Tutorial update and small fixes * Tutorials update + Tutorial FNO * Create a metric tracker callback * Update PINN for logging * Update plotter for plotting * Small fix LabelTensor * Small fix FNO --------- Co-authored-by: Dario Coscia <dariocoscia@cli-10-110-13-250.WIFIeduroamSTUD.units.it> Co-authored-by: Dario Coscia <dariocoscia@dhcp-176.eduroam.sissa.it>
602 lines
258 KiB
Plaintext
Vendored
602 lines
258 KiB
Plaintext
Vendored
{
|
|
"cells": [
|
|
{
|
|
"cell_type": "markdown",
|
|
"metadata": {},
|
|
"source": [
|
|
"# Tutorial 2: resolution of Poisson problem and usage of extra-features"
|
|
]
|
|
},
|
|
{
|
|
"cell_type": "markdown",
|
|
"metadata": {},
|
|
"source": [
|
|
"### The problem definition"
|
|
]
|
|
},
|
|
{
|
|
"cell_type": "markdown",
|
|
"metadata": {},
|
|
"source": [
|
|
"This tutorial presents how to solve with Physics-Informed Neural Networks a 2D Poisson problem with Dirichlet boundary conditions. Using extrafeatures.\n",
|
|
"\n",
|
|
"The problem is written as:\n",
|
|
"\\begin{equation}\n",
|
|
"\\begin{cases}\n",
|
|
"\\Delta u = \\sin{(\\pi x)} \\sin{(\\pi y)} \\text{ in } D, \\\\\n",
|
|
"u = 0 \\text{ on } \\Gamma_1 \\cup \\Gamma_2 \\cup \\Gamma_3 \\cup \\Gamma_4,\n",
|
|
"\\end{cases}\n",
|
|
"\\end{equation}\n",
|
|
"where $D$ is a square domain $[0,1]^2$, and $\\Gamma_i$, with $i=1,...,4$, are the boundaries of the square."
|
|
]
|
|
},
|
|
{
|
|
"cell_type": "markdown",
|
|
"metadata": {},
|
|
"source": [
|
|
"First of all, some useful imports."
|
|
]
|
|
},
|
|
{
|
|
"cell_type": "code",
|
|
"execution_count": 1,
|
|
"metadata": {},
|
|
"outputs": [],
|
|
"source": [
|
|
"import torch\n",
|
|
"from torch.nn import Softplus\n",
|
|
"\n",
|
|
"from pina.problem import SpatialProblem\n",
|
|
"from pina.operators import laplacian\n",
|
|
"from pina.model import FeedForward\n",
|
|
"from pina.solvers import PINN\n",
|
|
"from pina.trainer import Trainer\n",
|
|
"from pina.plotter import Plotter\n",
|
|
"from pina.geometry import CartesianDomain\n",
|
|
"from pina.equation import Equation, FixedValue\n",
|
|
"from pina import Condition, LabelTensor\n",
|
|
"from pina.callbacks import MetricTracker"
|
|
]
|
|
},
|
|
{
|
|
"cell_type": "markdown",
|
|
"metadata": {},
|
|
"source": [
|
|
"Now, the Poisson problem is written in PINA code as a class. The equations are written as *conditions* that should be satisfied in the corresponding domains. *truth_solution*\n",
|
|
"is the exact solution which will be compared with the predicted one."
