def test_minibatch_reassign():
    # Give a perfect initialization, but a large reassignment_ratio,
    # as a result all the centers should be reassigned and the model
    # should not longer be good
    for this_X in (X, X_csr):
        mb_k_means = MiniBatchKMeans(n_clusters=n_clusters, batch_size=1,
                                     random_state=42)
        mb_k_means.fit(this_X)
        centers_before = mb_k_means.cluster_centers_.copy()
        try:
            old_stdout = sys.stdout
            sys.stdout = StringIO()
            # Turn on verbosity to smoke test the display code
            _mini_batch_step(this_X, (X ** 2).sum(axis=1),
                             mb_k_means.cluster_centers_,
                             mb_k_means.counts_,
                             np.zeros(X.shape[1], np.double),
                             False, random_reassign=True, random_state=42,
                             reassignment_ratio=1, verbose=True)
        finally:
            sys.stdout = old_stdout
        centers_after = mb_k_means.cluster_centers_.copy()
        # Check that all the centers have moved
        assert_greater(((centers_before - centers_after)**2).sum(axis=1).min(),
                       .2)

def test_minibatch_reassign():
    # Give a perfect initialization, but a large reassignment_ratio,
    # as a result all the centers should be reassigned and the model
    # should not longer be good
    for this_X in (X, X_csr):
        mb_k_means = MiniBatchKMeans(n_clusters=n_clusters, batch_size=100, random_state=42)
        mb_k_means.fit(this_X)

        score_before = mb_k_means.score(this_X)
        try:
            old_stdout = sys.stdout
            sys.stdout = StringIO()
            # Turn on verbosity to smoke test the display code
            _mini_batch_step(
                this_X,
                (X ** 2).sum(axis=1),
                mb_k_means.cluster_centers_,
                mb_k_means.counts_,
                np.zeros(X.shape[1], np.double),
                False,
                distances=np.zeros(X.shape[0]),
                random_reassign=True,
                random_state=42,
                reassignment_ratio=1,
                verbose=True,
            )
        finally:
            sys.stdout = old_stdout
        assert_greater(score_before, mb_k_means.score(this_X))

    # Give a perfect initialization, with a small reassignment_ratio,
    # no center should be reassigned
    for this_X in (X, X_csr):
        mb_k_means = MiniBatchKMeans(
            n_clusters=n_clusters, batch_size=100, init=centers.copy(), random_state=42, n_init=1
        )
        mb_k_means.fit(this_X)
        clusters_before = mb_k_means.cluster_centers_
        # Turn on verbosity to smoke test the display code
        _mini_batch_step(
            this_X,
            (X ** 2).sum(axis=1),
            mb_k_means.cluster_centers_,
            mb_k_means.counts_,
            np.zeros(X.shape[1], np.double),
            False,
            distances=np.zeros(X.shape[0]),
            random_reassign=True,
            random_state=42,
            reassignment_ratio=1e-15,
        )
        assert_array_almost_equal(clusters_before, mb_k_means.cluster_centers_)

def test_minibatch_update_consistency():
    # Check that dense and sparse minibatch update give the same results
    rng = np.random.RandomState(42)
    old_centers = centers + rng.normal(size=centers.shape)

    new_centers = old_centers.copy()
    new_centers_csr = old_centers.copy()

    counts = np.zeros(new_centers.shape[0], dtype=np.int32)
    counts_csr = np.zeros(new_centers.shape[0], dtype=np.int32)

    x_squared_norms = (X ** 2).sum(axis=1)
    x_squared_norms_csr = row_norms(X_csr, squared=True)

    buffer = np.zeros(centers.shape[1], dtype=np.double)
    buffer_csr = np.zeros(centers.shape[1], dtype=np.double)

    # extract a small minibatch
    X_mb = X[:10]
    X_mb_csr = X_csr[:10]
    x_mb_squared_norms = x_squared_norms[:10]
    x_mb_squared_norms_csr = x_squared_norms_csr[:10]

    # step 1: compute the dense minibatch update
    old_inertia, incremental_diff = _mini_batch_step(
        X_mb, x_mb_squared_norms, new_centers, counts,
        buffer, 1, None, random_reassign=False)
    assert_greater(old_inertia, 0.0)

    # compute the new inertia on the same batch to check that it decreased
    labels, new_inertia = _labels_inertia(
        X_mb, x_mb_squared_norms, new_centers)
    assert_greater(new_inertia, 0.0)
    assert_less(new_inertia, old_inertia)

    # check that the incremental difference computation is matching the
    # final observed value
    effective_diff = np.sum((new_centers - old_centers) ** 2)
    assert_almost_equal(incremental_diff, effective_diff)

    # step 2: compute the sparse minibatch update
    old_inertia_csr, incremental_diff_csr = _mini_batch_step(
        X_mb_csr, x_mb_squared_norms_csr, new_centers_csr, counts_csr,
        buffer_csr, 1, None, random_reassign=False)
    assert_greater(old_inertia_csr, 0.0)

    # compute the new inertia on the same batch to check that it decreased
    labels_csr, new_inertia_csr = _labels_inertia(
        X_mb_csr, x_mb_squared_norms_csr, new_centers_csr)
    assert_greater(new_inertia_csr, 0.0)
    assert_less(new_inertia_csr, old_inertia_csr)

    # check that the incremental difference computation is matching the
    # final observed value
    effective_diff = np.sum((new_centers_csr - old_centers) ** 2)
    assert_almost_equal(incremental_diff_csr, effective_diff)

    # step 3: check that sparse and dense updates lead to the same results
    assert_array_equal(labels, labels_csr)
    assert_array_almost_equal(new_centers, new_centers_csr)
    assert_almost_equal(incremental_diff, incremental_diff_csr)
    assert_almost_equal(old_inertia, old_inertia_csr)
    assert_almost_equal(new_inertia, new_inertia_csr)