{ "cells": [ { "cell_type": "markdown", "id": "2ce28f2b-008b-4beb-ba51-5c478e4bdec7", "metadata": {}, "source": [ "# Symbolic Regression" ] }, { "cell_type": "markdown", "id": "d01a09d3-f78c-4b46-acd9-3a4250bb34c9", "metadata": {}, "source": [ "This example combines neural differential equations with regularised evolution to discover the equations\n", "\n", "$\\frac{\\mathrm{d} x}{\\mathrm{d} t}(t) = \\frac{y(t)}{1 + y(t)}$\n", "\n", "$\\frac{\\mathrm{d} y}{\\mathrm{d} t}(t) = \\frac{-x(t)}{1 + x(t)}$\n", "\n", "directly from data.\n", "\n", "**References:**\n", "\n", "This example appears as an example in Section 6.1 of:\n", "\n", "```bibtex\n", "@phdthesis{kidger2021on,\n", " title={{O}n {N}eural {D}ifferential {E}quations},\n", " author={Patrick Kidger},\n", " year={2021},\n", " school={University of Oxford},\n", "}\n", "```\n", "\n", "Whilst drawing heavy inspiration from:\n", "\n", "```bibtex\n", "@inproceedings{cranmer2020discovering,\n", " title={{D}iscovering {S}ymbolic {M}odels from {D}eep {L}earning with {I}nductive\n", " {B}iases},\n", " author={Cranmer, Miles and Sanchez Gonzalez, Alvaro and Battaglia, Peter and\n", " Xu, Rui and Cranmer, Kyle and Spergel, David and Ho, Shirley},\n", " booktitle={Advances in Neural Information Processing Systems},\n", " publisher={Curran Associates, Inc.},\n", " year={2020},\n", "}\n", "\n", "@software{cranmer2020pysr,\n", " title={PySR: Fast \\& Parallelized Symbolic Regression in Python/Julia},\n", " author={Miles Cranmer},\n", " publisher={Zenodo},\n", " url={http://doi.org/10.5281/zenodo.4041459},\n", " year={2020},\n", "}\n", "```\n", "\n", "This example is available as a Jupyter notebook [here](https://github.com/patrick-kidger/diffrax/blob/main/examples/symbolic_regression.ipynb)." ] }, { "cell_type": "code", "execution_count": 1, "id": "dea04fa4-a95b-47f8-b0eb-be297037bc7d", "metadata": {}, "outputs": [], "source": [ "import tempfile\n", "\n", "import equinox as eqx # https://github.com/patrick-kidger/equinox\n", "import jax\n", "import jax.numpy as jnp\n", "import optax # https://github.com/deepmind/optax\n", "import pysr # https://github.com/MilesCranmer/PySR\n", "import sympy\n", "import sympy2jax # https://github.com/google/sympy2jax\n", "\n", "\n", "# Note that PySR, which we use for symbolic regression, uses Julia as a backend.\n", "# You'll need to install a recent version of Julia if you don't have one.\n", "# (And can get funny errors if you have a too-old version of Julia already.)\n", "# You may also need to restart Python after running `pysr.install()` the first time.\n", "pysr.install(quiet=True)" ] }, { "cell_type": "markdown", "id": "4d26c41f-7682-4ad0-aa33-77e22b2768f8", "metadata": {}, "source": [ "Now two helpers. We'll use these in a moment; skip over them for now." ] }, { "cell_type": "code", "execution_count": 2, "id": "0d294688-3ea9-43ce-855f-cc587de908a5", "metadata": {}, "outputs": [], "source": [ "class Stack(eqx.Module):\n", " modules: list[eqx.Module]\n", "\n", " def __call__(self, x):\n", " assert x.shape[-1] == 2\n", " x0 = x[..., 0]\n", " x1 = x[..., 1]\n", " return jnp.stack([module(x0=x0, x1=x1) for module in self.modules], axis=-1)\n", "\n", "\n", "def quantise(expr, quantise_to):\n", " if isinstance(expr, sympy.Float):\n", " return expr.func(round(float(expr) / quantise_to) * quantise_to)\n", " elif isinstance(expr, sympy.Symbol):\n", " return expr\n", " else:\n", " return expr.func(*[quantise(arg, quantise_to) for arg in expr.args])" ] }, { "cell_type": "markdown", "id": "35245e23-17e0-4c7e-bb60-3e462b9b3b3c", "metadata": {}, "source": [ "Okay, let's get started.\n", "\n", "We start by running the [Neural ODE example](./neural_ode.ipynb).