|
|
]
|
|
},
|
|
{
|
|
"cell_type": "code",
|
|
"execution_count": 2,
|
|
"metadata": {},
|
|
"outputs": [],
|
|
"source": [
|
|
"class Poisson(SpatialProblem):\n",
|
|
" output_variables = ['u']\n",
|
|
" spatial_domain = CartesianDomain({'x': [0, 1], 'y': [0, 1]})\n",
|
|
"\n",
|
|
" def laplace_equation(input_, output_):\n",
|
|
" force_term = (torch.sin(input_.extract(['x'])*torch.pi) *\n",
|
|
" torch.sin(input_.extract(['y'])*torch.pi))\n",
|
|
" laplacian_u = laplacian(output_, input_, components=['u'], d=['x', 'y'])\n",
|
|
" return laplacian_u - force_term\n",
|
|
"\n",
|
|
" conditions = {\n",
|
|
" 'gamma1': Condition(location=CartesianDomain({'x': [0, 1], 'y': 1}), equation=FixedValue(0.)),\n",
|
|
" 'gamma2': Condition(location=CartesianDomain({'x': [0, 1], 'y': 0}), equation=FixedValue(0.)),\n",
|
|
" 'gamma3': Condition(location=CartesianDomain({'x': 1, 'y': [0, 1]}), equation=FixedValue(0.)),\n",
|
|
" 'gamma4': Condition(location=CartesianDomain({'x': 0, 'y': [0, 1]}), equation=FixedValue(0.)),\n",
|
|
" 'D': Condition(location=CartesianDomain({'x': [0, 1], 'y': [0, 1]}), equation=Equation(laplace_equation)),\n",
|
|
" }\n",
|
|
"\n",
|
|
" def poisson_sol(self, pts):\n",
|
|
" return -(\n",
|
|
" torch.sin(pts.extract(['x'])*torch.pi)*\n",
|
|
" torch.sin(pts.extract(['y'])*torch.pi)\n",
|
|
" )/(2*torch.pi**2)\n",
|
|
" \n",
|
|
" truth_solution = poisson_sol\n",
|
|
"\n",
|
|
"problem = Poisson()\n",
|
|
"\n",
|
|
"# let's discretise the domain\n",
|
|
"problem.discretise_domain(25, 'grid', locations=['D'])\n",
|
|
"problem.discretise_domain(25, 'grid', locations=['gamma1', 'gamma2', 'gamma3', 'gamma4'])"
|
|
]
|
|
},
|
|
{
|
|
"cell_type": "markdown",
|
|
"metadata": {},
|
|
"source": [
|
|
"### The problem solution "
|
|
]
|
|
},
|
|
{
|
|
"cell_type": "markdown",
|
|
"metadata": {},
|
|
"source": [
|
|
"After the problem, the feed-forward neural network is defined, through the class `FeedForward`. This neural network takes as input the coordinates (in this case $x$ and $y$) and provides the unkwown field of the Poisson problem. The residual of the equations are evaluated at several sampling points (which the user can manipulate using the method `CartesianDomain_pts`) and the loss minimized by the neural network is the sum of the residuals.\n",
|
|
"\n",
|
|
"In this tutorial, the neural network is composed by two hidden layers of 10 neurons each, and it is trained for 1000 epochs with a learning rate of 0.006. These parameters can be modified as desired."
|
|
]
|
|
},
|
|
{
|
|
"cell_type": "code",
|
|
"execution_count": 3,
|
|
"metadata": {
|
|
"scrolled": true
|
|
},
|
|
"outputs": [
|
|
{
|
|
"name": "stderr",
|
|
"output_type": "stream",
|
|
"text": [
|
|
"GPU available: False, used: False\n",
|
|
"TPU available: False, using: 0 TPU cores\n",
|
|
"IPU available: False, using: 0 IPUs\n",
|
|
"HPU available: False, using: 0 HPUs\n",
|
|
"/Users/dariocoscia/anaconda3/envs/pina/lib/python3.9/site-packages/lightning/pytorch/trainer/connectors/logger_connector/logger_connector.py:67: UserWarning: Starting from v1.9.0, `tensorboardX` has been removed as a dependency of the `lightning.pytorch` package, due to potential conflicts with other packages in the ML ecosystem. For this reason, `logger=True` will use `CSVLogger` as the default logger, unless the `tensorboard` or `tensorboardX` packages are found. Please `pip install lightning[extra]` or one of them to enable TensorBoard support by default\n",
|
|
" warning_cache.warn(\n",
|
|
"\n",
|
|
" | Name | Type | Params\n",
|
|
"----------------------------------------\n",
|
|
