pos], dim=-1),
|
|
x,
|
|
edge_index,
|
|
edge_attr,
|
|
# deg,
|
|
batch.boundary_mask,
|
|
batch.boundary_values,
|
|
)
|
|
_plot_mesh(batch.pos, x, out, y[:, i, :], batch.batch, i, self.current_epoch)
|
|
x = out
|
|
losses.append(self.loss(out, y[:, i, :]))
|
|
|
|
loss = torch.stack(losses).mean()
|
|
self._log_loss(loss, batch, "val")
|
|
return loss
|
|
|
|
def test_step(self, batch: Batch, batch_idx):
|
|
x, y, edge_index, edge_attr = self._preprocess_batch(batch)
|
|
|
|
deg = self._compute_deg(edge_index, edge_attr, x.size(0))
|
|
losses = []
|
|
for i in range(self.max_iters):
|
|
out = self._compute_model_steps(
|
|
x,
|
|
edge_index,
|
|
edge_attr.unsqueeze(-1),
|
|
deg,
|
|
batch.boundary_mask,
|
|
batch.boundary_values,
|
|
)
|
|
converged = self._check_convergence(out, x)
|
|
# _plot_mesh(batch.pos, y, out, batch.batch, i, batch_idx)
|
|
losses.append(self.loss(out, y).item())
|
|
if converged:
|
|
break
|
|
x = out
|
|
loss = self.loss(out, y)
|
|
# _plot_losses(losses, batch_idx)
|
|
self._log_loss(loss, batch, "test")
|
|
self.log(
|
|
"test/iterations",
|
|
i + 1,
|
|
on_step=False,
|
|
on_epoch=True,
|
|
prog_bar=True,
|
|
batch_size=int(batch.num_graphs),
|
|
)
|
|
|
|
def configure_optimizers(self):
|
|
optimizer = torch.optim.AdamW(self.parameters(), lr=1e-2)
|
|
return optimizer
|
|
|
|
def _impose_bc(self, x: torch.Tensor, data: Batch):
|
|
x[data.boundary_mask] = data.boundary_values
|
|
return x
|