{ "cells": [ { "cell_type": "markdown", "metadata": {}, "source": [ "# GPyTorch regression with derivative information\n", "\n", "## Introduction\n", "In this notebook, we show how to train a GP regression model in GPyTorch of an unknown function given function value and derivative observations. We consider modeling the function:\n", "\n", "$$\n", "\\begin{align}\n", " y &= \\sin(2x) + cos(x) + \\epsilon \\\\\n", " \\frac{dy}{dx} &= 2\\cos(2x) - \\sin(x) + \\epsilon \\\\ \n", " \\epsilon &\\sim \\mathcal{N}(0, 0.5)\n", "\\end{align}\n", "$$\n", "\n", "using 50 value and derivative observations." ] }, { "cell_type": "code", "execution_count": 1, "metadata": {}, "outputs": [], "source": [ "import torch\n", "import gpytorch\n", "import math\n", "from matplotlib import pyplot as plt\n", "import numpy as np\n", "\n", "%matplotlib inline\n", "%load_ext autoreload\n", "%autoreload 2" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "## Setting up the training data\n", "We use 50 uniformly distributed points in the interval $[0, 5 \\pi]$" ] }, { "cell_type": "code", "execution_count": 2, "metadata": {}, "outputs": [], "source": [ "lb, ub = 0.0, 5*math.pi\n", "n = 50\n", "\n", "train_x = torch.linspace(lb, ub, n).unsqueeze(-1)\n", "train_y = torch.stack([\n", " torch.sin(2*train_x) + torch.cos(train_x), \n", " -torch.sin(train_x) + 2*torch.cos(2*train_x)\n", "], -1).squeeze(1)\n", "\n", "train_y += 0.05 * torch.randn(n, 2)" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "## Setting up the model\n", "A GP prior on the function values implies a multi-output GP prior on the function values and the partial derivatives, see 9.4 in http://www.gaussianprocess.org/gpml/chapters/RW9.pdf for more details. This allows using a `MultitaskMultivariateNormal` and `MultitaskGaussianLikelihood` to train a GP model from both function values and gradients. The resulting RBF kernel that models the covariance between the values and partial derivatives has been implemented in `RBFKernelGrad` and the extension of a constant mean is implemented in `ConstantMeanGrad`.\n", "\n", "The `RBFKernelGrad` is generally worse conditioned than the `RBFKernel`, so we place a lower bound on the noise parameter to keep the smallest eigenvalues of the kernel matrix away from zero." ] }, { "cell_type": "code", "execution_count": 3, "metadata": {}, "outputs": [], "source": [ "class GPModelWithDerivatives(gpytorch.models.ExactGP):\n", " def __init__(self, train_x, train_y, likelihood):\n", " super(GPModelWithDerivatives, self).__init__(train_x, train_y, likelihood)\n", " self.mean_module = gpytorch.means.ConstantMeanGrad()\n", " self.base_kernel = gpytorch.kernels.RBFKernelGrad()\n", " self.covar_module = gpytorch.kernels.ScaleKernel(self.base_kernel)\n", "\n", " def forward(self, x):\n", " mean_x = self.mean_module(x)\n", " covar_x = self.covar_module(x)\n", " return gpytorch.distributions.MultitaskMultivariateNormal(mean_x, covar_x)\n", "\n", "likelihood = gpytorch.likelihoods.MultitaskGaussianLikelihood(num_tasks=2) # Value + Derivative\n", "model = GPModelWithDerivatives(train_x, train_y, likelihood)" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "The model training is similar to training a standard