"0 | _loss | MSELoss | 0 \n",
|
|
"1 | _neural_net | Network | 151 \n",
|
|
"----------------------------------------\n",
|
|
"151 Trainable params\n",
|
|
"0 Non-trainable params\n",
|
|
"151 Total params\n",
|
|
"0.001 Total estimated model params size (MB)\n"
|
|
]
|
|
},
|
|
{
|
|
"name": "stdout",
|
|
"output_type": "stream",
|
|
"text": [
|
|
"Epoch 999: : 1it [00:00, 129.50it/s, v_num=45, mean_loss=0.00196, gamma1_loss=0.0093, gamma2_loss=0.000146, gamma3_loss=8.16e-5, gamma4_loss=0.000201, D_loss=8.44e-5] "
|
|
]
|
|
},
|
|
{
|
|
"name": "stderr",
|
|
"output_type": "stream",
|
|
"text": [
|
|
"`Trainer.fit` stopped: `max_epochs=1000` reached.\n"
|
|
]
|
|
},
|
|
{
|
|
"name": "stdout",
|
|
"output_type": "stream",
|
|
"text": [
|
|
"Epoch 999: : 1it [00:00, 101.25it/s, v_num=45, mean_loss=0.00196, gamma1_loss=0.0093, gamma2_loss=0.000146, gamma3_loss=8.16e-5, gamma4_loss=0.000201, D_loss=8.44e-5]\n"
|
|
]
|
|
}
|
|
],
|
|
"source": [
|
|
"# make model + solver + trainer\n",
|
|
"model = FeedForward(\n",
|
|
" layers=[10, 10],\n",
|
|
" func=Softplus,\n",
|
|
" output_dimensions=len(problem.output_variables),\n",
|
|
" input_dimensions=len(problem.input_variables)\n",
|
|
")\n",
|
|
"pinn = PINN(problem, model, optimizer_kwargs={'lr':0.006, 'weight_decay':1e-8})\n",
|
|
"trainer = Trainer(pinn, max_epochs=1000, callbacks=[MetricTracker()])\n",
|
|
"\n",
|
|
"# train\n",
|
|
"trainer.train()"
|
|
]
|
|
},
|
|
{
|
|
"cell_type": "markdown",
|
|
"metadata": {},
|
|
"source": [
|
|
"Now the *Plotter* class is used to plot the results.\n",
|
|
"The solution predicted by the neural network is plotted on the left, the exact one is represented at the center and on the right the error between the exact and the predicted solutions is showed. "
|
|
]
|
|
},
|
|
{
|
|
"cell_type": "code",
|
|
"execution_count": 4,
|
|
"metadata": {},
|
|
"outputs": [
|
|
{
|
|
"data": {
|
|
"image/png": "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",
|
|
"text/plain": [
|
|
"<Figure size 1600x600 with 6 Axes>"
|
|
]
|
|
},
|
|
"metadata": {},
|
|
"output_type": "display_data"
|
|
}
|
|
],
|
|
"source": [
|
|
"plotter = Plotter()\n",
|
|
"plotter.plot(trainer)"
|
|
]
|
|
},
|
|
{
|
|
"cell_type": "markdown",
|
|
"metadata": {},
|
|
"source": [
|
|
"### The problem solution with extra-features"
|
|
]
|
|
},
|
|
{
|
|
"cell_type": "markdown",
|
|
"metadata": {},
|
|
"source": [
|
|
"Now, the same problem is solved in a different way.\n",
|
|
"A new neural network is now defined, with an additional input variable, named extra-feature, which coincides with the forcing term in the Laplace equation. \n",
|
|
"The set of input variables to the neural network is:\n",
|
|
"\n",
|
|
"\\begin{equation}\n",
|
|
"[x, y, k(x, y)], \\text{ with } k(x, y)=\\sin{(\\pi x)}\\sin{(\\pi y)},\n",
|
|
"\\end{equation}\n",
|
|
"\n",
|
|
"where $x$ and $y$ are the spatial coordinates and $k(x, y)$ is the added feature. \n",
|
|
"\n",
|
|
"This feature is initialized in the class `SinSin`, which needs to be inherited by the `torch.nn.Module` class and to have the `forward` method. After declaring such feature, we can just incorporate in the `FeedForward` class thanks to the `extra_features` argument.\n",
|
|
"**NB**: `extra_features` always needs a `list` as input, you you have one feature just encapsulated it in a class, as in the next cell.\n",
|
|
"\n",
|
|
"Finally, we perform the same training as before: the problem is `Poisson`, the network is composed by the same number of neurons and optimizer parameters are equal to previous test, the only change is the new extra feature."
