{ "cells": [ { "cell_type": "markdown", "id": "fef2775c-7f84-4f1d-b191-00d24c3c2160", "metadata": {}, "source": [ "# Simple Koopman pipeline\n", "\n", "This example uses simulation data from a simple mass-spring-damper to demonstrate how to use `KoopmanPipeline`, the most important class in `pykoop`. A Koopman pipeline consists of a set of lifting functions, which transform states into a higher dimensional space, and a regressor, that performs linear regression in that space." ] }, { "cell_type": "code", "execution_count": 1, "id": "f13aba65-eb88-42e9-8e1c-ba4fdf5f35df", "metadata": {}, "outputs": [], "source": [ "# Imports\n", "from matplotlib import pyplot as plt\n", "from sklearn.preprocessing import MaxAbsScaler, StandardScaler\n", "\n", "import pykoop\n", "\n", "# Set plot defaults\n", "plt.rc('lines', linewidth=2)\n", "plt.rc('axes', grid=True)\n", "plt.rc('grid', linestyle='--')" ] }, { "cell_type": "markdown", "id": "981b1158-5e69-42db-91b5-1a89a102d1bf", "metadata": {}, "source": [ "Load example data from the library. `eg` is a `dict` containing training data, validation data, and a few related parameters." ] }, { "cell_type": "code", "execution_count": 2, "id": "0690bdf5-b7fc-457c-8a5a-68fea34f341e", "metadata": {}, "outputs": [], "source": [ "eg = pykoop.example_data_msd()" ] }, { "cell_type": "markdown", "id": "73164194-751a-4e19-bc16-55c85023b5d2", "metadata": {}, "source": [ "Create the Koopman pipeline. It accepts a list of lifting functions, which are composed to form the final transformation. Each lifting function has a string name, which can be used for cross-validating over lifting function parameters." ] }, { "cell_type": "code", "execution_count": 3, "id": "76b73844-93d8-49aa-be68-e58d047a246e", "metadata": {}, "outputs": [], "source": [ "kp = pykoop.KoopmanPipeline(\n", " lifting_functions=[\n", " ('ma', pykoop.SkLearnLiftingFn(MaxAbsScaler())),\n", " ('pl', pykoop.PolynomialLiftingFn(order=2)),\n", " ('ss', pykoop.SkLearnLiftingFn(StandardScaler())),\n", " ],\n", " regressor=pykoop.Edmd(alpha=1),\n", ")" ] }, { "cell_type": "markdown", "id": "3ede1b44-219e-4638-8cef-d8d187879403", "metadata": {}, "source": [ "Fit the pipeline. The data format consists of an optional episode feature, which groups individual experiments, states, and exogenous inputs. The `fit()` method needs to know if there is an episode feature, and how many inputs there are." ] }, { "cell_type": "code", "execution_count": 4, "id": "22adedb5-8cfc-4109-bb9e-52e640da8e07", "metadata": {}, "outputs": [ { "data": { "text/plain": [ "KoopmanPipeline(lifting_functions=[('ma',\n", " SkLearnLiftingFn(transformer=MaxAbsScaler())),\n", " ('pl', PolynomialLiftingFn(order=2)),\n", " ('ss',\n", " SkLearnLiftingFn(transformer=StandardScaler()))],\n", " regressor=Edmd(alpha=1))" ] }, "execution_count": 4, "metadata": {}, "output_type": "execute_result" } ], "source": [ "kp.fit(\n", " eg['X_train'],\n", " n_inputs=eg['n_inputs'],\n", " episode_feature=eg['episode_feature'],\n", ")" ] }, { "cell_type": "markdown", "id": "c33281df-3c16-4ea9-9427-49a9b606d03d", "metadata": {}, "source": [ "This is the matrix approximation of the Koopman operator. It needs to be transposed because `scikit-learn` puts time on the first axis and features on the second." ] }, { "cell_type": "code", "execution_count": 5, "id": "1a45fd15-acfb-471d-8be6-0ea82389b486", "metadata": {}, "outputs": [ { "data": { "text/plain": [ "array([[ 9.83821877e-01, 7.37395653e-02, 1.14097993e-03,\n", " 3.56958678e-04, -1.31762746e-02, 2.69532030e-02,\n", " -2.12135080e-03, 1.96144371e-02, -7.31646538e-03],\n", " [-1.51394823e-01, 8.44690217e-01, 1.40948997e-04,\n", " -6.45670543e-03, -2.39662729e-02, 1.77247216e-01,\n", " 4.60499946e-03, 3.76649110e-02, -1.53728812e-02],\n", " [ 1.16916967e-03, 2.29797118e-04, 9.58696283e-01,\n", " 9.28153811e-02, 1.75676259e-02, 3.23235312e-04,\n", " 7.60167038e-02, -2.28520931e-03, -8.46063574e-03],\n", " [-3.81899442e-04, -7.09619495e-03, -1.96816403e-01,\n", " 7.84429668e-01, 6.79723612e-02, 8.86434652e-03,\n", " 2.05522321e-01, 7.49812518e-02, -1.85952809e-02],\n", " [-3.65272037e-03, -1.08771186e-02, 3.38235034e-02,\n", " -1.83559626e-01, 6.86068831e-01, 1.16734342e-02,\n", " -6.15615705e-02, 3.28055216e-01, -4.72169811e-03]])" ] }, "execution_count": 5, "metadata": {}, "output_type": "execute_result" } ], "source": [ "kp.regressor_.coef_.T" ] }, { "cell_type": "markdown", "id": "1c350f8c-3916-4c17-bf98-bc89c29a86b9", "metadata": {}, "source": [ "Given the fit pipeline, predict the system's response to new initial conditions and inputs." ] }, { "cell_type": "code", "execution_count": 6, "id": "36e4b8e6-ffa1-4362-8bf1-795616d2daed", "metadata": {}, "outputs": [], "source": [ "X_pred = kp.predict_trajectory(eg['x0_valid'], eg['u_valid'])" ] }, { "cell_type": "markdown", "id": "56772b09-5b95-4832-bd0c-b15bda61118c", "metadata": {}, "source": [ "Score the episode against a validation set." ] }, { "cell_type": "code", "execution_count": 7, "id": "9cb89b94-f8ef-45ec-b1c7-7f34024adf33", "metadata": {}, "outputs": [], "source": [ "score = kp.score(eg['X_valid'])" ] }, { "cell_type": "markdown", "id": "23c3a595-5de4-41c5-933f-c3fd104cf2b5", "metadata": {}, "source": [ "Plot the prediction versus the validation set." ] }, { "cell_type": "code", "execution_count": 8, "id": "fc59ebe5-9e19-4345-852a-f72d72aebf81", "metadata": {}, "outputs": [ { "data": { "text/plain": [ "" ] }, "execution_count": 8, "metadata": {}, "output_type": "execute_result" }, { "data": { "image/png": "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\n", "text/plain": [ "
" ] }, "metadata": { "needs_background": "light" }, "output_type": "display_data" } ], "source": [ "fig, ax = plt.subplots(\n", " kp.n_states_in_ + kp.n_inputs_in_,\n", " 1,\n", " constrained_layout=True,\n", " sharex=True,\n", " figsize=(6, 6),\n", ")\n", "# Plot true trajectory\n", "ax[0].plot(eg['t'], eg['X_valid'][:, 1], label='True trajectory')\n", "ax[1].plot(eg['t'], eg['X_valid'][:, 2])\n", "ax[2].plot(eg['t'], eg['X_valid'][:, 3])\n", "# Plot predicted trajectory\n", "ax[0].plot(eg['t'], X_pred[:, 1], '--', label='Predicted trajectory')\n", "ax[1].plot(eg['t'], X_pred[:, 2], '--')\n", "# Add labels\n", "ax[-1].set_xlabel('$t$')\n", "ax[0].set_ylabel('$x(t)$')\n", "ax[1].set_ylabel(r'$\\dot{x}(t)$')\n", "ax[2].set_ylabel('$u$')\n", "ax[0].set_title(f'True and predicted states; MSE={-1 * score:.2e}')\n", "ax[0].legend(loc='upper right')" ] } ], "metadata": { "kernelspec": { "display_name": "pykoop", "language": "python", "name": "pykoop" }, "language_info": { "codemirror_mode": { "name": "ipython", "version": 3 }, "file_extension": ".py", "mimetype": "text/x-python", "name": "python", "nbconvert_exporter": "python", "pygments_lexer": "ipython3", "version": "3.10.6" } }, "nbformat": 4, "nbformat_minor": 5 }