GP regression model" ] }, { "cell_type": "code", "execution_count": 4, "metadata": {}, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "Iter 1/50 - Loss: 71.141 lengthscale: 0.693 noise: 0.693\n", "Iter 2/50 - Loss: 69.100 lengthscale: 0.744 noise: 0.644\n", "Iter 3/50 - Loss: 66.347 lengthscale: 0.797 noise: 0.598\n", "Iter 4/50 - Loss: 64.771 lengthscale: 0.845 noise: 0.554\n", "Iter 5/50 - Loss: 63.744 lengthscale: 0.886 noise: 0.513\n", "Iter 6/50 - Loss: 61.682 lengthscale: 0.928 noise: 0.474\n", "Iter 7/50 - Loss: 59.820 lengthscale: 0.961 noise: 0.437\n", "Iter 8/50 - Loss: 57.801 lengthscale: 0.987 noise: 0.402\n", "Iter 9/50 - Loss: 56.894 lengthscale: 1.004 noise: 0.370\n", "Iter 10/50 - Loss: 54.522 lengthscale: 1.010 noise: 0.340\n", "Iter 11/50 - Loss: 53.263 lengthscale: 1.006 noise: 0.311\n", "Iter 12/50 - Loss: 50.900 lengthscale: 0.998 noise: 0.285\n", "Iter 13/50 - Loss: 49.472 lengthscale: 0.986 noise: 0.260\n", "Iter 14/50 - Loss: 47.405 lengthscale: 0.980 noise: 0.238\n", "Iter 15/50 - Loss: 46.851 lengthscale: 0.982 noise: 0.217\n", "Iter 16/50 - Loss: 43.638 lengthscale: 0.991 noise: 0.198\n", "Iter 17/50 - Loss: 42.900 lengthscale: 1.002 noise: 0.180\n", "Iter 18/50 - Loss: 39.969 lengthscale: 1.021 noise: 0.164\n", "Iter 19/50 - Loss: 38.408 lengthscale: 1.040 noise: 0.149\n", "Iter 20/50 - Loss: 35.881 lengthscale: 1.059 noise: 0.135\n", "Iter 21/50 - Loss: 34.669 lengthscale: 1.078 noise: 0.122\n", "Iter 22/50 - Loss: 32.928 lengthscale: 1.097 noise: 0.111\n", "Iter 23/50 - Loss: 30.690 lengthscale: 1.113 noise: 0.100\n", "Iter 24/50 - Loss: 28.567 lengthscale: 1.127 noise: 0.091\n", "Iter 25/50 - Loss: 27.540 lengthscale: 1.138 noise: 0.082\n", "Iter 26/50 - Loss: 24.865 lengthscale: 1.142 noise: 0.074\n", "Iter 27/50 - Loss: 23.273 lengthscale: 1.141 noise: 0.067\n", "Iter 28/50 - Loss: 20.533 lengthscale: 1.147 noise: 0.061\n", "Iter 29/50 - Loss: 19.787 lengthscale: 1.144 noise: 0.055\n", "Iter 30/50 - Loss: 16.676 lengthscale: 1.146 noise: 0.050\n", "Iter 31/50 - Loss: 14.890 lengthscale: 1.151 noise: 0.045\n", "Iter 32/50 - Loss: 13.735 lengthscale: 1.158 noise: 0.040\n", "Iter 33/50 - Loss: 11.772 lengthscale: 1.171 noise: 0.036\n", "Iter 34/50 - Loss: 9.266 lengthscale: 1.182 noise: 0.033\n", "Iter 35/50 - Loss: 7.507 lengthscale: 1.193 noise: 0.030\n", "Iter 36/50 - Loss: 5.724 lengthscale: 1.195 noise: 0.027\n", "Iter 37/50 - Loss: 5.030 lengthscale: 1.195 noise: 0.024\n", "Iter 38/50 - Loss: 1.297 lengthscale: 1.207 noise: 0.022\n", "Iter 39/50 - Loss: -0.072 lengthscale: 1.211 noise: 0.020\n", "Iter 40/50 - Loss: -2.852 lengthscale: 1.208 noise: 0.018\n", "Iter 41/50 - Loss: -4.053 lengthscale: 1.200 noise: 0.016\n", "Iter 42/50 - Loss: -5.160 lengthscale: 1.198 noise: 0.014\n", "Iter 43/50 - Loss: -8.217 lengthscale: 1.208 noise: 0.013\n", "Iter 44/50 - Loss: -8.949 lengthscale: 1.216 noise: 0.012\n", "Iter 45/50 - Loss: -11.805 lengthscale: 1.228 noise: 0.011\n", "Iter 46/50 - Loss: -14.472 lengthscale: 1.230 noise: 0.010\n", "Iter 47/50 - Loss: -17.141 lengthscale: 1.228 noise: 0.009\n", "Iter 48/50 - Loss: -16.575 lengthscale: 1.215 noise: 0.008\n", "Iter 49/50 - Loss: -18.488 lengthscale: 1.204 noise: 0.007\n", "Iter 