|
|
]
|
|
},
|
|
{
|
|
"cell_type": "code",
|
|
"execution_count": 5,
|
|
"metadata": {},
|
|
"outputs": [
|
|
{
|
|
"name": "stderr",
|
|
"output_type": "stream",
|
|
"text": [
|
|
"GPU available: False, used: False\n",
|
|
"TPU available: False, using: 0 TPU cores\n",
|
|
"IPU available: False, using: 0 IPUs\n",
|
|
"HPU available: False, using: 0 HPUs\n",
|
|
"\n",
|
|
" | Name | Type | Params\n",
|
|
"----------------------------------------\n",
|
|
"0 | _loss | MSELoss | 0 \n",
|
|
"1 | _neural_net | Network | 161 \n",
|
|
"----------------------------------------\n",
|
|
"161 Trainable params\n",
|
|
"0 Non-trainable params\n",
|
|
"161 Total params\n",
|
|
"0.001 Total estimated model params size (MB)\n"
|
|
]
|
|
},
|
|
{
|
|
"name": "stdout",
|
|
"output_type": "stream",
|
|
"text": [
|
|
"Epoch 999: : 1it [00:00, 112.55it/s, v_num=46, mean_loss=2.73e-7, gamma1_loss=1.13e-6, gamma2_loss=7.1e-8, gamma3_loss=4.69e-8, gamma4_loss=6.81e-8, D_loss=4.65e-8] "
|
|
]
|
|
},
|
|
{
|
|
"name": "stderr",
|
|
"output_type": "stream",
|
|
"text": [
|
|
"`Trainer.fit` stopped: `max_epochs=1000` reached.\n"
|
|
]
|
|
},
|
|
{
|
|
"name": "stdout",
|
|
"output_type": "stream",
|
|
"text": [
|
|
"Epoch 999: : 1it [00:00, 92.69it/s, v_num=46, mean_loss=2.73e-7, gamma1_loss=1.13e-6, gamma2_loss=7.1e-8, gamma3_loss=4.69e-8, gamma4_loss=6.81e-8, D_loss=4.65e-8] \n"
|
|
]
|
|
}
|
|
],
|
|
"source": [
|
|
"class SinSin(torch.nn.Module):\n",
|
|
" \"\"\"Feature: sin(x)*sin(y)\"\"\"\n",
|
|
" def __init__(self):\n",
|
|
" super().__init__()\n",
|
|
"\n",
|
|
" def forward(self, x):\n",
|
|
" t = (torch.sin(x.extract(['x'])*torch.pi) *\n",
|
|
" torch.sin(x.extract(['y'])*torch.pi))\n",
|
|
" return LabelTensor(t, ['sin(x)sin(y)'])\n",
|
|
"\n",
|
|
"\n",
|
|
"# make model + solver + trainer\n",
|
|
"model_feat = FeedForward(\n",
|
|
" layers=[10, 10],\n",
|
|
" func=Softplus,\n",
|
|
" output_dimensions=len(problem.output_variables),\n",
|
|
" input_dimensions=len(problem.input_variables)+1\n",
|
|
")\n",
|
|
"pinn_feat = PINN(problem, model_feat, extra_features=[SinSin()], optimizer_kwargs={'lr':0.006, 'weight_decay':1e-8})\n",
|
|
"trainer_feat = Trainer(pinn_feat, max_epochs=1000, callbacks=[MetricTracker()])\n",
|
|
"\n",
|
|
"# train\n",
|
|
"trainer_feat.train()"
|
|
]
|
|
},
|
|
{
|
|
"cell_type": "markdown",
|
|
"metadata": {},
|
|
"source": [
|
|
"The predicted and exact solutions and the error between them are represented below.\n",
|
|
"We can easily note that now our network, having almost the same condition as before, is able to reach additional order of magnitudes in accuracy."
