{ "cells": [ { "metadata": {}, "cell_type": "markdown", "source": [ "# Example 0: The Simplest Neuroptimiser\n", "This example demonstrates how to use the Neuroptimiser library to solve a dummy optimisation problem." ], "id": "8adf569f028839ad" }, { "metadata": {}, "cell_type": "markdown", "source": [ "## 1. Setup\n", "Import minimal necessary libraries." ], "id": "fb33b2b07c748040" }, { "cell_type": "code", "id": "initial_id", "metadata": { "collapsed": true, "ExecuteTime": { "end_time": "2025-06-17T15:46:20.277580Z", "start_time": "2025-06-17T15:46:20.247614Z" } }, "source": [ "import matplotlib.pyplot as plt\n", "import numpy as np\n", "\n", "from neuroptimiser import NeurOptimiser" ], "outputs": [], "execution_count": 1 }, { "metadata": {}, "cell_type": "markdown", "source": [ "## 2. Quick problem and optimiser setup\n", "We define a simple optimisation problem with a fitness function and bounds." ], "id": "f3a7caa4f24c16a2" }, { "metadata": { "ExecuteTime": { "end_time": "2025-06-17T15:46:20.283982Z", "start_time": "2025-06-17T15:46:20.281964Z" } }, "cell_type": "code", "source": [ "problem_function = lambda x: np.linalg.norm(x)\n", "problem_bounds = np.array([[-5.0, 5.0], [-5.0, 5.0]])" ], "id": "ffcbc9470bdb0f38", "outputs": [], "execution_count": 2 }, { "metadata": {}, "cell_type": "markdown", "source": "Then, we instantiate the Neuroptimiser with the default configurations.", "id": "f3bc216156714f0c" }, { "metadata": { "ExecuteTime": { "end_time": "2025-06-17T15:46:20.408721Z", "start_time": "2025-06-17T15:46:20.406328Z" } }, "cell_type": "code", "source": "optimiser = NeurOptimiser()", "id": "8c6f19c161e212f2", "outputs": [], "execution_count": 3 }, { "metadata": { "ExecuteTime": { "end_time": "2025-06-17T15:46:20.421458Z", "start_time": "2025-06-17T15:46:20.419413Z" } }, "cell_type": "code", "source": [ "# Show the overall configuration parameters of the optimiser\n", "print(\"DEFAULT CONFIG PARAMS:\\n\", optimiser.config_params, \"\\n\")\n", "print(\"DEFAULT CORE PARAMS:\\n\", optimiser.core_params)" ], "id": "cdf20b3ac7c0167f", "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "DEFAULT CONFIG PARAMS:\n", " {'num_iterations': 300, 'num_neighbours': 1, 'seed': 69, 'function': None, 'search_space': array([[-1, 1],\n", " [-1, 1]]), 'unit_topology': '2dr', 'core_params': {}, 'num_agents': 10, 'neuron_topology': '2dr', 'num_dimensions': 2, 'spiking_core': 'TwoDimSpikingCore'} \n", "\n", "DEFAULT CORE PARAMS:\n", " [{'coeffs': 'random', 'seed': None, 'approx': 'rk4', 'thr_alpha': 1.0, 'thr_k': 0.05, 'ref_mode': 'pg', 'thr_max': 1.0, 'spk_cond': 'fixed', 'dt': 0.01, 'hs_operator': 'fixed', 'name': 'linear', 'thr_min': 1e-06, 'max_steps': 100, 'noise_std': 0.1, 'alpha': 1.0, 'thr_mode': 'diff_pg', 'spk_alpha': 0.25, 'is_bounded': False, 'hs_variant': 'fixed'}, {'coeffs': 'random', 'seed': None, 'approx': 'rk4', 'thr_alpha': 1.0, 'thr_k': 0.05, 'ref_mode': 'pg', 'thr_max': 1.0, 'spk_cond': 'fixed', 'dt': 0.01, 'hs_operator': 'fixed', 'name': 'linear', 'thr_min': 1e-06, 'max_steps': 100, 'noise_std': 0.1, 'alpha': 1.0, 'thr_mode': 'diff_pg', 'spk_alpha': 0.25, 'is_bounded': False, 'hs_variant': 'fixed'}, {'coeffs': 'random', 'seed': None, 'approx': 'rk4', 'thr_alpha': 1.0, 'thr_k': 0.05, 'ref_mode': 'pg', 'thr_max': 1.0, 'spk_cond': 'fixed', 'dt': 0.01, 'hs_operator': 'fixed', 'name': 'linear', 'thr_min': 1e-06, 'max_steps': 100, 'noise_std': 0.1, 'alpha': 1.0, 'thr_mode': 'diff_pg', 'spk_alpha': 0.25, 'is_bounded': False, 'hs_variant': 'fixed'}, {'coeffs': 'random', 'seed': None, 'approx': 'rk4', 'thr_alpha': 1.0, 'thr_k': 0.05, 'ref_mode': 'pg', 'thr_max': 1.0, 'spk_cond': 'fixed', 'dt': 0.01, 'hs_operator': 