test/test_jax.py

import unittest
import numpy as np
from pysr import sympy2jax, PySRRegressor
import pandas as pd
from jax import numpy as jnp
from jax import random
import sympy


class TestJAX(unittest.TestCase):
    def setUp(self):
        np.random.seed(0)

    def test_sympy2jax(self):
        x, y, z = sympy.symbols("x y z")
        cosx = 1.0 * sympy.cos(x) + y
        key = random.PRNGKey(0)
        X = random.normal(key, (1000, 2))
        true = 1.0 * jnp.cos(X[:, 0]) + X[:, 1]
        f, params = sympy2jax(cosx, [x, y, z])
        self.assertTrue(jnp.all(jnp.isclose(f(X, params), true)).item())

    def test_pipeline_pandas(self):
        X = pd.DataFrame(np.random.randn(100, 10))
        equations = pd.DataFrame(
            {
                "Equation": ["1.0", "cos(x1)", "square(cos(x1))"],
                "MSE": [1.0, 0.1, 1e-5],
                "Complexity": [1, 2, 3],
            }
        )

        equations["Complexity MSE Equation".split(" ")].to_csv(
            "equation_file.csv.bkup", sep="|"
        )

        model = PySRRegressor(
            equation_file="equation_file.csv",
            output_jax_format=True,
            variable_names="x1 x2 x3".split(" "),
        )

        model.fit(X, y=np.ones(X.shape[0]), from_equation_file=True)
        model.refresh()
        jformat = model.jax()

        np.testing.assert_almost_equal(
            np.array(jformat["callable"](jnp.array(X), jformat["parameters"])),
            np.square(np.cos(X.values[:, 1])),  # Select feature 1
            decimal=4,
        )

    def test_pipeline(self):
        X = np.random.randn(100, 10)
        equations = pd.DataFrame(
            {
                "Equation": ["1.0", "cos(x1)", "square(cos(x1))"],
                "MSE": [1.0, 0.1, 1e-5],
                "Complexity": [1, 2, 3],
            }
        )

        equations["Complexity MSE Equation".split(" ")].to_csv(
            "equation_file.csv.bkup", sep="|"
        )

        model = PySRRegressor(
            equation_file="equation_file.csv",
            output_jax_format=True,
            variable_names="x1 x2 x3".split(" "),
        )

        model.fit(X, y=np.ones(X.shape[0]), from_equation_file=True)
        model.refresh()
        jformat = model.jax()

        np.testing.assert_almost_equal(
            np.array(jformat["callable"](jnp.array(X), jformat["parameters"])),
            np.square(np.cos(X[:, 1])),  # Select feature 1
            decimal=4,
        )