|
|
]
|
|
},
|
|
{
|
|
"cell_type": "code",
|
|
"execution_count": 6,
|
|
"metadata": {},
|
|
"outputs": [
|
|
{
|
|
"data": {
|
|
"image/png": "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",
|
|
"text/plain": [
|
|
"<Figure size 1600x600 with 6 Axes>"
|
|
]
|
|
},
|
|
"metadata": {},
|
|
"output_type": "display_data"
|
|
}
|
|
],
|
|
"source": [
|
|
"plotter.plot(trainer_feat)"
|
|
]
|
|
},
|
|
{
|
|
"cell_type": "markdown",
|
|
"metadata": {},
|
|
"source": [
|
|
"### The problem solution with learnable extra-features"
|
|
]
|
|
},
|
|
{
|
|
"cell_type": "markdown",
|
|
"metadata": {},
|
|
"source": [
|
|
"We can still do better!\n",
|
|
"\n",
|
|
"Another way to exploit the extra features is the addition of learnable parameter inside them.\n",
|
|
"In this way, the added parameters are learned during the training phase of the neural network. In this case, we use:\n",
|
|
"\n",
|
|
"\\begin{equation}\n",
|
|
"k(x, \\mathbf{y}) = \\beta \\sin{(\\alpha x)} \\sin{(\\alpha y)},\n",
|
|
"\\end{equation}\n",
|
|
"\n",
|
|
"where $\\alpha$ and $\\beta$ are the abovementioned parameters.\n",
|
|
"Their implementation is quite trivial: by using the class `torch.nn.Parameter` we cam define all the learnable parameters we need, and they are managed by `autograd` module!"
|
|
]
|
|
},
|
|
{
|
|
"cell_type": "code",
|
|
"execution_count": 7,
|
|
"metadata": {},
|
|
"outputs": [
|
|
{
|
|
"name": "stderr",
|
|
"output_type": "stream",
|
|
"text": [
|
|
"GPU available: False, used: False\n",
|
|
"TPU available: False, using: 0 TPU cores\n",
|
|
"IPU available: False, using: 0 IPUs\n",
|
|
"HPU available: False, using: 0 HPUs\n",
|
|
"\n",
|
|
" | Name | Type | Params\n",
|
|
"----------------------------------------\n",
|
|
"0 | _loss | MSELoss | 0 \n",
|
|
"1 | _neural_net | Network | 161 \n",
|
|
"----------------------------------------\n",
|
|
"161 Trainable params\n",
|
|
"0 Non-trainable params\n",
|
|
"161 Total params\n",
|
|
"0.001 Total estimated model params size (MB)\n"
|
|
]
|
|
},
|
|
{
|
|
"name": "stdout",
|
|
"output_type": "stream",
|
|
"text": [
|
|
"Epoch 999: : 1it [00:00, 91.07it/s, v_num=47, mean_loss=2.11e-6, gamma1_loss=1.03e-5, gamma2_loss=4.17e-8, gamma3_loss=4.28e-8, gamma4_loss=5.65e-8, D_loss=6.21e-8] "
|
|
]
|
|
},
|
|
{
|
|
"name": "stderr",
|
|
"output_type": "stream",
|
|
"text": [
|
|
"`Trainer.fit` stopped: `max_epochs=1000` reached.\n"
|
|
]
|
|
},
|
|
{
|
|
"name": "stdout",
|
|
"output_type": "stream",
|
|
"text": [
|
|
"Epoch 999: : 1it [00:00, 76.19it/s, v_num=47, mean_loss=2.11e-6, gamma1_loss=1.03e-5, gamma2_loss=4.17e-8, gamma3_loss=4.28e-8, gamma4_loss=5.65e-8, D_loss=6.21e-8]\n"
|
|
]
|
|
}
|
|
],
|
|
"source": [
|
|
"class SinSinAB(torch.nn.Module):\n",
|
|
" \"\"\" \"\"\"\n",
|
|
" def __init__(self):\n",
|
|
" super().__init__()\n",
|
|
" self.alpha = torch.nn.Parameter(torch.tensor([1.0]))\n",
|
|
" self.beta = torch.nn.Parameter(torch.tensor([1.0]))\n",