'fixed', 'name': 'linear', 'thr_min': 1e-06, 'max_steps': 100, 'noise_std': 0.1, 'alpha': 1.0, 'thr_mode': 'diff_pg', 'spk_alpha': 0.25, 'is_bounded': False, 'hs_variant': 'fixed'}, {'coeffs': 'random', 'seed': None, 'approx': 'rk4', 'thr_alpha': 1.0, 'thr_k': 0.05, 'ref_mode': 'pg', 'thr_max': 1.0, 'spk_cond': 'fixed', 'dt': 0.01, 'hs_operator': 'fixed', 'name': 'linear', 'thr_min': 1e-06, 'max_steps': 100, 'noise_std': 0.1, 'alpha': 1.0, 'thr_mode': 'diff_pg', 'spk_alpha': 0.25, 'is_bounded': False, 'hs_variant': 'fixed'}, {'coeffs': 'random', 'seed': None, 'approx': 'rk4', 'thr_alpha': 1.0, 'thr_k': 0.05, 'ref_mode': 'pg', 'thr_max': 1.0, 'spk_cond': 'fixed', 'dt': 0.01, 'hs_operator': 'fixed', 'name': 'linear', 'thr_min': 1e-06, 'max_steps': 100, 'noise_std': 0.1, 'alpha': 1.0, 'thr_mode': 'diff_pg', 'spk_alpha': 0.25, 'is_bounded': False, 'hs_variant': 'fixed'}, {'coeffs': 'random', 'seed': None, 'approx': 'rk4', 'thr_alpha': 1.0, 'thr_k': 0.05, 'ref_mode': 'pg', 'thr_max': 1.0, 'spk_cond': 'fixed', 'dt': 0.01, 'hs_operator': 'fixed', 'name': 'linear', 'thr_min': 1e-06, 'max_steps': 100, 'noise_std': 0.1, 'alpha': 1.0, 'thr_mode': 'diff_pg', 'spk_alpha': 0.25, 'is_bounded': False, 'hs_variant': 'fixed'}, {'coeffs': 'random', 'seed': None, 'approx': 'rk4', 'thr_alpha': 1.0, 'thr_k': 0.05, 'ref_mode': 'pg', 'thr_max': 1.0, 'spk_cond': 'fixed', 'dt': 0.01, 'hs_operator': 'fixed', 'name': 'linear', 'thr_min': 1e-06, 'max_steps': 100, 'noise_std': 0.1, 'alpha': 1.0, 'thr_mode': 'diff_pg', 'spk_alpha': 0.25, 'is_bounded': False, 'hs_variant': 'fixed'}, {'coeffs': 'random', 'seed': None, 'approx': 'rk4', 'thr_alpha': 1.0, 'thr_k': 0.05, 'ref_mode': 'pg', 'thr_max': 1.0, 'spk_cond': 'fixed', 'dt': 0.01, 'hs_operator': 'fixed', 'name': 'linear', 'thr_min': 1e-06, 'max_steps': 100, 'noise_std': 0.1, 'alpha': 1.0, 'thr_mode': 'diff_pg', 'spk_alpha': 0.25, 'is_bounded': False, 'hs_variant': 'fixed'}, {'coeffs': 'random', 'seed': None, 'approx': 'rk4', 'thr_alpha': 1.0, 'thr_k': 0.05, 'ref_mode': 'pg', 'thr_max': 1.0, 'spk_cond': 'fixed', 'dt': 0.01, 'hs_operator': 'fixed', 'name': 'linear', 'thr_min': 1e-06, 'max_steps': 100, 'noise_std': 0.1, 'alpha': 1.0, 'thr_mode': 'diff_pg', 'spk_alpha': 0.25, 'is_bounded': False, 'hs_variant': 'fixed'}]\n" ] } ], "execution_count": 4 }, { "metadata": {}, "cell_type": "markdown", "source": [ "## 3. Optimisation process\n", "We proceed to solve the optimisation problem using the `solve` method of the `NeurOptimiser` process. In this example, we enable the debug mode to get more detailed output during the optimisation process." ], "id": "23ebe893982271b9" }, { "metadata": { "ExecuteTime": { "end_time": "2025-06-17T15:46:25.228158Z", "start_time": "2025-06-17T15:46:20.442500Z" } }, "cell_type": "code", "source": [ "optimiser.solve(\n", " obj_func=problem_function,\n", " search_space=problem_bounds,\n", " debug_mode=True\n", ")" ], "id": "15d44d1385cd700b", "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "[neuropt:log] Debug mode is enabled. Monitoring will be activated.\n", "[neuropt:log] Parameters are set up.\n", "[neuropt:log] Initial positions and topologies are set up.\n", "[neuropt:log] Tensor contraction layer, neighbourhood manager, and high-level selection unit are created.\n", "[neuropt:log] Population of nheuristic units is created.\n", "[neuropt:log] Connections between nheuristic units and auxiliary processes are established.\n", "[neuropt:log] Monitors are set up.\n", "[neuropt:log] Starting simulation with 300 iterations...