\n", "Then we extract the learnt neural vector field, and symbolically regress across this.\n", "Finally we fine-tune the resulting symbolic expression.\n" ] }, { "cell_type": "code", "execution_count": 3, "id": "b0525ef3-979f-4cf5-b15f-530eae7b8ae0", "metadata": {}, "outputs": [], "source": [ "def main(\n", " symbolic_dataset_size=2000,\n", " symbolic_num_populations=100,\n", " symbolic_population_size=20,\n", " symbolic_migration_steps=4,\n", " symbolic_mutation_steps=30,\n", " symbolic_descent_steps=50,\n", " pareto_coefficient=2,\n", " fine_tuning_steps=500,\n", " fine_tuning_lr=3e-3,\n", " quantise_to=0.01,\n", "):\n", " #\n", " # First obtain a neural approximation to the dynamics.\n", " # We begin by running the previous example.\n", " #\n", "\n", " # Runs the Neural ODE example.\n", " # This defines the variables `ts`, `ys`, `model`.\n", " print(\"Training neural differential equation.\")\n", " %run neural_ode.ipynb\n", "\n", " #\n", " # Now symbolically regress across the learnt vector field, to obtain a Pareto\n", " # frontier of symbolic equations, that trades loss against complexity of the\n", " # equation. Select the \"best\" from this frontier.\n", " #\n", "\n", " print(\"Symbolically regressing across the vector field.\")\n", " vector_field = model.func.mlp # noqa: F821\n", " dataset_size, length_size, data_size = ys.shape # noqa: F821\n", " in_ = ys.reshape(dataset_size * length_size, data_size) # noqa: F821\n", " in_ = in_[:symbolic_dataset_size]\n", " out = jax.vmap(vector_field)(in_)\n", " with tempfile.TemporaryDirectory() as tempdir:\n", " symbolic_regressor = pysr.PySRRegressor(\n", " niterations=symbolic_migration_steps,\n", " ncyclesperiteration=symbolic_mutation_steps,\n", " populations=symbolic_num_populations,\n", " population_size=symbolic_population_size,\n", " optimizer_iterations=symbolic_descent_steps,\n", " optimizer_nrestarts=1,\n", " procs=1,\n", " model_selection=\"score\",\n", " progress=False,\n", " tempdir=tempdir,\n", " temp_equation_file=True,\n", " )\n", " symbolic_regressor.fit(in_, out)\n", " best_expressions = [b.sympy_format for b in symbolic_regressor.get_best()]\n", "\n", " #\n", " # Now the constants in this expression have been optimised for regressing across\n", " # the neural vector field. This was good enough to obtain the symbolic expression,\n", " # but won't quite be perfect -- some of the constants will be slightly off.\n", " #\n", " # To fix this we now plug our symbolic function back into the original dataset\n", " # and apply gradient descent.\n", " #\n", "\n", " print(\"\\nOptimising symbolic expression.\")\n", "\n", " symbolic_fn = Stack([sympy2jax.SymbolicModule(expr) for expr in best_expressions])\n", " symbolic_model = eqx.tree_at(lambda m: m.func.mlp, model, symbolic_fn) # noqa: F821\n", "\n", " @eqx.filter_grad\n", " def grad_loss(symbolic_model):\n", " vmap_model = jax.vmap(symbolic_model, in_axes=(None, 0))\n", " pred_ys = vmap_model(ts, ys[:, 0]) # noqa: F821\n", " return jnp.mean((ys - pred_ys) ** 2) # noqa: F821\n", "\n", " optim = optax.adam(fine_tuning_lr)\n", " opt_state = optim.init(eqx.filter(symbolic_model, eqx.is_inexact_array))\n", "\n", " @eqx.filter_jit\n", " def make_step(symbolic_model, opt_state):\n", " grads = grad_loss(symbolic_model)\n", " updates, opt_state = optim.update(grads, opt_state)\n", " symbolic_model = eqx.apply_updates(symbolic_model, updates)\n", " return symbolic_model, opt_state\n", "\n", " for _ in range(fine_tuning_steps):\n", " symbolic_model, opt_state = make_step(symbolic_model, opt_state)\n", "\n", " #\n", " # Finally we round each constant to the nearest multiple of `quantise_to`.