|
|
"\n",
|
|
"\n",
|
|
" def forward(self, x):\n",
|
|
" t = (\n",
|
|
" self.beta*torch.sin(self.alpha*x.extract(['x'])*torch.pi)*\n",
|
|
" torch.sin(self.alpha*x.extract(['y'])*torch.pi)\n",
|
|
" )\n",
|
|
" return LabelTensor(t, ['b*sin(a*x)sin(a*y)'])\n",
|
|
"\n",
|
|
"\n",
|
|
"# make model + solver + trainer\n",
|
|
"model_lean= FeedForward(\n",
|
|
" layers=[10, 10],\n",
|
|
" func=Softplus,\n",
|
|
" output_dimensions=len(problem.output_variables),\n",
|
|
" input_dimensions=len(problem.input_variables)+1\n",
|
|
")\n",
|
|
"pinn_lean = PINN(problem, model_lean, extra_features=[SinSin()], optimizer_kwargs={'lr':0.006, 'weight_decay':1e-8})\n",
|
|
"trainer_learn = Trainer(pinn_lean, max_epochs=1000)\n",
|
|
"\n",
|
|
"# train\n",
|
|
"trainer_learn.train()"
|
|
]
|
|
},
|
|
{
|
|
"cell_type": "markdown",
|
|
"metadata": {},
|
|
"source": [
|
|
"Umh, the final loss is not appreciabily better than previous model (with static extra features), despite the usage of learnable parameters. This is mainly due to the over-parametrization of the network: there are many parameter to optimize during the training, and the model in unable to understand automatically that only the parameters of the extra feature (and not the weights/bias of the FFN) should be tuned in order to fit our problem. A longer training can be helpful, but in this case the faster way to reach machine precision for solving the Poisson problem is removing all the hidden layers in the `FeedForward`, keeping only the $\\alpha$ and $\\beta$ parameters of the extra feature."
|
|
]
|
|
},
|
|
{
|
|
"cell_type": "code",
|
|
"execution_count": 8,
|
|
"metadata": {},
|
|
"outputs": [
|
|
{
|
|
"name": "stderr",
|
|
"output_type": "stream",
|
|
"text": [
|
|
"GPU available: False, used: False\n",
|
|
"TPU available: False, using: 0 TPU cores\n",
|
|
"IPU available: False, using: 0 IPUs\n",
|
|
"HPU available: False, using: 0 HPUs\n",
|
|
"\n",
|
|
" | Name | Type | Params\n",
|
|
"----------------------------------------\n",
|
|
"0 | _loss | MSELoss | 0 \n",
|
|
"1 | _neural_net | Network | 4 \n",
|
|
"----------------------------------------\n",
|
|
"4 Trainable params\n",
|
|
"0 Non-trainable params\n",
|
|
"4 Total params\n",
|
|
"0.000 Total estimated model params size (MB)\n"
|
|
]
|
|
},
|
|
{
|
|
"name": "stdout",
|
|
"output_type": "stream",
|
|
"text": [
|
|
"Epoch 999: : 1it [00:00, 149.45it/s, v_num=48, mean_loss=1.34e-16, gamma1_loss=6.66e-16, gamma2_loss=2.6e-18, gamma3_loss=4.84e-19, gamma4_loss=2.59e-18, D_loss=4.84e-19] "
|
|
]
|
|
},
|
|
{
|
|
"name": "stderr",
|
|
"output_type": "stream",
|
|
"text": [
|
|
"`Trainer.fit` stopped: `max_epochs=1000` reached.\n"
|
|
]
|
|
},
|
|
{
|
|
"name": "stdout",
|
|
"output_type": "stream",
|
|
"text": [
|
|
"Epoch 999: : 1it [00:00, 117.81it/s, v_num=48, mean_loss=1.34e-16, gamma1_loss=6.66e-16, gamma2_loss=2.6e-18, gamma3_loss=4.84e-19, gamma4_loss=2.59e-18, D_loss=4.84e-19]\n"
|
|
]
