test_neuropil.py

"""Tests for the functions in neuropil.py.

Author: Sander Keemink (swkeemink@scimail.eu)
Created: 2015-11-06
"""

import warnings

import numpy as np

from .. import neuropil as npil
from .base_test import BaseTestCase


class NeuropilMixin:
    """Neuropil functions testing class."""

    def setUp(self):
        """Set up the basic variables."""
        # length of arrays in tests
        self.n_obs = 100
        self.shape_desired = (2, self.n_obs)
        # setup basic x values for fake data
        self.x = np.linspace(0, np.pi * 2, self.n_obs)
        # Generate simple source data
        self.data = np.array([np.sin(self.x), np.cos(3 * self.x)]) + 1
        # Mix to make test data
        self.W = np.array([[0.2, 0.8], [0.8, 0.2]])
        self.S = np.dot(self.W, self.data)

    def run_method(self, method, expected_converged=None, **kwargs):
        """Test a single method, with specific parameters."""
        # Run the separation routine
        S_sep, S_matched, A_sep, convergence = npil.separate(
            self.S, sep_method=method, **kwargs
        )
        # Ensure output shapes are as expected
        self.assert_equal(S_sep.shape, self.shape_desired)
        self.assert_equal(S_matched.shape, self.shape_desired)
        self.assert_equal(S_sep.shape, self.shape_desired)
        # If specified, assert that the result is as expected
        if expected_converged is not None:
            self.assert_equal(convergence["converged"], expected_converged)
        return convergence["converged"]

    def test_method(self):
        self.run_method(self.method, expected_converged=True, max_tries=1)

    def test_method_loud(self):
        capture_pre = self.capsys.readouterr()  # Clear stdout
        self.run_method(self.method, expected_converged=True, max_tries=1, verbosity=1)
        capture_post = self.recapsys(capture_pre)  # Capture and then re-output
        self.assertTrue("converged after" in capture_post.out)

    def test_method_quiet(self):
        capture_pre = self.capsys.readouterr()  # Clear stdout
        self.run_method(self.method, expected_converged=True, max_tries=1, verbosity=0)
        capture_post = self.recapsys(capture_pre)  # Capture and then re-output
        self.assert_equal(capture_post.out, "")

    def test_reduce_dim(self):
        self.run_method(self.method, expected_converged=True, max_tries=1, n=2)

    def test_manual_seed(self):
        self.run_method(
            self.method,
            expected_converged=True,
            max_tries=1,
            random_state=0,
        )

    def test_retry_loud(self):
        capture_pre = self.capsys.readouterr()  # Clear stdout
        with warnings.catch_warnings():
            warnings.simplefilter("ignore")
            converged = self.run_method(
                self.method, max_iter=1, max_tries=3, verbosity=1
            )
        capture_post = self.recapsys(capture_pre)  # Capture and then re-output
        if not converged:
            self.assertTrue("Attempt 1 failed to converge at " in capture_post.out)
            self.assertTrue("Trying a new random state." in capture_post.out)
            self.assertTrue(
                "aximum number of allowed tries reached" in capture_post.out
            )

    def test_retry_quiet(self):
        capture_pre = self.capsys.readouterr()  # Clear stdout
        with warnings.catch_warnings():
            warnings.simplefilter("ignore")
            self.run_method(self.method, max_iter=1, max_tries=3, verbosity=0)
        capture_post = self.recapsys(capture_pre)  # Capture and then re-output
        self.assert_equal(capture_post.out, "")


class TestNeuropilNMF(BaseTestCase, NeuropilMixin):
    def setUp(self):
        NeuropilMixin.setUp(self)
        self.method = "nmf"

    def test_nmf_manual_alpha(self):
        self.run_method(self.method, expected_converged=True, alpha=0.2)

    def test_badmethod(self):
        with self.assertRaises(ValueError):
            npil.separate(self.S, sep_method="bogus_method")


class TestNeuropilICA(BaseTestCase, NeuropilMixin):
    def setUp(self):
        NeuropilMixin.setUp(self)
        self.method = "ica"


class TestNeuropilFA(BaseTestCase, NeuropilMixin):
    def setUp(self):
        NeuropilMixin.setUp(self)
        self.method = "FactorAnalysis"