\n", " #\n", "\n", " trained_expressions = []\n", " for symbolic_module in symbolic_model.func.mlp.modules:\n", " expression = symbolic_module.sympy()\n", " expression = quantise(expression, quantise_to)\n", " trained_expressions.append(expression)\n", "\n", " print(f\"Expressions found: {trained_expressions}\")" ] }, { "cell_type": "code", "execution_count": 4, "id": "042fd565-825a-40fb-a4da-25e3e0da106a", "metadata": { "tags": [] }, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "Training neural differential equation.\n", "Step: 0, Loss: 0.16657482087612152, Computation time: 11.210124731063843\n", "Step: 100, Loss: 0.01115578692406416, Computation time: 0.002620220184326172\n", "Step: 200, Loss: 0.006481764372438192, Computation time: 0.0026247501373291016\n", "Step: 300, Loss: 0.0013819701271131635, Computation time: 0.003179311752319336\n", "Step: 400, Loss: 0.0010746140033006668, Computation time: 0.0031697750091552734\n", "Step: 499, Loss: 0.0007994902553036809, Computation time: 0.0031609535217285156\n", "Step: 0, Loss: 0.028307927772402763, Computation time: 11.210363626480103\n", "Step: 100, Loss: 0.005411561578512192, Computation time: 0.020294666290283203\n", "Step: 200, Loss: 0.004366496577858925, Computation time: 0.022084712982177734\n", "Step: 300, Loss: 0.0018046485492959619, Computation time: 0.022309064865112305\n", "Step: 400, Loss: 0.001767474808730185, Computation time: 0.021766185760498047\n", "Step: 499, Loss: 0.0011962582357227802, Computation time: 0.022264480590820312\n" ] }, { "data": { "image/png": "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\n", "text/plain": [ "
" ] }, "metadata": {}, "output_type": "display_data" }, { "name": "stdout", "output_type": "stream", "text": [ "Symbolically regressing across the vector field.\n" ] }, { "name": "stdout", "output_type": "stream", "text": [ "Started!\n", "\n", "Cycles per second: 5.190e+03\n", "Head worker occupation: 3.3%\n", "Progress: 434 / 800 total iterations (54.250%)\n", "==============================\n", "Best equations for output 1\n", "Hall of Fame:\n", "-----------------------------------------\n", "Complexity Loss Score Equation\n", "1 4.883e-02 1.206e+00 x1\n", "3 2.746e-02 2.877e-01 (x1 + -0.14616892)\n", "5 6.162e-04 1.899e+00 (x1 / (x1 - -1.0118991))\n", "7 4.476e-04 1.598e-01 ((x1 / 0.92953163) / (x1 + 1.0533974))\n", "9 3.997e-04 5.664e-02 (((x1 * 1.0935224) + -0.008988203) / (x1 + 1.0716586))\n", "13 3.364e-04 4.306e-02 (x1 * ((((x0 * -0.94923264) / 11.808947) - -1.087501) / (x1 + 1.0548282)))\n", "15 3.062e-04 4.714e-02 (x1 * ((((x0 * (-1.1005011 - x1)) / 13.075972) - -1.0955853) / (x1 + 1.0604433)))\n", "\n", "==============================\n", "Best equations for output 2\n", "Hall of Fame:\n", "-----------------------------------------\n", "Complexity Loss Score Equation\n", "1 1.588e-01 -1.000e-10 -0.002322703\n", "3 2.034e-02 1.028e+00 (0.14746223 - x0)\n", "5 1.413e-03 1.333e+00 (x0 / (-1.046938 - x0))\n", "7 6.958e-04 3.543e-01 (x0 / ((x0 + 1.1405994) / -1.1647526))\n", "9 2.163e-04 5.841e-01 (((x0 + -0.026584703) / (x0 + 1.2191753)) * -1.2456053)\n", "11 2.163e-04 7.749e-06 ((x0 - 0.026545616) / (((x0 / -1.2450436) + -0.9980602) - -0.019172505))\n", "\n", "==============================\n", "Press 'q' and then to stop execution early.\n", "\n", "Optimising symbolic expression.\n", "Expressions found: [x1/(x1 + 1.0), x0/(-x0 - 1.0)]\n" ] } ], "source": [ "main()" ] } ], "metadata": { "kernelspec": { "display_name": "py38", "language": "python", "name": "py38" }, "language_info": { "codemirror_mode": { "name": "ipython", "version": 3 }, "file_extension": ".py", "mimetype": "text/x-python", "name": "python", "nbconvert_exporter": "python", "pygments_lexer": "ipython3", "version": "3.8.16" } }, "nbformat": 4, "nbformat_minor": 5 }