50/50 - Loss: -20.305 lengthscale: 1.207 noise: 0.007\n" ] } ], "source": [ "# this is for running the notebook in our testing framework\n", "import os\n", "smoke_test = ('CI' in os.environ)\n", "training_iter = 2 if smoke_test else 50\n", "\n", "\n", "# Find optimal model hyperparameters\n", "model.train()\n", "likelihood.train()\n", "\n", "# Use the adam optimizer\n", "optimizer = torch.optim.Adam(model.parameters(), lr=0.1) # Includes GaussianLikelihood parameters\n", "\n", "# \"Loss\" for GPs - the marginal log likelihood\n", "mll = gpytorch.mlls.ExactMarginalLogLikelihood(likelihood, model)\n", "\n", "for i in range(training_iter):\n", " optimizer.zero_grad()\n", " output = model(train_x)\n", " loss = -mll(output, train_y)\n", " loss.backward()\n", " print('Iter %d/%d - Loss: %.3f lengthscale: %.3f noise: %.3f' % (\n", " i + 1, training_iter, loss.item(),\n", " model.covar_module.base_kernel.lengthscale.item(),\n", " model.likelihood.noise.item()\n", " )) \n", " optimizer.step()" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "Model predictions are also similar to GP regression with only function values, butwe need more CG iterations to get accurate estimates of the predictive variance" ] }, { "cell_type": "code", "execution_count": 5, "metadata": {}, "outputs": [ { "data": { "image/png": "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\n", "text/plain": [ "
" ] }, "metadata": { "needs_background": "light" }, "output_type": "display_data" } ], "source": [ "# Set into eval mode\n", "model.train()\n", "model.eval()\n", "likelihood.eval()\n", "\n", "# Initialize plots\n", "f, (y1_ax, y2_ax) = plt.subplots(1, 2, figsize=(12, 6))\n", "\n", "# Make predictions\n", "with torch.no_grad(), gpytorch.settings.max_cg_iterations(50):\n", " test_x = torch.linspace(lb, ub, 500)\n", " predictions = likelihood(model(test_x))\n", " mean = predictions.mean\n", " lower, upper = predictions.confidence_region()\n", " \n", "# Plot training data as black stars\n", "y1_ax.plot(train_x.detach().numpy(), train_y[:, 0].detach().numpy(), 'k*')\n", "# Predictive mean as blue line\n", "y1_ax.plot(test_x.numpy(), mean[:, 0].numpy(), 'b')\n", "# Shade in confidence \n", "y1_ax.fill_between(test_x.numpy(), lower[:, 0].numpy(), upper[:, 0].numpy(), alpha=0.5)\n", "y1_ax.legend(['Observed Values', 'Mean', 'Confidence'])\n", "y1_ax.set_title('Function values')\n", "\n", "# Plot training data as black stars\n", "y2_ax.plot(train_x.detach().numpy(), train_y[:, 1].detach().numpy(), 'k*')\n", "# Predictive mean as blue line\n", "y2_ax.plot(test_x.numpy(), mean[:, 1].numpy(), 'b')\n", "# Shade in confidence \n", "y2_ax.fill_between(test_x.numpy(), lower[:, 1].numpy(), upper[:, 1].numpy(), alpha=0.5)\n", "y2_ax.legend(['Observed Derivatives', 'Mean', 'Confidence'])\n", "y2_ax.set_title('Derivatives')\n", "\n", "None" ] }, { "cell_type": "code", "execution_count": null, "metadata": {}, "outputs": [], "source": [] }, { "cell_type": "code", "execution_count": null, "metadata": {}, "outputs": [], "source": [] } ], "metadata": { "kernelspec": { "display_name": "Python 3", "language": "python", "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": "3.7.3" } }, "nbformat": 4, "nbformat_minor": 2 }