|
|
}
|
|
],
|
|
"source": [
|
|
"# make model + solver + trainer\n",
|
|
"model_lean= FeedForward(\n",
|
|
" layers=[],\n",
|
|
" func=Softplus,\n",
|
|
" output_dimensions=len(problem.output_variables),\n",
|
|
" input_dimensions=len(problem.input_variables)+1\n",
|
|
")\n",
|
|
"pinn_learn = PINN(problem, model_lean, extra_features=[SinSin()], optimizer_kwargs={'lr':0.006, 'weight_decay':1e-8})\n",
|
|
"trainer_learn = Trainer(pinn_learn, max_epochs=1000, callbacks=[MetricTracker()])\n",
|
|
"\n",
|
|
"# train\n",
|
|
"trainer_learn.train()"
|
|
]
|
|
},
|
|
{
|
|
"cell_type": "markdown",
|
|
"metadata": {},
|
|
"source": [
|
|
"In such a way, the model is able to reach a very high accuracy!\n",
|
|
"Of course, this is a toy problem for understanding the usage of extra features: similar precision could be obtained if the extra features are very similar to the true solution. The analyzed Poisson problem shows a forcing term very close to the solution, resulting in a perfect problem to address with such an approach.\n",
|
|
"\n",
|
|
"We conclude here by showing the graphical comparison of the unknown field and the loss trend for all the test cases presented here: the standard PINN, PINN with extra features, and PINN with learnable extra features."
|
|
]
|
|
},
|
|
{
|
|
"cell_type": "code",
|
|
"execution_count": 9,
|
|
"metadata": {},
|
|
"outputs": [
|
|
{
|
|
"data": {
|
|
"image/png": "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",
|
|
"text/plain": [
|
|
"<Figure size 1600x600 with 6 Axes>"
|
|
]
|
|
},
|
|
"metadata": {},
|
|
"output_type": "display_data"
|
|
}
|
|
],
|
|
"source": [
|
|
"plotter.plot(trainer_learn)"
|
|
]
|
|
},
|
|
{
|
|
"cell_type": "code",
|
|
"execution_count": 10,
|
|
"metadata": {},
|
|
"outputs": [
|
|
{
|
|
"data": {
|
|
"image/png": "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",
|
|
"text/plain": [
|
|
"<Figure size 1600x600 with 1 Axes>"
|
|
]
|
|
},
|
|
"metadata": {},
|
|
"output_type": "display_data"
|
|
}
|
|
],
|
|
"source": [
|
|
"import matplotlib.pyplot as plt\n",
|
|
"\n",
|
|
"plt.figure(figsize=(16, 6))\n",
|
|
"plotter.plot_loss(trainer, label='Standard')\n",
|
|
"plotter.plot_loss(trainer_feat, label='Static Features')\n",
|
|
"plotter.plot_loss(trainer_learn, label='Learnable Features')\n",
|
|
"\n",
|
|
"plt.grid()\n",
|
|
"plt.legend()\n",
|
|
"plt.show()"
|
|
]
|
|
}
|
|
],
|
|
"metadata": {
|
|
"interpreter": {
|
|
"hash": "56be7540488f3dc66429ddf54a0fa9de50124d45fcfccfaf04c4c3886d735a3a"
|
|
},
|
|
"kernelspec": {
|
|
"display_name": "Python 3.9.16 64-bit ('dl': conda)",
|
|
"name": "python3"
|
|
},
|
|
"language_info": {
|
|
"codemirror_mode": {
|
|
"name": "ipython",
|
|
"version": 3
|
|
},
|
|
"file_extension": ".py",
|
|
"mimetype": "text/x-python",
|
|
"name": "python",
|
|
"nbconvert_exporter": "python",
|
|
"pygments_lexer": "ipython3",
|
|
"version": "3.9.16"
|
|
}
|
|
},
|
|
"nbformat": 4,
|
|
"nbformat_minor": 5
|
|
}
|