\n", "... step: 0, best fitness: 2.025650978088379\n", "... step: 30, best fitness: 1.113487958908081\n", "... step: 60, best fitness: 0.15552423894405365\n", "... step: 90, best fitness: 0.1322910338640213\n", "... step: 120, best fitness: 0.024050353094935417\n", "... step: 150, best fitness: 0.024050353094935417\n", "... step: 180, best fitness: 0.0138082941994071\n", "... step: 210, best fitness: 0.013055730611085892\n", "... step: 240, best fitness: 0.013055730611085892\n", "... step: 270, best fitness: 0.013055730611085892\n", "... step: 299, best fitness: 0.013055730611085892\n", "[neuropt:log] Simulation completed. Fetching monitor data... done\n" ] }, { "data": { "text/plain": [ "(array([-0.00682959, 0.0111269 ]), array([0.01305573]))" ] }, "execution_count": 5, "metadata": {}, "output_type": "execute_result" } ], "execution_count": 5 }, { "metadata": {}, "cell_type": "markdown", "source": [ "## (Optional) 4. Results processing and visualisation\n", "We process the results obtained from the optimiser and visualise the absolute error in fitness values over the optimisation steps." ], "id": "8b985328ea61fba7" }, { "metadata": { "ExecuteTime": { "end_time": "2025-06-17T15:46:25.249446Z", "start_time": "2025-06-17T15:46:25.244931Z" } }, "cell_type": "code", "source": [ "# Recover the results from the optimiser\n", "fp = optimiser.results[\"fp\"]\n", "fg = optimiser.results[\"fg\"]\n", "positions = np.array(optimiser.results[\"p\"])\n", "best_position = np.array(optimiser.results[\"g\"])\n", "v1 = np.array(optimiser.results[\"v1\"])\n", "v2 = np.array(optimiser.results[\"v2\"])\n", "\n", "# Calculate the absolute error in fitness values\n", "efp = np.abs(np.array(fp))\n", "efg = np.abs(np.array(fg))\n", "\n", "# Convert the spikes to integer type\n", "spikes = np.array(optimiser.results[\"s\"]).astype(int)\n", "\n", "# Print some minimal information about the results\n", "print(f\"fg: {fg[-1][0]:.4f}, f*: {0.0:.4f}, error: {efg[-1][0]:.4e}\")\n", "print(f\"norm2(g - x*): {np.linalg.norm(best_position[-1]):.4e}\")\n", "print(f\"{v1.min():.4f} <= v1 <= {v1.max():.4f}\")\n", "print(f\"{v2.min():.4f} <= v2 <= {v2.max():.4f}\")" ], "id": "2ce73f98f2651bcb", "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "fg: 0.0131, f*: 0.0000, error: 1.3056e-02\n", "norm2(g - x*): 1.3056e-02\n", "-1.1394 <= v1 <= 1.2176\n", "-0.8110 <= v2 <= 0.9703\n" ] } ], "execution_count": 6 }, { "metadata": { "ExecuteTime": { "end_time": "2025-06-17T15:46:25.638387Z", "start_time": "2025-06-17T15:46:25.270365Z" } }, "cell_type": "code", "source": [ "fig, ax = plt.subplots(figsize=(6.9, 6.9*0.618))\n", "\n", "plt.plot(efp, color=\"silver\", alpha=0.5)\n", "plt.plot(np.max(efp, axis=1), '--', color=\"red\", label=r\"Max.\")\n", "plt.plot(np.average(efp, axis=1), '--', color=\"black\", label=r\"Mean\")\n", "plt.plot(np.median(efp, axis=1), '--', color=\"blue\", label=r\"Median\")\n", "plt.plot(efg, '--', color=\"green\", label=r\"Min.\")\n", "\n", "plt.xlabel(r\"Step, $t$\")\n", "plt.ylabel(r\"Abs. Error, $\\varepsilon_f$\")\n", "\n", "lgd = plt.legend(ncol=2, loc=\"lower left\")\n", "\n", "plt.xscale(\"log\")\n", "plt.yscale(\"log\")\n", "\n", "ax.patch.set_alpha(0)\n", "fig.tight_layout()" ], "id": "359cbb6cd7c18d57", "outputs": [ { "data": { "text/plain": [ "
" ], "image/png": "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" }, "metadata": {}, "output_type": "display_data" } ], "execution_count": 7 } ], "metadata": { "kernelspec": { "display_name": "Python 3", "language": "python", "name": "python3" }, "language_info": { "codemirror_mode": { "name": "ipython", "version": 2 }, "file_extension": ".py", "mimetype": "text/x-python", "name": "python", "nbconvert_exporter": "python", "pygments_lexer": "ipython2", "version": "2.7.6" } }, "nbformat": 4, "nbformat_minor": 5 }